Journal of Environmental Management 230 (2019) 245–254

Contents lists available at ScienceDirect

Journal of Environmental Management

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

Research article

Evaluation of direct and alternating current on nitrate removal using a T

continuous electrocoagulation process: Economical and environmental
approaches through RSM
Elnaz Karamati-Niaragha, Mohammad Reza Alavi Moghaddama,∗,
Mohammad Mahdi Emamjomehb, Ebrahim Nazlabadia
a
Civil and Environmental Engineering Department, Amirkabir University of Technology (AUT), Hafez Ave., Tehran, 15875-4413, Iran
b
Social Determinant of Health Research Center, Qazvin University of Medical Sciences, Qazvin, Iran

ARTICLE INFO ABSTRACT

Keywords: This study aims to investigate the effects of alternating current (AC) and direct current (DC) for nitrate removal
Alternating current and its operating costs by using a continuous electrocoagulation (CEC) process. For this purpose, two series of 31
Direct current experiments, which were designed by response surface method (RSM), were carried out in both cases of the AC
Nitrate removal and the DC modes. In each series, the effect of selected parameters, namely, initial nitrate concentration, inlet
Operating cost
flow rate, current density and initial pH along with their interactions on the nitrate removal efficiency as well as
Response surface method
its operating costs, as responses, were investigated separately. According to the analysis of variance (ANOVA),
there is a reasonable agreement between achieving results and the experimental data for both responses. The
nitrate removal in the AC mode was slightly more efficient than that of the DC mode. In addition, the average
operating costs of the DC mode, including the energy and the electrode consumption for the CEC process were
achieved 54 US$/(kg nitrate removed); whereas this amount was calculated 29 US$/(kg nitrate removed) for the
AC mode. Therefore, the average of the operating costs was improved more than 40% using the AC mode, which
was mainly related to reduction of aluminum electrode consumption.

1. Introduction be limited, mainly due to concentrating the pollution instead of re-

Preserving the quality of drinking water resources presents one of
the major challenges of the 21st century (Pulkka et al., 2014). Nitrate
pollution, one of the serious environmental problems, has been in-
creasing constantly in the last recent years, due to excessive application
of fertilizers, discharge of municipal/industrial wastewater, animal
wastes and septic systems (Azadeghan et al., 2014). Furthermore,
health effects of high nitrate concentration in drinking water are most
significantly linked to methemoglobinemia, also known as “blue-baby
syndrome”, that affects infants (Ghafari et al., 2008). The maximum
acceptable level of nitrate announced by WHO drinking water guideline
is 50 mg/L as NO3− (WHO, 2013).
In order to keep nitrate within the approved ranges, different
treatment methods in terms of physical, chemical and biological
methods are used to remove this anion (Moussa et al., 2017; Pulkka
et al., 2014). Applying the common physicochemical treatments such as
ion exchange, electrodialysis, reverse osmosis and nanofiltration may
moving it and they are also considered as expensive methods. The
sensitivity of biological methods to temperature, long duration of
treatment and low C:N ratios of waters are the major demerits of these
approaches (Moussa et al., 2017; Yehya et al., 2015).
Electrocoagulation (EC), is reported as a promising physicochemical
technology to remove various kinds of pollutants (Hakizimana et al.,
2017; Moussa et al., 2017; Nariyan et al., 2018), which also represents
efficient ability in nitrate removal (Emamjomeh and Sivakumar, 2009;
Govindan et al., 2015; Hashim et al., 2017a; Moussa et al., 2017;
Nazlabadi and Alavi Moghaddam, 2017; Pulkka et al., 2014; Xu et al.,
2018; Yehya et al., 2014). Advantages of the EC process, namely,
flexibility, environmental compatibility, energy efficiency, cost-effec-
tiveness, better amenability to automation and also the ability to cope
with various kinds of pollutants make it more requesting (Moussa et al.,
2017; Pulkka et al., 2014). In this process, coagulant and metallic hy-
droxide species are generated in situ by electro-dissolution of sacrificial
anode materials triggered by electric current applied through the

Corresponding author.
∗

E-mail addresses: elnazkaramati@gmail.com (E. Karamati-Niaragh), alavim@yahoo.com (M.R. Alavi Moghaddam),

m_emamjomeh@yahoo.com (M.M. Emamjomeh), ebrahim.nazlabadi@gmail.com (E. Nazlabadi).

https://doi.org/10.1016/j.jenvman.2018.09.091
Received 15 May 2018; Received in revised form 5 September 2018; Accepted 25 September 2018
0301-4797/ © 2018 Elsevier Ltd. All rights reserved.
E. Karamati-Niaragh et al.
electrodes (Moussa et al., 2017; Pulkka et al., 2014).
In the EC process, direct current (DC) is generally applied to pro-
duce the electric current through the electrodes. In this condition, an
impermeable oxide layer may form on the cathode as well as corrosion
of anodes and inhibit the effective current transport between electrodes
and therefore the efficiency of the EC process may diminish.
Furthermore, flows of current in one direction can reduce the lifetime of
the electrodes (Ghanizadeh et al., 2016; Moussa et al., 2017). The
passivity of the electrodes can be minimized using the alternating
current (AC) mode in which cathode and anode can be switched peri-
odically. Thus, the delay in cathode passivation and the anode dete-
rioration confirms acceptable electrode life (Vasudevan and Lakshmi,
2011). The AC mode was also considered as one of the proper alter-
native in the EC process for different types of pollutants namely, textile
dye (Tiaibaa et al., 2017), fluoride (Ghanizadeh et al., 2016), oily water
(Cerqueira et al., 2014), lead and zinc (Mansoorian et al., 2014), copper
(Kamaraj et al., 2013), cadmium (Vasudevan and Lakshmi, 2011) and
dyes (Eyvaz et al., 2009).
Ghanizadeh et al. (2016) showed that DC mode had higher efficacy
than that of AC mode for floride removal based on statistical analysis.
However, Mansoorian et al. (2014) proved that the operating costs
which related to energy consumption and electrode corrosion by using
AC mode reduced for lead and zinc removal. Cerqueira et al. (2014)
also indicated that application of AC mode of EC process for oil and
grease removal promoted a lower electrode consumption as compared
to using DC mode. Kamaraj et al. (2013) showed that the energy con-
sumption reduced for copper removal as well as removal efficiency
improved by using AC mode. Vasudevan and Lakshmi (2011) reported
that the cadmium removal for AC mode was efficient than that of DC
mode and the energy consumption reduced in AC mode, which lead to
operating costs improvement. Eyvaz et al. (2009) also found that the AC
mode was more efficient for dye removal.
Nitrate removal by using the EC process have been studied by other
research groups (Emamjomeh and Sivakumar, 2009; Govindan et al.,
2015; Hashim et al., 2017a; Kumar and Goel, 2010; Lacasa et al., 2013;
Lacasa et al., 2011; Moussa et al., 2017; Nazlabadi and Alavi
Moghaddam, 2017; Pulkka et al., 2014; Xu et al., 2018; Yehya et al.,
2014). Also, some studies indicate that the operating cost of nitrate
removal is of high expenses using the EC process (Lacasa et al., 2013;
Yehya et al., 2015). However, to the best of our knowledge, evaluation
of AC mode of a continuous electrocoagulation (CEC) process has not
been applied for nitrate pollution to reduce its operating costs. In ad-
dition, RSM as a statistical-mathematical method could be used to
evaluate the selected parameters and their interactions between re-
sponses for the CEC process (Hakizimana et al., 2017; Hendaoui et al.,
2018). As is known from the literature review, RSM method is applied
to evaluate EC process in batch systems for nitrate removal
(Emamjomeh et al., 2017; Nazlabadi and Alavi Moghaddam, 2017),
however, this method has not been simultaneously applied for nitrate
removal efficiency and its operating costs using AC and DC modes in the
CEC process.
The main objective of this research is an evaluation of the operating
costs for nitrate removal using the CEC process. For this purpose, the
electrode and the energy consumption using the AC and DC modes were
compared. Hence, two series of 31 experiments were carried out in both
cases through the RSM to identify the relationship between two re-
sponses (nitrate removal and its operating costs) and the selected ef-
fective parameters.
2. Materials and methods
2.1. Experimental setup
A continuous EC process was used for nitrate removal as schema-
tically shown in Fig. 1. An electrolytic cell was made from Plexiglas
with an effective volume of 2.4 L. In the cell, aluminum plate electrodes
246
Journal of Environmental Management 230 (2019) 245–254
(dimension 100 × 70 × 1 mm) were connected in a monopolar parallel
mode to the DC (Micro, PW4053R, 0–5A, 0–40 V) and the AC power
supply (0–5A, 0–40 V, 50 Hz). The inter-electrode gaps were kept con-
stant (10 mm) for all of the experiments. A peristaltic pump (Heidolph,
PD 5201, Germany) was used to allow influent control from a reservoir
tank to feed the electrolytic cell. Effluent from the electrolytic cell en-
ters to a sedimentation tank to remove the produced suspended solids.
2.2. Experimental techniques
Sodium nitrate (NaNO3) was dissolved in tap water for preparing
the required initial concentration (50–250 mg/L as NO3−). In order to
adjust the initial pH (2–10), sulphuric acid (2N) and sodium hydroxide
(5N) were added to the solutions. To increase the conductivity of the
solutions, sodium sulfate (Na2SO4) was used as a supporting electrolyte.
The pH and the conductivity of the solutions were monitored during
and at the end of the CEC process using a pH-meter (pH 340i, WTW,
Germany) and a conductivity-meter (Cond 340, WTW, Germany). All of
the experiments were conducted at the ambient temperature. At the end
of each experiment, samples were taken from the effluent and were
analyzed for nitrate concentration measurements. For this achievement,
UV–Vis spectrophotometer (HACH, DR4000, USA) at a wavelength of
500 nm was applied according to the standard methods for examination
of water and wastewater (AWWA, 1998). The performance of the CEC
process was assessed in terms of the contents of nitrate concentration
before and after treatment which was calculated using Eq. (1):
Cr Ct
Nitrate Removal Efficiency (RE%) = ( ) × 100
Cr (1)
where Cr and Ct are the nitrate concentration in raw and the treated
solutions (mg/L-NO3-), respectively. The accuracy of the experiments
was also assured using the random repetition of experiments. A scan-
ning electron microscopy (SEM) (Seron, AIS2100, South Korea) was
also used to evaluate the morphology of the electrodes after electrolysis
(AC and DC modes). The SEM images of electrodes were related to those
ones which were applied after all 31 runs in each current modes.
2.3. Economic analysis
Feasibility study of the EC process in large scale application, mainly
depends on its cost-effectiveness. The accurate operating cost of the EC
process comprises the cost of chemicals, electrode consumption, energy
consumption, sludge dewatering/disposal, maintenance, pertaining to
labor and fixed costs (Hakizimana et al., 2017; Hashim et al., 2017b).
According to the literature, costs of electrodes and energy consumption
were considerable for preliminary cost evaluation of EC process due to
their major role in high expenses (Behbahani et al., 2011; Yehya et al.,
2014), especially for removing low concentration of nitrate (Yehya
et al., 2014). The Eq. (2) was used for estimation of these operating
costs US$/(kg NO3− removed):
Operating Cost (OC)=a×Cenergy + b×Celectrodes (2)
where Cenergy (kWh/(kg NO3− removed)) and Celectrodes (kg Al/(kg
NO3− removed)) are energy and electrode consumption for nitrate re-
moval, respectively. Besides, a and b, the coefficient of the Iranian
market in 2016, are described below:
Coefficient a: industrial electricity price = 0.0222 US$/kWh
Coefficient b: Wholesale Al electrode price = 1.56 US$/kg Al
The electric energy consumption of the EC process was deduced as a
function of operation time through Eq. (3) as follows (Behbahani et al.,
2011; Hakizimana et al., 2017; Hashim et al., 2017b):
E. Karamati-Niaragh et al. Journal of Environmental Management 230 (2019) 245–254

Fig. 1. Continuous monopolar electrocoagulation process: 1) Reservoir tank; 2) Peristaltic pump; 3) Electrolytic cell; 4) Aluminum electrodes; 5) DC & AC power
supply; 6) Sedimentation tank and 7) Effluent.

t
levels of the independent parameters are presented according to the
E= V·I dt RSM model. For statistical calculations, the variables Xi were coded as
(3)
xi according to the following equation (Hakizimana et al., 2017;
0

where E is the electric energy consumption (kWh/(kg NO3− removed)), Behbahani et al., 2011):
V is the voltage (V), I is the current intensity (A) and t is the electrolysis
(Xi X 0)
time (hrs). In case of continuous EC, the energy consumption of the xi =
X (5)
pump was also calculated in the present study, however mechanical
power for mixing purposes was neglected. To obtain the mass depletion where X0 is the value of Xi at center point and δX represents the stage
of the electrode corrosion due to the anodic oxidation, weight loss of change.
the electrodes was calculated by subtracting the electrodes' weight at The statistical software Design Expert 7 was applied to analyze the
the beginning and at the end of each experiment. The electrodes were experimental data and generate a second-order polynomial model, thus
also cleaned after each run in order to prevent surface pollution of the ANOVA, regression coefficients, three dimensional curves of the re-
electrodes which might affect the electrodes passivation. sponse surfaces and residual plots were developed. The empirical model
of RSM represented by a second order polynomial regression is calcu-
2.4. Experimental design and data analysis lated through Eq. (6):
k k k k
In the present study, the central composite design (CCD) method of Y= b0 + bi xi + bii x 2i + b ijxixj
RSM, commonly used by other researchers, was selected to design the i= 1 i= 1 j= 1 i= 1 (6)
experiments, as well as to generate a quadratic model of the CEC pro-
where Y is the required responses of the CEC process (nitrate removal
cess. In this method, minimum numbers of experiments are achieved
efficiency and operating costs in this study) and b0, bi, bii, bij, xi and xj
and interaction between parameters is also analyzed (Karimifard and
are the constant coefficient, the linear coefficient, the quadratic coef-
Alavi Moghaddam, 2018; Behbahani et al., 2011; Emamjomeh et al.,
ficient, the interaction coefficient and the coded values of the factors,
2017; Georgiou et al., 2014; Nazlabadi and Alavi Moghaddam, 2017;
respectively. In this study, only a two-way interaction has been con-
Nazlabadi and Alavi Moghaddam, 2014; Song et al., 2015; Varank and
sidered. The statistical significance of the produced models was eval-
Sabuncu, 2015). According to our previous work and obtained sig-
uated through ANOVA with a confidence level of 95%. Residual plots
nificant parameters (Karamati Niaragh et al., 2017), four parameters
and coefficients of determination (R2, Adjusted R2 and Predicted R2)
including the initial nitrate concentration, inlet flow rate, current
were also presented to express the quality of the fitted models.
density and initial pH were selected in the present study to evaluate the
combined effects of these factors as well as present a mathematical
model. Thirty-one experiments were designed by employing the para- 3. Results and discussion
meters with five-levels. These experiments consist of 16 factorial ex-
periments, 8 axial experiments on the axis at a distance of ± α from the 3.1. Development of regression model and its validation for AC and DC
center, and 7 replicates at the center point. The value of α for rotat- modes
ability was calculated as brought in Eq. (4):
The results of the conducted experiments corresponding to the RSM
= (Nf )1/4 (4) design for nitrate removal efficiency and operating costs in both cases
of DC and AC modes are presented in Table 2. The modified quadratic
where Nf is the number of points in the cube portion of the design and is
model for nitrate removal efficiency (RE%) and operating costs (OC) in
equal to 2k, and k is the number of factors. Thus, the α was calculated to
terms of the coded values for DC and AC mode are given by Eqs. (7a),
be 2 in the present study. In Table 1, selected factors and their five
(b) and (8a), (b), respectively:
For DC mode:
Table 1
Levels of the parameters according to RSM for the CEC process. YRE = 52.49 7.01x1 7.04x2 + 5.17x3 + 1.98x 4 4.64x1 x2 + 2.34x1 x3
Variables Unit Factors Levels 2.77x1 x 4 + 2.72x2 x3 2.14x2 x 4 2.62x12 + 1.63x 22 1.29x 32
-α −1 0 +1 +α + 3.77x42 (7a)

Initial nitrate concentration mg/L X1 50 100 150 200 250

YOC = 43.49 12.13x1 4.10x2 + 5.41x3 0.64x 4 1.40x1 x3 0.7x1 x 4
Inlet flow rate mL/min X2 40 60 80 100 120
Current density A/m2 X3 45 60 75 90 105 0.80x2 x 4 + 0.98x3 x 4 + 8.50x12 11.25x 22 + 8.67x42 (7b)
Initial pH – X4 2 4 6 8 10
For AC mode:

247
E. Karamati-Niaragh et al. Journal of Environmental Management 230 (2019) 245–254

Table 2
RSM experimental design and its observed values in both cases of DC and AC modes.
Run number Initial nitrate Inlet Current Initial pH Direct current (DC) Alternating current (AC)
concentration (x1) flow rate density (x3) (x4)
a
(x2) RE (%) RE (kg NO3− a
OC (US$/kg RE (%) RE (kg NO3− OC (US$/kg
removed) × 104 NO3− removed) × 104 NO3−
removed) removed)

1 0 0 −2 0 34.67 0.52 48.29 35.33 0.53 30.51

2 0 0 +2 0 60 0.9 47.24 60 0.9 29.11
3 +1 +1 +1 +1 61 1.22 33.42 61.5 1.23 19.23
4 0 0 0 0 49.33 0.74 45.98 50.67 0.76 28.73
5 +1 −1 −1 +1 56 1.12 36.60 55 1.1 21.50
6 0 −2 0 0 74.67 1.12 52.57 76.67 1.15 27.34
7 0 0 0 0 48 0.72 45.75 50 0.75 30.03
8 +1 +1 −1 +1 55 1.1 24.50 52.5 1.05 15.62
9 +2 0 0 0 28 0.7 54.19 28.8 0.72 32.35
10 0 0 0 0 55.33 1.01 40.29 66.67 1 20.83
11 −2 0 0 0 56 2.8 100.81 60 0.3 67.97
12 +1 +1 +1 −1 50 1 40.20 47.5 0.95 26.43
13 −1 +1 −1 −1 52 0.52 52.02 55 0.55 33.36
14 −1 −1 −1 −1 57 0.57 62.96 60 0.6 37.10
15 0 0 0 −2 30.67 0.46 75.94 34 0.51 46.58
16 0 0 0 +2 57.33 0.86 80.42 62 0.93 45.72
17 0 0 0 0 52.44 0.79 43.05 54 0.81 28.62
18 0 0 0 0 53.33 0.8 43.88 54.67 0.82 24.04
19 +1 +1 −1 −1 45 0.9 33.48 30 0.6 22.22
20 −1 +1 +1 +1 66 0.66 55.98 70 0.7 31.88
21 −1 +1 −1 +1 58 0.58 49.56 59 0.59 31.80
22 +1 −1 +1 −1 59 1.18 45.80 55 1.1 26.04
23 −1 −1 −1 +1 68 0.68 55.37 69 0.69 29.63
24 +1 −1 −1 −1 54 1.08 37.56 50 1 20.99
25 0 +2 0 0 43.33 0.65 54.71 44 0.66 30.56
26 +1 −1 +1 +1 64 1.28 44.81 57.5 1.15 27.19
27 −1 −1 +1 +1 71 0.71 74.45 75 0.75 41.01
28 0 0 0 0 50.67 0.76 44.10 52 0.78 25.01
29 0 0 0 0 56.67 0.85 41.38 53.33 0.8 13.66
30 −1 +1 +1 −1 60 0.6 64.84 61 0.61 36.43
31 −1 −1 +1 −1 56 0.56 95.33 65 0.65 50.30

a
RE: Removal Efficiency, OC: Operating Costs.

YRE = 52.93 7.91x1 6.71x2 + 5.36x3 + 3.34x 4 2.11x1 x 4 + 2.64x2 x3 statistically significant (P ˂0.0001), and implies that the modified
second-order quadratic model is fitted well with the experimental re-
+ 0.028x2 x 4 0.69x3 x 4 2.13x12 + 1.85x 22 1.32x 32 + 3.50x42
sults. The results also specified that all the selected independent para-
(8a) meters were significant for removal efficiency of both cases of DC and
YOC = 25.45 7.89x1 2.66x2 + 3.26x3 11.29x 4 + 0.41x1 x2 + 1.03x1 x 4 AC models. Also, the most important independent process variables for
operating costs were found to be initial nitrate concentration (x1), inlet
1.16x2 x3 0.66x3 x 4 + 6.18x12 7.74x 22 + 5.18x42 (8b)
flow rate (x2) and current density (x3) (detailed data not shown).
The ANOVA results for DC and AC modes are also summarised in According to Table 3, the lack of fit for nitrate removal efficiency (P-
Table 3. As indicated in Table 3, the quadratic model of nitrate removal value of 0.5180 and 0.1339 for DC and AC, respectively) and operating
efficiency and operating costs in both cases of DC and AC modes were costs (P-value of 0.1592 and 0.5044 for DC and AC, respectively) were

Table 3
ANOVA results and determination coefficients for RSM model in both cases of DC and AC mode.
Current Response Source DF Adj SS Adj MS F P Coefficient of determination (%) 0.8 Pred-R2≤ Adj-R2 ≤1.2 Pred-R2 &
Adj-R2≤ R2
R2 Adj-R2 Pred-R2

DC RE (%) Model 13 2388.70 183.75 23.12 ˂0.0001 96.8 92.59 78.80 ✓

Lack of Fit 6 48 8 1.02 0.5180
Pure Error 4 31.48 7.87 – –
Total 23 2468.18 – – –
OC (US$/kg NO3− Model 11 7094.74 644.98 83.89 ˂0.0001 98.6 97.4 92.1 ✓
removed) Lack of Fit 7 73.21 10.46 2.35 0.1592
Pure Error 6 26.74 4.46 – –
Total 24 7194.68 – – –
AC RE(%) Model 13 3394.27 261.10 42.58 ˂0.0001 98.2 95.9 83.5 ✓
Lack of Fit 6 51.01 8.50 3.30 0.1339
Pure Error 4 10.30 2.58 – –
Total 23 3455.58 – – –
OC(US$/kg NO3− Model 11 3035.02 275.91 23.56 ˂0.0001 95.6 91.5 80 ✓
removed) Lack of Fit 8 96.11 12.01 1.08 0.5044
Pure Error 4 44.40 11.10 – –
Total 23 3175.52 – – –

248
E. Karamati-Niaragh et al.
(a)
Direct Current (DC)
(a)
Alternating Current (AC)
Fig. 2. Normal probability plots in both cases of DC and AC modes for (a) nitrate removal efficiency and (b) operating costs.
not significant; thus, the data was fitted well to the model. In addition,
high values of R2 were achieved 96.8% and 98.6% for removal effi-
ciency and operating costs of DC, respectively, indicated a satisfactory
fit of the quadratic model to the experimental data. In the case of AC,
values of R2 were found to be 98.2% and 95.6% for nitrate removal
efficiency and operating costs, respectively. The values of adjusted R2
(Adj-R2) were also lower than the R2 values and the Adj-R2 and pre-
dicted R2 (Pred-R2) were within approximately 0.20, which indicated
reasonable agreement.
Furthermore, the adequacies of the models were controlled by
analyzing the residuals. For this purpose, the normal probability plots
for nitrate removal efficiency and operating costs, which indicates
whether the residuals follow a normal distribution, are illustrated in
Fig. 2. As seen in Fig. 2, the normal probability was relatively satisfied
as the points in the plot follows from fairly straight line. All of the plots
in Fig. 2 revealed that the model for both DC and AC modes were found
to be adequate to describe the nitrate removal efficiency and its relative
operating costs using CCD method of RSM.
3.2. Evaluation of nitrate removal for AC and DC modes
Generally, the most important parameters, which affect the per-
formance of the CEC process, are initial concentration of nitrate, cur-
rent density, inlet flow rate and initial pH of the solution. The main
parameters effects on the CEC process performance as well as their
contribution ratio were elaborately discussed in our previous work
(Karamati Niaragh et al., 2017).
The significant relationships between nitrate removal efficiency and
the two-way interaction terms of the DC mode for the CEC process are
also shown in Fig. 3. According to Fig. 3(a), a higher removal efficiency
occurred for low initial nitrate concentration and low inlet flow rate. At
a constant nitrate removal efficiency, the required inlet flow rate re-
duced when the initial nitrate concentration increased. The similar
results were also observed for AC mode (data not shown here). The
removal efficiency was achieved higher than 60% when the initial ni-
trate concentration and the inlet flow rate were 100 mg/L and 100 mL/
min, respectively. The detention time required for the inlet flow rate of
100 mL/min was 48 min. This result is in agreement with the results
achieved by Hashim et al. in batch EC process, which revealed that the
required time to reduce the initial nitrate concentration of 100 mg/L to
249
Journal of Environmental Management 230 (2019) 245–254
(b)
(b)
the allowable WHO limitations for drinking water was 55 min (Hashim
et al., 2017a). However, at the initial nitrate concentration of 150 mg/
L, the inlet flow rate of 100 mL/min was not enough for the CEC process
to meet these limitations.
The literature highlighted that current density, which determines
the anodic dissolution rate and hydrogen gas generation, affected the
EC performance (Behbahani et al., 2011; Koparal and Oğütveren, 2002;
Moussa et al., 2017). Fig. 3(b) and (c) shows that higher nitrate removal
rates were achieved with higher current density. In the current study,
the experiments were carried out in different ranges of the current
densities between 60 and 90 A/m2 to evaluate the influence of this
main parameter on the CEC process for nitrate removal. For instance,
according to Fig. 3(c), in constant initial nitrate concentration of
100 mg/L, initial pH of 7 and the inlet flow rate of 100 mL/min, the
nitrate removal efficiency improved about 15%. The improving effects
of increased current density were also observed for AC mode in the
same conditions (data not shown here). However, the results of this
study along with literature (Behbahani et al., 2011; Emamjomeh et al.,
2017; Hashim et al., 2017a, 2017b; Karamati Niaragh et al., 2017;
Nazlabadi and Alavi Moghaddam, 2017) showed that high amounts of
current density had a significant negative effect on the energy con-
sumptions, and consequently on the operating costs. Therefore, by
considering the operating costs, the optimum current density for nitrate
removal efficiency using the CEC process for both cases of DC and AC
modes was calculated to be 70–80 A/m2 in the present study.
Fig. 3(d) illustrates that at a constant initial nitrate concentration,
by increasing the initial pH, nitrate removal rate increases. This can be
related to the reaction between Al metal and hydroxide ions. In this
work, the optimum removal efficiencies were obtained in the initial pH
range of 7–8 for both AC and DC modes. The results of this study have
been recently observed by the other research groups (Hashim et al.,
2017a; Yehya et al., 2014; Kumar and Goel, 2010; Lakshmi et al., 2013;
Nazlabadi and Moghaddam, 2017). Conversely, some research groups
have been reported that the alkaline pH (9–11) is the optimum pH for
nitrate removal (Koparal and Oğütveren, 2002; Emamjomeh et al.,
2017; Emamjomeh and Sivakumar, 2009). But, Safari and Yehya study
groups showed that the rate of nitrate removal had not changed sig-
nificantly when the pH increases from 7 to 10 (Yehya et al., 2014; Safari
et al., 2015).
E. Karamati-Niaragh et al. Journal of Environmental Management 230 (2019) 245–254

(a) (b)
al
)

al
)
(%

(%
Nitrate remov

Nitrate remov
y

cy
nc

en
effic ie

effici
In (mL
co

Ini trati

let /m
nc

tia on
en

flo in)
nsity
l n (m

w
te nt de
itr

w ra Curre m2 )

rat
ate g/L)

t flo (A/

e
Inle /min)
(mL

(c) (d)

al
(% )
Nitrate remov
al
)

y
efficiency (%
Nitrate remov

efficienc
co

In trat
nc
iti io
co

en
al
In ntra
nc

nit (m
iti ti
e

l pH
al on

rat g/L

/m ) Initia
n
ni (m

2
e

(A
tra g

nt density
te /L

Curre
)
)

Fig. 3. Relation between nitrate removal efficiency and the interaction terms using 3d plots (DC mode): (a) the interaction between initial nitrate concentration and
inlet flow rate; (b) the interaction between inlet flow rate and the current density; (c) the initial nitrate concentration and the current density; (d) the initial nitrate
concentration and the initial pH.

3.3. Analysis of operating costs for AC and DC modes
From the literature, it is obvious that the application of EC for ni-
trate removal is of high operating costs, especially, it is expensive for
treatment of water containing low nitrate concentrations (Yehya et al.,
2014). However, for the EC lab scale units, the electrode material and
the electrical energy consumption are of significant effect on operating
costs (Yehya et al., 2014; Hashim et al., 2017a; Emamjomeh et al.,
2017). In addition, a review of the existing literature indicated that the
electrode material and energy consumption could be more cost-effec-
tive if the AC mode is used instead of the DC mode during the EC
process. Furthermore, the application of AC can prohibit the formation
of the impermeable layer on the surface of the anodes, which is con-
sidered as a demerits in DC mode of the EC process (Ghanizadeh et al.,
2016; Cerqueira et al., 2014; Mansoorian et al., 2014; Kamaraj et al.,
2013; Vasudevan and Lakshmi, 2011; Eyvaz et al., 2009).
To investigate the techno-economical effects of AC and DC mode,
two series of 31 experiments have been carried out using the AC mode
as well as the DC mode. According to Table 2, nitrate removal efficiency
in the AC mode is slightly higher (1%) than the DC mode in most of the
experiments; however, it does not mean that the AC mode could be an
efficient alternative. In agreement with this result, other research
groups have reported that the high performance of AC in the removal of
textile dye (Tiaibaa et al., 2017), oily water (Cerqueira et al., 2014),
lead and zinc (Mansoorian et al., 2014), copper (Kamaraj et al., 2013),
cadmium (Vasudevan and Lakshmi, 2011) and dyes (Eyvaz et al.,
2009), but Ghanizadeh research group has confirmed that the efficacy
of AC mode was not statistically significant for fluoride removal
(Ghanizadeh et al., 2016). Therefore, in the present study, minimizing
the electrodes passivity using the AC mode had no significant effect on
nitrate removal efficiency during the CEC process. It may due to an
insignificant increase in the resistance of the electrolyte in the applied
detention times.
In conjunction with operating costs, Fig. 4 illustrated that the re-
moving nitrate operating costs of DC mode are higher than AC mode for
the CEC process. The average operating cost of all experiments for DC
mode was achieved 54 US$/(kg NO3− removed); while this amount
was obtained 29 US$/(kg NO3− removed) using AC mode. As a result,
the operating costs of nitrate removal efficiency using AC mode for the
CEC process was improved about 40%. To elaborate how the costs were
attributed to the amount of energy and electrode used in the CEC
process, the energy and the electrode consumption of all experiments

250
E. Karamati-Niaragh et al. Journal of Environmental Management 230 (2019) 245–254

120 AC

(US$/kg nitrate removed)

DC
100

Operating Cost 80

60

40

20

0
0 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

Number of experiment
Fig. 4. The operating costs of the CEC process through 31 experiments designed by RSM in both cases of DC and AC modes.

3
AC
2.5 DC
(kWh/gr nitrate removed)
Energy consumption

1.5

0.5

0
0 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

Number of experiment
Fig. 5. The energy consumption of the CEC process through 31 experiments designed by RSM in both cases of DC and AC modes.

50 AC
45 DC
(kg Al/kg nitrate removed)
Electrode consumption

40
35
30
25
20
15
10
5
0
0 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

Number of experiment
Fig. 6. The electrode consumption of the CEC process through 31 experiments designed by RSM in both cases of DC and AC modes.

are presented in Figs. 5 and 6, respectively, in both cases of AC and DC
mode.
According to Fig. 5, the energy consumption was slightly higher in
the case of AC mode in current intensities between 2 and 4A for nitrate
removal. Thus, it indicated that the energy consumption had may no
relation with the impermeable oxide layer formed on cathodes in DC
mode, which increase the resistivity of the electrolyte. This result is in
agreement with Cerqueira research work on oily water treatment,
which demonstrated that the energy consumption was higher in the
case of AC mode at current intensities lower than 3A (Cerqueira et al.,
2014). Conversely, other research groups indicated that the energy
consumption in the case of AC is lower than DC mode of EC process for
removal of fluoride (Ghanizadeh et al., 2016), lead and zinc
(Mansoorian et al., 2014), copper (Kamaraj et al., 2013), cadmium
(Vasudevan and Lakshmi, 2011) and dyes (Eyvaz et al., 2009). The
higher energy consumption in AC mode may due to the type of

251
E. Karamati-Niaragh et al.
H2O
NO2
-
e after 31 runs, which is presented in Fig. 8(a) and (b), respectively. It can
+2 -
e
-2
+6 e- N2
NO3
+8
e- -
e
-3
NH3
H2O
Fig. 7. Schematic reduction pathway of nitrate anions.
pollutants and its complexes of removal mechanisms. The process of
nitrate reduction is summarised in Fig. 7 (Govindan et al., 2015;
Hashim et al., 2017a).
Also, since the first step in nitrate removal is related to nitrate re-
duction, the key role on nitrate removal mechanism is the electricity,
which has significant effect on energy consumption (Yehya et al.,
2014).
As depicted in Fig. 6, conducting of all experiments in constant si-
tuations for both AC and DC modes confirmed that the consumption of
the aluminum electrode was significantly lower in the AC mode as
compared to that of the DC mode. According to the obtained results, the
average electrode consumption of the experiments using the DC was
about 4 times higher than that of the AC mode. In other words, after
same conditions of the CEC process, the average amount of aluminum
electrode consumed in DC mode was 19.96 gr/(gr NO3−removed),
whereas in the AC mode it was achieved 4.96 gr/(gr NO3− removed).
Therefore, the DC mode of the CEC process consumed the electrode
much faster and the AC mode performed efficiently. This was probably
related to the uniform dissolution of the anode and the cathode during
the CEC process using AC mode. It means that the higher electrode
consumption by DC mode refers to one directional flows of DC mode,
which allows the irregular wear on the electrode plates due to the
oxidation at the same preferential points of the plates. The reversal of
current in the AC mode induces the uniformly points of electrodes, thus,
(a)
Fig. 8. SEM images of aluminum anodes: (a) Alternating current; (b) Direct current.
Journal of Environmental Management 230 (2019) 245–254
H2O provides a longer life for electrodes, which is considered as one of the
disadvantages of the EC process. To obtain more insight into the effect
of AC mode, the morphology of the aluminum anode surface after the
CEC process was evaluated by SEM for both cases of AC and DC modes
be observed that less disordered pores formed in the case of AC which
confirmed the uniform dissolving of the electrodes. While, for the DC
mode, many tortuous pores of the electrode's surface shows the alu-
minum oxide may form.
4. Conclusions
The alternating current was applied in the present study to reduce
H2O the operating costs of the CEC process for nitrate removal. The per-
formance and the operating costs of the CEC process using both cases of
the AC and the DC modes was also compared. For this purpose, two
series of 31 experiments, which were designed by RSM, were carried
out in both current modes. The second modified quadratic models were
also developed for the nitrate removal efficiency and the operating
costs for both current modes, which were statistically significant.
According to ANOVA results, the models fitted well with the experi-
mental results. High R2 values for AC and DC modes were 96.8% and
98.6% for nitrate removal efficiency and 98.2% and 95.6% for oper-
ating costs, respectively. The R2, Adj-R2 and Pred-R2 were also in a
reasonable agreement. Hence, the RSM method has been approved as a
powerful tool to evaluate the CEC process. The performance of nitrate
removal in the AC mode was slightly higher than that of the DC mode in
more than 80% of the experiments. However, to choose the AC mode as
an effective alternative, the performance of the CEC process has to be
evaluated in conjunction with its operating costs. The operating costs of
the DC mode was higher than that of the AC mode for nitrate removal
efficiency using the CEC process. The average operating cost of the
experiments using the DC mode was achieved 54 US$/(kg nitrate re-
moved); while this amount was obtained 29 US$/(kg nitrate removed)
for the AC mode. The significant reduction of the operating costs was
related to the electrode consumption in this study rather than the en-
ergy consumption. The average electrode consumption of the experi-
ments using the DC mode was more than 4 times higher than that of the
AC mode. As a result, the operating costs was improved more than 40%
using the AC mode. Therefore, the AC mode achieved to be a promising
cost-effective alternative for nitrate removal using the CEC process.
(b)
252
E. Karamati-Niaragh et al. Journal of Environmental Management 230 (2019) 245–254

(a) (b)

oval
oval

)
)

efficien (%
efficiency (%

Nitrate rem
Nitrate rem

cy
In (mL
In nce g/L

le
co (m
iti nt )

t f /mi
al rat

nsity

lo
te nt de
ni io

w ra

w n)
Curre m2 )
tra n

t flo

ra
Inle /min) (A/
te

te
(mL

(c)
oval
(%)
Nitrate rem
effi ci en cy

In nce g/L
co (m
iti nt )
al rat
ni io

al pH
tra n

Initi
te

Acknowledgments
The authors would like to thank the Amirkabir University of
Technology for financial support. In addition, the authors wish to ap-
preciate Mr. Mahyar Ghannadi and Ms. Pariya Sefatjoo (former M.Sc
students of AUT) for their respected assistance during the experiments.
References
AWWA (American Water Works Association), APHA (American Public Health
Association), WEF (Water Environment Federation), 1998. Standard Methods for the
Examination of Water and Wastewater. Washington, DC, USA.
Azadeghan, M.M., Alavi Moghaddam, M.R., Maknoon, R., 2014. Public awareness and
performance regarding nitrate pollution in nitrate-polluted area of Tehran, Iran.
Environ. Eng. Manag. J. (EEMJ) 13 (3), 611–617.
Behbahani, M., Alavi Moghaddam, M.R., Arami, M., 2011. Techno-economical evaluation
of fluoride removal by electrocoagulation process: optimization through response
surface methodology. Desalination 271 (1–3), 209–218.
Cerqueira, A.A., Souza, P.S.A., Marques, M.R.C., 2014. Effects of direct and alternating
current on the treatment of oily water in an electroflocculation process. Braz. J.
Chem. Eng. 31 (3), 693–701.
Emamjomeh, M.M., Jamali, H.A., Moradnia, M., 2017. Optimization of nitrate removal
efficiency and energy consumption using a batch monopolar electrocoagulation:
prediction by RSM Method. J. Environ. Eng. 143 (7), 4017022.
Emamjomeh, M.M., Sivakumar, M., 2009. Denitrification using a monopolar electro-
coagulation/flotation (ECF) process. J. Environ. Manag. 91 (2), 516–522.
Eyvaz, M., Kirlaroglu, M., Aktas, T.S., Yuksel, E., 2009. The effects of alternating current
electrocoagulation on dye removal from aqueous solutions. Chem. Eng. J. 153 (1–3),
16–22.
Georgiou, S.D., Stylianou, S., Aggarwal, M., 2014. A class of composite designs for re-
sponse surface methodology. Comput. Stat. Data Anal. 71, 1124–1133.
Ghafari, S., Hasan, M., Aroua, M.K., 2008. Bio-electrochemical removal of nitrate from
water and wastewater-A review. Bioresour. Technol. 99 (10), 3965–3974.
Ghanizadeh, G., Shariati neghab, G., Salem, M., Khalagi, K., 2016. Taguchi experimental
design for electrocoagulation process using alternating and direct current on fluoride
removal from water. Desalination Water Treat. 57 (27), 12675–12683.
Govindan, K., Noel, M., Mohan, R., 2015. Removal of nitrate ion from water by elec-
trochemical approaches. J. Water Process Eng. 6, 58–63.
Hakizimana, J.N., Gourich, B., Chafi, M., Stiriba, Y., Vial, C., Drogui, P., Naja, J., 2017.
Electrocoagulation process in water treatment: a review of electrocoagulation mod-
eling approaches. Desalination 404, 1–21.
Hashim, K.S., Shaw, A., Al Khaddar, R., Pedrola, M.O., Phipps, D., 2017a. Energy efficient
electrocoagulation using a new flow column reactor to remove nitrate from drinking
water: experimental, statistical, and economic approach. J. Environ. Manag. 196,
224–233.
Hashim, K.S., Shaw, A., Al Khaddar, R., Pedrola, M.O., Phipps, D., 2017b. Iron removal,
energy consumption and operating cost of electrocoagulation of drinking water using
a new flow column reactor. J. Environ. Manag. 189, 98–108.
Hendaoui, K., Ayari, F., Rayana, I.B., Ammar, R.B., Darragi, F., Trabelsi-Ayadi, M., 2018.
Real indigo dyeing effluent decontamination using continuous electrocoagulation
cell: study and optimization using Response Surface Methodology. Process Saf.
Environ. Protect. 116, 578–589.
Kamaraj, R., Ganesan, P., Lakshmi, J., Vasudevan, S., 2013. Removal of copper from
water by electrocoagulation process-effect of alternating current (AC) and direct
current (DC). Environ. Sci. Pollut. Control Ser. 20 (1), 399–412.
Karamati-Niaragh, E., Alavi Moghaddam, M.R., Emamjomeh, M.M., 2017. Techno-eco-
nomical evaluation of nitrate removal using continuous flow electro-coagulation
process: optimization by Taguchi model. Water Sci. Technol. Water Supply 17 (6),
1703–1711.
Karimifard, S., Alavi Moghaddam, M.R., 2018. Application of response surface metho-
dology in physicochemical removal of dyes from wastewater: a critical review. Sci.

253
E. Karamati-Niaragh et al.
Total Environ. 640, 772–797.
Koparal, A.S., Oğütveren, Ü.B., 2002. Removal of nitrate from water by electroreduction
and electrocoagulation. J. Hazard Mater. 89 (1), 83–94.
Kumar, N.S., Goel, S., 2010. Factors influencing arsenic and nitrate removal from
drinking water in a continuous flow electrocoagulation (EC) process. J. Hazard
Mater. 173 (1–3), 528–533.
Lacasa, E., Cañizares, P., Sáez, C., Martínez, F., Rodrigo, M.A., 2013. Modelling and cost
evaluation of electro-coagulation processes for the removal of anions from water.
Separ. Purif. Technol. 107, 219–227.
Lacasa, E., Cañizares, P., Sáez, C., Fernández, F.J., Rodrigo, M.A., 2011. Removal of ni-
trates from groundwater by electrocoagulation. Chem. Eng. J. 171 (3), 1012–1017.
Lakshmi, J., Sozhan, G., Vasudevan, S., 2013. Recovery of hydrogen and removal of ni-
trate from water by electrocoagulation process. Environ. Sci. Pollut. Control Ser. 20
(4), 2184–2192.
Mansoorian, H.J., Mahvi, A.H., Jafari, A.J., 2014. Removal of lead and zinc from battery
industry wastewater using electrocoagulation process: influence of direct and alter-
nating current by using iron and stainless steel rod electrodes. Separ. Purif. Technol.
135, 165–175.
Moussa, D.T., El-Naas, M.H., Nasser, M., Al-Marri, M.J., 2017. A comprehensive review of
electrocoagulation for water treatment: potentials and challenges. J. Environ. Manag.
186, 24–41.
Nariyan, E., Sillanpää, M., Wolkersdorfer, C., 2018. Uranium removal from Pyhäsalmi/
Finland mine water by batch electrocoagulation and optimization with the response
surface methodology. Separ. Purif. Technol. 193, 386–397.
Nazlabadi, E., Alavi Moghaddam, M.R., 2017. Simulation of nitrate removal in a batch
flow electrocoagulation-flotation (ECF) process by response surface method (RSM).
Trends in Asian Water Environmental Science and Technology. Springer, Cham, pp.
49–60.
Nazlabadi, E., Alavi Moghaddam, M.R., 2014. Simulation of operation cost for nitrate
removal using response surface method in electrocoagulation process. Int. J. Environ.
254
Journal of Environmental Management 230 (2019) 245–254
Eng. 1 (2), 91–95.
Pulkka, S., Martikainen, M., Bhatnagar, A., Sillanpää, M., 2014. Electrochemical methods
for the removal of anionic contaminants from water - a review. Separ. Purif. Technol.
132, 252–271.
Safari, M., Rezaee, A., Ayati, B., Jonidi-Jafari, A., 2015. Simultaneous removal of nitrate
and its intermediates by use of bipolar electrochemistry. Res. Chem. Intermed. 41 (3),
1365–1372.
Song, P., Yang, Z., Zeng, G., Yang, X., Xu, H., Huang, J., Wang, L., 2015. Optimization,
kinetics, isotherms, and thermodynamics studies of antimony removal in electro-
coagulation process. Water Air Soil Pollut. 226 (11), 380.
Tiaibaa, M., Merzoukb, B., Amourd, A., Mazoura, M., Leclerce, J.P., Lapicquee, F., 2017.
Influence of electrodes connection mode and type of current in electrocoagulation
process on the removal of a textile dye. Desalination Water Treat. 73, 330–338.
Varank, G., Sabuncu, M.E., 2015. Application of central composite design approach for
dairy wastewater treatment by electrocoagulation using iron and aluminum elec-
trodes: modeling and optimization. Desalination Water Treat. 56 (1), 33–54.
Vasudevan, S., Lakshmi, J., 2011. Effects of alternating and direct current in electro-
coagulation process on the removal of cadmium from water: a novel approach. Separ.
Purif. Technol. 80 (3), 643–651.
WHO, 2013. Guidelines for Drinking-Water Quality, 4th ed. Water Sanitation Health,
Geneva, Switzerland.
Xu, D., Li, Y., Yin, L., Ji, Y., Niu, J., Yu, Y., 2018. Electrochemical removal of nitrate in
industrial wastewater. Front. Environ. Sci. Eng. 12 (1), 9.
Yehya, T., Balla, W., Chafi, M., Audonnet, F., Vial, C., Essadki, A., Gourich, B., 2015.
Assessment of denitrification using electrocoagulation process. Can. J. Chem. Eng. 93
(2), 241–248.
Yehya, T., Chafi, M., Balla, W., Vial, C., Essadki, A., Gourich, B., 2014. Experimental
analysis and modeling of denitrification using electrocoagulation process. Separ.
Purif. Technol. 132, 644–654.