i An update to this article is included at the end

Journal of Water Process Engineering 49 (2022) 103074

Contents lists available at ScienceDirect

Journal of Water Process Engineering

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

Optimization of electrocoagulation process for treatment of rice mill

effluent using response surface methodology
A. Ronaldo Anuf a, *, K. Ramaraj a, Vishnu Sankar Sivasankarapillai b,
Ragupathy Dhanusuraman b, J. Prakash Maran c, G. Rajeshkumar d, Abbas Rahdar e, *,
Ana M. Díez-Pascual f, *
a
Department of Biotechnology, Centre for Research, Kamaraj College of Engineering and Technology, S.P.G.C. Nagar, Virudhunagar, Tamil Nadu, India
b
Nano Electrochemistry Lab (NEL), Department of Chemistry, National Institute of Technology Puducherry, Karaikal 609609, India
c
Department of Food Science and Nutrition, Periyar University, Salem 636 011, Tamil Nadu, India
d
Department of Mechanical Engineering, PSG Institute of Technology and Applied Research, Coimbatore, Tamil Nadu, India
e
Department of Physics, University of Zabol, Zabol, Iran
f
Universidad de Alcalá, Facultad de Ciencias, Departamento de Química Analítica, Química Física e Ingeniería Química, Ctra. Madrid-Barcelona, Km. 33.6, 28805
Alcalá de Henares, Madrid, Spain

A R T I C L E I N F O A B S T R A C T

Keywords: The present work explores the impact of electro coagulation (EC) method on the treatment of waste from rice mill
Electrocoagulation industries using two different electrode materials (Iron (Fe) and Aluminum (Al)). The influence of different
Electrode parameters such as inter-electrode distance (4–7 cm), effluent pH (6–8), current density (10–30 mA/cm2) and
Effluent
treatment time (20–40 min) on the reduction of chemical oxygen demand (COD), total dissolved solids (TDS) and
Box-Behnken design
total soluble solids (TSS) of rice mill effluent (RME) was evaluated through batch experimental runs using Box-
Behnken design. Results reveal that the percentage removal of COD, TDS and TSS increased up to an inter-
electrode distance of 6 cm, pH of 7, current density of 20 mA/cm2 and treatment time of 30 min and then
decreased for both electrodes. In addition, mathematical models were developed for both electrodes in order to
predict the experimental data. A numerical optimization method was applied to find out the optimal operating
parameters to treat RME, and the percentage removal of COD, TDS and TSS was found to be 94.79, 96.62 and
88.76 %, using the Al electrode, as well as 76.63, 78.51 and 72.03 %, for the Fe electrode, respectively. The
comparison of the results attained demonstrate that the Al electrode is more suitable to treat RME than Fe using
EC method.

1. Introduction water consuming process, and soaking of paddy requires a huge quantity
of water. Approximately six hundred billion liters of nutrient rich
The tremendous progress in industrialization has greatly accelerated effluent was released from rice mill in every year to produce nearly 500
the release of pollutants into the environment. This could cause negative MMT of paddy [4]. Release of this effluent directly into the nearby
effects to the society and has now turned to be a major threat to the aquatic bodies or land without appropriate treatment strategy is a most
environmental sustainability [1]. The release of industrial effluent into common practice which was a severe concern for the last two decades.
the nearby water bodies could lead to severe effects like eutrophication This leads to deterioration of ground water quality and can have many
and eventually contribute to mortality of the aquatic life [2]. Rice serves adverse effects on the environment [5]. The water discharged into irri­
as the major feeding crop throughout the world. In India, rice is gated fields after soaking and cleaning can lead to health risk and can
considered as the staple food and is the integral part of the diet. Pro­ also impose threat to the quality of the crop plants due to the presence of
duction of edible rice involves a process called milling to remove husk high organic and inorganic content [6]. Due to the alkaline nature of the
and rice bran from paddy rice. Prior to milling, parboiling is performed effluent with higher content of chemical oxygen demand (COD), total
to reduce grain breakage and to avoid nutritional loss [3], which is a soluble solids (TSS) and organic matter, it cannot be used for any

* Corresponding authors.
E-mail addresses: [email protected] (A.R. Anuf), [email protected] (A. Rahdar), [email protected] (A.M. Díez-Pascual).

https://doi.org/10.1016/j.jwpe.2022.103074
Received 5 June 2022; Received in revised form 24 July 2022; Accepted 15 August 2022
Available online 22 August 2022
2214-7144/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
A.R. Anuf et al. Journal of Water Process Engineering 49 (2022) 103074

purposes without treating it with the proper method [7]. dimensions of 50 mm × 60 mm and an effective surface area of 25 cm2
Treatment of industrial effluents can be carried out by different were used individually in order to compare the influence of the electrode
techniques such as physical, chemical and biological methods. Treat­ material on the RME treatment. Proper inter-electrode distance was kept
ment using chemical methodologies like use of coagulants is ineffective to minimize the electrical resistance of the electrodes. Current density
due to the high cost of the reagent and the low COD removal rate. In was maintained at a particular level with the aid of a DC power supply
addition, the possibility of generation of secondary pollutants is high (0–30 V, 0–2 A). The cathode and anode ends were connected to the
using chemical coagulants [8]. Treatment by biological method requires power supply and the experiments were performed at room tempera­
a long-time span for the complete treatment and post treatment process. ture. A schematic representation of the experimental set up is provided
Electrocoagulation (EC) is a widely used method in wastewater treat­ in Fig. 1.
ment due to its advantages such as lower set up cost, cheaper electrode
material and shorter reaction time than the other mentioned methods, 2.3. Experimental procedure
combined with the simultaneous removal of organic and inorganic
pollutants as well as suspended solids, versatility and easiness of auto­ The collected RME was poured into the reactor and the electrodes
mation [9–11]. Moreover, it offers the benefits of coagulation, floccu­ were placed in the proper position according to the experimental design.
lation and electrochemistry in a single process [12,13]. In order to attain a uniform mixing, a magnetic stirrer was used that kept
EC unit consists of an anode and cathode material that are used to a constant stirring speed of 2000 rpm. Experiments were performed at
carry out the oxidation and reduction reaction simultaneously. When different inter-electrode distance (4–7 cm), effluent pH (6–8), current
the anode is subjected to a suitable current, dissolution of the electrodes density (10–30 mA/cm2) and treatment time (20–40 min). After each
take place and leads to the generation of metal ions, which act as batch of experiments, impurities present on the electrode surface were
effective coagulants and facilitate charge neutralization over the col­ removed by treating with hydrochloric acid solution (15 %) for 2 min
loids in an effluent [14]. The metal and hydroxyl ions generated react followed by distilled water. Settling of flocs was carried out by trans­
together to form metal hydroxide precipitates that promote colloidal ferring the effluent to another beaker and kept undisturbed for 20 min.
adsorption and the removal of contaminants from the aqueous phase The flocs were removed and the treated effluent samples were analyzed
through flocs [15]. Different process parameters like electrode material, to determine the TSS, TDS and COD content.
pH, current intensity, electrolyte concentration and process time could
influence the EC process in order to affect/improve the efficiency of 2.4. Analytical procedure
COD, total dissolved solids (TDS) and TSS removal. Therefore, optimi­
zation of these parameters would likely improve the efficiency of the EC The efficacy of the EC process using two different electrodes was
process. In the present study, Response surface methodology (RSM) has evaluated by examining the TDS, TSS and COD content following the
been applied to optimize the process variables that can influence the Standard protocol for Examination of Waste Water (APHA, AWWA
treatment process. In particular, the data obtained from the experiments 2012) after each experimental run.
were analyzed for optimization of processing parameters with respect to
the responses [16]. The use of RSM aids to find out the interaction be­ 2.5. Mathematical model development
tween independent variables, to predict the responses of the system to
any new condition by generating an appropriate mathematical model The optimization of the experimental process parameters including
and to reduce costs by decreasing the number of experiments. inter-electrode distance (4–7 cm), effluent pH (6–8), current density
Box–Behnken Design (BBD) is a commonly used RSM design that helps (10–30 mA/cm2) and treatment time (20–40 min) on EC of RME was
to avoid experiments performed under process conditions that could carried out by applying a four factor, three level Box Behnken design
yield unsatisfied experimental results [17,18]. Hence, the present study (BBD). The influence of the process variables on the selected responses
focuses on the design and development of an EC unit for the effective (COD, TDS and TSS removal) and their interaction effect were exam­
removal of COD, TDS and TSS from RME using two different electrode ined. The total number of experimental runs was calculated by applying
materials (Iron (Fe) and Aluminum (Al)), evaluates and optimizes the the following eq. [19].
impact of processing parameters (inter-electrode distance (4–7 cm),
N = 2F(F − 1) + C0 (1)
effluent pH (6–8), current density (10–30 mA/cm2) and treatment time
(20–40 min)). To the best of our knowledge, this is the first research
work on the development of an EC unit that can use two electrodes (Fe
and Al) without any modification of the experimental set up.

2. Materials and methods

2.1. Materials

The RME used in this work was trapped before being drained out
from parboiling process from a paddy processing industry located in
Madurai District, Tamilnadu, India. The effluent samples were collected
in an air-tight container and stored at 4 ◦ C to avoid any degradation
before the analysis. The effluent characteristics were analyzed using
APHA standard methods and found to be as follows: turbid, yellowish
color, pH of 6.1, TDS of 8130 mg/L, electrical conductivity of 11,614
micro-mho/cm, initial COD of 450 mg/L respectively.

2.2. Electro coagulation unit

A glass reactor with capacity of two liters that enables to place the
electrode inside was used to carry out the EC process. Two different
electrode (Fe and Al) materials with a plate thickness of 5 mm, Fig. 1. Schematic representation of the electrocoagulation experimental setup.

2
A.R. Anuf et al. Journal of Water Process Engineering 49 (2022) 103074

where F denotes the total number of factors and C0 the total number of strongly depends on the solution pH. Fig. 2 exhibits the relationship
central points. The developed design consists of 29 experimental runs between the treatment efficiency and effluent pH. It is evident that the
and the experiments were performed randomly. The functional relation percentage removal of COD, TDS and TSS removal increased linearly
between the selected responses and the independent variables was with increasing initial pH and reached a maximum at a pH of 7. This is
evaluated by a non-linear regression method using a second order attributed to the increase in the formation of Al(OH)3/Fe(OH)3 ions at
polynomial equation as shown below: such pH. The percentage removal of COD, TDS and TSS improved from
48.13 to 74.11 %, 54.14 to 78.29 % and 46.60 to 72.75 %, respectively,
∑
k ∑
k k ∑
∑ k
using the Fe electrode. Similar results were observed using the Al elec­
Z = β0 + βb xb + βbb x2b + βab xa xb + e (2)
a
trode: the percentage removal increased significantly, from 68.50 to
b=1 b=1 <j=2
94.72 % for COD, 68.75 to 96.11 % for TDS and 88.33 to 89.24 % for
where Z is the selected response; xa and xb are variables (a and b range TSS. Thus, the electro coagulation performed at a suitable pH promotes
from 1 to k); β0 is the model intercept coefficient; βa, βb and βab are the effective and immediate adsorption of all soluble organic content in the
interaction coefficients of linear, quadratic and the second-order terms; effluent and their entrapment as colloidal material, which in turn fa­
k is the number of independent parameters and ei the error. cilitates the reduction of COD, TDS and TSS. A further increase in the pH
The efficacy of the developed model was investigated using ANOVA. results in the formation of Al(OH) 4 -/Fe(OH) 4 - that dissolves rapidly in
Fisher F-test was used to determine the statistical significance of the the effluent and prevents the formation of flocs [20,14], thereby
developed model. The Probability value was used to evaluate model reducing the percentage removal of COD, TDS and TSS.
terms with a 95 % confidence level. The entire statistical analysis was
done using Design Expert Statistical Software package 13.0.0 (Stat Ease 3.1.2. Effect of treatment time
Inc., Minneapolis, USA). Treatment time plays a crucial effect on the effluent treatment using
EC process since it determines the cost effectiveness of the entire process
[21]. Electrolysis facilitates the release of coagulants due to electro-
3. Results and discussions
dissolution of anodes. The process removal efficiency depends on the
metal ion concentration generated from the electrodes. Upon increasing
The interaction effect between four independent experimental pa­
reaction time, the metal ion concentration and hydroxide flocs increase.
rameters (detailed in Section 2.3) on the EC of the RME was studied
In the current study, the reaction time was ranged from 20 to 40 min.
using two different electrodes (Fe and Al) applying BBD. Twenty-nine
The influence of reaction time on electro coagulation is displayed in
statistically designed experiments were performed (Table 1) to deter­
Fig. 3. Results demonstrate that on increasing reaction time from 20 to
mine the optimum parameters for the EC process.
32 min, the percentage removal of COD, TDS and TSS improved from
62.33 to 74.11 %, 66.95 to 77.12 % and 60.72 to 71.59 % respectively
3.1. Effect of process parameters on the EC of the RME for the Fe electrode. Similar response was observed using the Al elec­
trode: the percentage removal of COD, TDS and TSS increased from
3.1.1. Effect of initial pH 82.97 to 94.33 %, 82.06 to 96.11 % and 76.18 to 88.53 %, respectively,
To evaluate the influence of the effluent initial pH on COD, TDS and which is mainly attributed to the increase in metal ion formation at the
TSS removal efficiency, this parameter was adjusted prior to each initial stage of the electrolysis.
experimental run using sodium hydroxide (NaOH) or hydrochloric acid
(HCl) solutions. The generation of metal hydroxides and their stability 3.1.3. Effect of current density
Current density determines the efficiency of the EC process, which
Table 1 influences the coagulant dosage and bubble generation rates, and aids to
BBD design for the EC treatment of the RME. increase rapidly pollutant removal [22,23]. The ratio of the electrical
current to the electrode surface area is defined as current density. The
Std. order pH (X1) TT (min, X2) CD (mA/cm2, X3) IED (cm, X4)
effect of the applied current density on the percentage removal of COD,
1 6 20 20 5.5
TDS and TSS was investigated, and the results are included in Fig.4.
2 8 20 20 5.5
3 6 40 20 5.5
Current density directly influences the effluent mixing property and the
4 8 40 20 5.5 mass transfer rate at electrode surface. For both Fe and Al electrodes, it
5 7 30 10 4 was found that the treatment efficiency improved as the initial current
6 7 30 30 4 density was increased from 10 to 21 mA/cm2. Further rise in the current
7 7 30 10 7
density decreased significantly the percentage removal of TDS, TSS and
8 7 30 30 7
9 6 30 20 4 COD using both electrodes. This behavior might be attributed to sec­
10 8 30 20 4 ondary reactions that occur at high current density, which lead to colloid
11 6 30 20 7 charge reversal and thus cause re-dispersion of the colloids. Moreover,
12 8 30 20 7 higher current density could also result in a reduction of the electrode
13 7 20 10 5.5
14 7 40 10 5.5
lifetime [24].
15 7 20 30 5.5
16 7 40 30 5.5 3.1.4. Effect of inter-electrode distance
17 6 30 10 5.5 Inter-electrode distance is a key process variable that controls the
18 8 30 10 5.5
efficiency of the EC process [25]. In the current study, the effect of inter-
19 6 30 30 5.5
20 8 30 30 5.5 electrode distance on EC was studied by modifying the distance from 4
21 7 20 20 4 to 7 cm at an interval of 1.5 cm, and the responses are depicted in Fig. 5.
22 7 40 20 4 An effective response in treatment efficiency was observed for the
23 7 20 20 7 shortest electrode distances. As the inter-electrode distance was
24 7 40 20 7
25 7 30 20 5.5
increased linearly from 4 to 5.9 cm, the percentage removal of COD, TDS
26 7 30 20 5.5 and TSS improved from 45.11 to 75.72 %, 55.75 to 78.71 % and 47.09 to
27 7 30 20 5.5 71.59 %, respectively, using the Fe electrode. Similar response was
28 7 30 20 5.5 observed using the Al electrode: the percentage removal of COD, TDS
29 7 30 20 5.5
and TSS improved from 67.90 to 94.33 %, 71.49 to 96.11 % and 62.86 to

3
A.R. Anuf et al. Journal of Water Process Engineering 49 (2022) 103074

Fig. 2. Effect of pH on COD, TDS and TSS removal using Fe and Al electrodes.

4
A.R. Anuf et al. Journal of Water Process Engineering 49 (2022) 103074

Fig. 3. Effect of treatment time on COD, TDS and TSS removal using Fe and Al electrodes.

5
A.R. Anuf et al. Journal of Water Process Engineering 49 (2022) 103074

Fig. 4. Effect of current density on COD, TDS and TSS removal using Fe and Al electrodes.

6
A.R. Anuf et al. Journal of Water Process Engineering 49 (2022) 103074

Fig. 5. Effect of inter-electrode distance on COD, TDS and TSS removal using Fe and Al electrodes.

7
A.R. Anuf et al. Journal of Water Process Engineering 49 (2022) 103074

88.54 %, respectively, as the electrode distance increased from 4 to 5.8 TSS, respectively, for the Fe electrode, and 970.07, 114.37 and 59.92 for
cm. This behavior could be attributed to a rapid rise in the anion the Al electrode, respectively,) and its associated lower p-values (p <
discharge at the anode surface and an increased oxidation rate [26]. At 0.0001) demonstrate the validity of the developed mathematical models
higher inter-electrode distances (5.9 cm for the Fe electrode and 5.8 cm (Table 2).
for the Al one), a decrease in treatment efficiency was observed for both The statistical analysis yielded the following values for the coeffi­
electrodes. This could be ascribed to the fact that the floc distance cient of determination (R2): 0.993, 0.996 and 0.981 for the percentage
growths with increasing inter-electrode distance and could eventually removal of COD, TDS and TSS using the Fe electrode, as well as 0.999,
lead to a decrease in COD, TDS and TSS removal [27]. At higher dis­ 0.991 and 0.983 for the Al electrode. Analogously, the values of the
tances the effective interaction between the oxidants and coagulants adjusted coefficient of determination (adj-R2) were: 0.986, 0.993 and
could become weaker, hence leading to a drop in the process efficiency 0.962 for the percentage removal of COD, TDS and TSS using the Fe
[28]. electrode, as well as 0.997, 0.983 and 0.967 for the Al electrode.
Regarding the predicted coefficient of determination (pre-R2), the
values obtained were: 0.964, 0.983 and 0.893 for the percentage
3.2. Mathematical model development removal of COD, TDS and TSS using the Fe electrode, as well as 0.995,
0.951 and 0.906 for the Al electrode. These coefficients measure how
The results obtained from BBD were evaluated using a multiple well a statistical model predicts an outcome. The better a model is at
regression analysis method. The relationship between the independent making predictions, the closer these coefficients will be to 1. Therefore,
process parameters and the responses were expressed using three the very high values obtained herein corroborate that the selected model
different empirical models derived from the experimental data in order is the most appropriate to depict the relationship between the process
to assess the correlation between the responses and experimental pa­ parameters and the responses. The percentage coefficient of variance
rameters. The final model, in terms of coded factors, is shown below: (CV) values were also calculated: 3.78, 1.82 and 5.89 for the percentage
For the Fe electrode removal of COD, TDS and TSS for the Fe electrode, as well as 0.992, 2.89

Y1 = 74.59 + (9.61*X1 ) − (1.17* X2 ) − (4.92*X3 ) + (7.37*X4 ) + (12.76*X1 X2 ) + (3.25*X1 X3 ) + (0.32*X1 X4 ) − (4.10*X2 X3 ) − (5.60*X2 X4 )
( ) ) ( ) ( )
− (2.62*X3 X4 ) − 17.80*X1 2 − 13.43*X2 2 − 6.81*X3 2 − 21.24*X4 2 (3)

Y2 = 77.13 + (8.31*X1 ) − (0.86*X2 ) − (2.41*X3 ) + (5.94*X4 ) + (13.91*X1 X2 ) + (4.40*X1 X3 ) − (1.22*X1 X4 ) − (3.08*X2 X3 ) − (6.37*X2 X4 )
( ) ( ) ( ) ( )
− (2.93*X3 X4 ) − 14.67*X1 2 − 11.03*X2 2 − 0.86*X3 2 − 14.25*X4 2 (4)

Y3 = 71.59 + (8.65*X1 ) − (0.93*X2 ) − (3.45*X3 ) + (5.76*X4 ) + (10.43*X1 X2 ) + (5.07*X1 X3 ) − (0.73*X1 X4 ) − (3.93*X2 X3 ) − (8.68*X2 X4 )
( ) ( ) ( ) ( )
− (1.59*X3 X4 ) − 16.36*X1 2 − 11.80*X2 2 − 8.40*X3 2 − 18.76*X4 2 (5)

For the Al electrode

Y1 = − 1028.46 + (201.79*X1 ) + (1.89*X2 ) + (2.51*X3 ) + (118.15*X4 ) + (1.25*X1 X2 ) + (0.37*X1 X3 ) − (0.30*X1 X4 ) − (0.038*X2 X3 ) − (0.387*X2 X4 )
( ) ( ) ( ) ( )
− (0.19*X3 X4 ) − 16.87*X1 2 − 0.13*X2 2 − 0.07*X3 2 − 8.72*X4 2 (6)
(
Y2 = − 1082.16 + (206.20*X1 ) + (3.85*X2 ) + (1.71*X3 ) + 121.78*X4) + (1.14*X1 X2 ) + (0.50*X1 X3 ) − (0.14*X1 X4 )*(0.053*X1 X3 *0.58*X2 X4 )
( ) ( ) ( ) ( )
− (0.14*X3 X4 ) − 17.18*X1 2 − 0.12*X2 2 − 0.07*X3 2 − 8.75*X4 2 (7)

Y3 = + 96.12 + (8.39*X1 ) − (1.13*X2 ) − (1.99*X3 ) + (6.10*X4 ) + (12.96*X1 X2 ) + (4.67*X1 X3 ) − (1.90*X1 X4 ) − (3.41*X2 X3 ) − (8.27*X2 X4 )
( ) ( ) ( ) ( )
− (3.61*X3 X4 ) − 19.01*X1 2 − 15.18*X2 2 − 5.79*X3 2 − 18.51*X4 2 (8)

and 4.36 for the Al electrode. The low values obtained indicate a high
where,Y1, Y2, Y3, denote the COD (%), TDS (%), TSS (%) removal, and reliability in the experimental results. The value of adequate precision
X1, X2, X3 and X4 denote the pH, reaction time, current density and was higher than 26 for all the responses, which demonstrates the ac­
electrode distance, respectively. curacy of the developed models.
Pareto analysis of variance (ANOVA) was performed in order to
evaluate the experimental derived results. The higher model F-value
(150.73, 303.79 and 52.49 for the percentage removal of COD, TDS and

8
A.R. Anuf et al. Journal of Water Process Engineering 49 (2022) 103074

Table 2
ANOVA for the responses.
Fe electrode Al electrode

COD (%) TDS (%) TSS (%) COD (%) TDS (%) TSS (%)

Source F value p-value F value p-value F value p-value F value p-value F value p-value F value p-value

Model 150.73 <0.0001 303.79 <0.0001 52.49 <0.0001 970.07 <0.0001 114.37 <0.0001 59.92 <0.0001
X1 310.13 <0.0001 686.25 <0.0001 109.33 <0.0001 1970.92 <0.0001 194.73 <0.0001 125.45 <0.0001
X2 4.6 0.05 7.29 0.0173 1.39 0.2577 91.22 <0.0001 3.51 0.0818 0.1516 0.7029
X3 81.36 <0.0001 57.47 <0.0001 16.93 0.0011 200.76 <0.0001 11.01 0.0051 16.67 0.0011
X4 182.45 <0.0001 350.25 <0.0001 48.54 <0.0001 1137.64 <0.0001 103.15 <0.0001 53.93 <0.0001
X12 182.17 <0.0001 640.65 <0.0001 53 <0.0001 1288.04 <0.0001 154.97 <0.0001 65.59 <0.0001
X13 11.83 0.004 64.08 <0.0001 12.54 0.0033 113.65 <0.0001 20.11 0.0005 12.64 0.0032
X14 0.1127 0.742 4.91 0.0438 0.2579 0.6195 1.75 0.2068 3.34 0.089 0.0232 0.8812
X23 18.75 0.0007 31.35 <0.0001 7.06 0.0187 118.74 <0.0001 10.73 0.0055 14.27 0.002
X24 35.07 <0.0001 134.3 <0.0001 36.71 <0.0001 276.73 <0.0001 63.09 <0.0001 38.79 <0.0001
X34 7.65 0.0152 28.32 0.0001 1.23 0.2858 69.48 <0.0001 12.03 0.0038 2.43 0.1417
X21 574.66 <0.0001 1155.24 <0.0001 210.94 <0.0001 3781.29 <0.0001 540.63 <0.0001 240.13 <0.0001
X22 327.01 <0.0001 652.81 <0.0001 110.78 <0.0001 2344.68 <0.0001 344.74 <0.0001 130.5 <0.0001
X23 84.18 <0.0001 3.97 0.0663 56.24 <0.0001 846.62 <0.0001 50.18 <0.0001 50.55 <0.0001
X24 818.04 <0.0001 1089.47 <0.0001 277.28 <0.0001 5116.95 <0.0001 513.04 <0.0001 315.06 <0.0001
R2 0.9934 0.9967 0.9813 0.999 0.9913 0.9836
Adj- R2 0.9868 0.9934 0.9626 0.9979 0.9827 0.9672
Pre- R2 0.9644 0.9836 0.8931 0.9955 0.9519 0.9068
CV (%) 3.78 1.82 5.89 0.992 2.89 4.36
Adeq.Pre. 40.95 61.70 24.37 106.61 37.84 25.70

3.3. Diagnostics of model adequacy into account the practical feasibility, the optimal conditions were
slightly modified, and experiments were carried out under these con­
In general, the model needs to be validated in order to confirm that it ditions: a pH of 7.3, a current density of 19 mA/cm2, an inter-electrode
provides an accurate approximation to the actual (experimental) values. distance of 5.8 cm, and an electrolysis time of 32 min, for the Fe elec­
Investigating and optimizing the parameter without evaluating the trode, as well as a pH of 7.3, a current density of 21 mA/cm2, an inter-
satisfactory fitness of the model could lead to misleading and poor re­ electrode distance of 5.6 cm, and an electrolysis time of 31 min for the Al
sults. In this regard, diagnostic plots like normalized plot and the parity electrode. Results revealed that the percentages of removal of COD, TDS
plot between predicted and actual values aid to validate the satisfactory and TSS (76.63, 78.56 and 72.03 for the Fe electrode, as well as 94.79,
fitness of the developed model and to analyze the relationship between 96.62 and 88.76 for the Al electrode, respectively) were very close to the
the actual and predicted values. The parity plots for the selected re­ predicted values.
sponses and the two electrodes are provided in Fig. 6. As can be
observed, for both electrodes, the values lie close to a straight line,
which corroborates that the experimental derived results are in accor­ 3.5. Cost analysis (CA)
dance with the predicted ones. Overall, results suggest that the devel­
oped models for the Fe and Al electrodes can be applied to identify the The following data was used for estimating the operating cost:
optimal experimental conditions that provide the best percentage Rectifier installation cost (800 $), EC tank installation cost (100 $ for a
reduction of COD, TDS and TSS. capacity of 100 m3/day), maintenance cost (0.003 $/m3), electricity cost
Additionally, the normality of the residuals was examined. The (0.085 $/h), labor cost (0.005 $/m3), transportation and disposal of
normal distribution of the residuals for the different responses selected sludge (0.01$/kg), Fe and Al electrode cost (3.41$/kg and 2.23$/kg,
was assessed using the normal probability plot, as depicted in Fig. 7 for respectively). Cost calculation was derived from the equation described
both electrodes. As can be observed, the values lie in a straight line and by Sridhar et al. [29] taking into account the modified optimal condi­
the residuals are small, which corroborates that the data were normally tions for both electrodes detailed in the previous section. Overall, the
distributed, hence that the developed models are able to predict the data operating cost for the Fe electrode was 4.27 $/m3 of effluent, while for
with high precision. the Al one was 2.36$/m3.

4. Conclusion
3.4. Optimization and authentication of process parameters and responses
Environmental concerns are the major driving force accounting for
Derringer's desired function methodology was applied to identify the the development of novel strategies for treatment of effluents. EC pro­
optimum experimental conditions to attain the highest percentage cess was successfully used to treat the RME using Fe and Al electrodes. A
removal of COD, TDS, and TSS for both electrodes. The best conditions four factor, three level BBD was used to investigate and optimize the
for the Fe electrode were found to be a pH of 7.29, a current density of process parameters (inter-electrode distance (4–7 cm), effluent pH
18.64 mA/cm2, an inter-electrode distance of 5.8 cm, and an electrolysis (6–8), current density (10–30 mA/cm2) and treatment time (20–40
time of 31.33 min, which led to percentage removals of 77.04, 79.18 and min)). All the process parameters had a significant effect on the per­
73.13 for COD, TDS and TSS, respectively. Analogously, using the Al centage removal of COD, TDS and TSS from RME using both electrodes.
electrode, a pH of 7.29, a current density of 20.89 mA/cm2, an inter- Experimental data were statistically analyzed, and mathematical models
electrode distance of 5.6 cm, and an electrolysis time of 30.92 min were were developed for the responses (COD, TDS and TSS). The optimal
the optimal conditions, that led to the highest percentages of removal: conditions to treat the RME using EC process were derived and vali­
95.46, 97.11 and 89.83 for COD, TDS and TSS. the Al electrode. Taking dated. Validation experiments carried out under the optimal condition

9
A.R. Anuf et al. Journal of Water Process Engineering 49 (2022) 103074

Fig. 6. Parity plots between predicted and actual values for the different responses using Fe and Al electrodes.

10
A.R. Anuf et al. Journal of Water Process Engineering 49 (2022) 103074

Fig. 7. Normal probability plots for the different responses using both electrodes.

revealed that the percentages of removal of COD, TDS and TSS were Funding
higher for the Al electrode (94.79, 96.62 and 88.76, respectively) than
for the Fe electrode (76.63, 78.56 and 72.03). Similarly, the calculated Financial support from the Community of Madrid within the
operating cost was lower for the Al electrode when compared with the framework of the multi-year agreement with the University of Alcalá in
Fe electrode. Results clearly demonstrate that the EC process using the the line of action “Stimulus to Excellence for Permanent University
Al electrode can be applied to effectively reduce the COD, TDS and TSS Professors”, Ref. EPU-INV/2020/012, is gratefully acknowledged.
content, and the treated water could also be used for agriculture
purposes.

11
A.R. Anuf et al. Journal of Water Process Engineering 49 (2022) 103074

Declaration of competing interest [13] B. Mallesh, A review of electrocoagulation process for wastewater treatment, Int. J.
ChemTech Res. 11 (2018) 289–302.
[14] E. El-Ashtoukhy, Y. El-Taweel, O. Abdelwahab, E. Nassef, Treatment of
The authors declare the following financial interests/personal re­ petrochemical wastewater containing phenolic compounds by electrocoagulation
lationships which may be considered as potential competing interests: using a fixed bed electrochemical reactor, Int. J. Electrochem. Sci. 8 (2018)
Ana Maria Diez Pascual reports financial support was provided by 1534–1550.
[15] I.D. Tegladza, Q. Xu, K. Xu, G. Lv, J. Lu, Electrocoagulation processes: a general
University of Alcala. review about role of electro-generated flocs in pollutant removal, Process Saf.
Environ. Prot. 146 (2021) 169–189.
Data availability [16] S. Karimifard, M.R.A. Moghaddam, Application of response surface methodology in
physicochemical removal of dyes from wastewater: a critical review, Sci. Total
Environ. 640 (2018) 772–797.
Data will be made available on request. [17] F. Erraib, K. El Ass, Optimization of electrocoagulation operating parameters for
COD removal from olive mill wastewater: application of Box-Behnken design,
Mediterr.J.Chem. 9 (2019) 212–221.
References [18] M. Bajpai, S.S. Katoch, Reduction of COD from real graywater by electro-
coagulation using Fe electrode: optimization through Box-Behnken design, Mater.
[1] T. Shindhal, P. Rakholiya, S. Varjani, A. Pandey, H.H. Ngo, W. Guo, H.Y. Ng, M. TodayProc. 43 (2018) 303–307.
J. Taherzadeh, A critical review on advances in the practices and perspectives for [19] K. Thirugnanasambandham, V. Sivakumar, Removal of ecotoxicological matters
the treatment of dye industry wastewater, Bioengineered 12 (2021) 70–87. from tannery wastewater using electrocoagulation reactor: modelling and
[2] S. Bhuvaneshwari, F. Majeed, E. Jose, A. Mohan, Different treatment optimization, Desalin. Water Treat. 57 (2016) 3871–3880.
methodologies and reactors employed for dairy effluent treatment-a review, J. [20] K. Thirugnanasambandham, V. Sivakumar, J.P. Maran, Response surface modelling
Water Process Eng. 46 (2022), 102622. and optimization of treatment of meat industry wastewater using electrochemical
[3] G. Behera, P. Sutar, A comprehensive review of mathematical modeling of paddy treatment method, J. Taiwan Inst. Chem. Eng. 46 (2015) 160–167.
parboiling and drying: effects of modern techniques on process kinetics and rice [21] A. Tahreen, M.S. Jami, F. Ali, Role of electrocoagulation in wastewater treatment: a
quality, Trends Food Sci. Technol. 75 (2018) 206–230. developmental review, J.Water Process Eng. 37 (2020), 101440.
[4] J. Umamaheswari, S. Shanthakumar, Paddy-soaked rice mill wastewater treatment [22] R. Katal, H. Pahlavanzadeh, Influence of different combinations of aluminum and
by phycoremediation and feasibility study on use of algal biomass as biofertilizer, iron electrode on electrocoagulation efficiency: application to the treatment of
J. Chem. Technol. Biotechnol. 96 (2021) 394–403. paper mill wastewater, Desalination 265 (2011) 199–205.
[5] Z. Arif, N.K. Sethy, P. Mishra, B. Verma, in: Impact on groundwater quality [23] C. Akarsu, E.U. Deveci, C. Gönen, O. Madenli, Treatment of slaughterhouse
resources due to industrial effluent. groundwater geochemistry: pollution and wastewater by electrocoagulation and electroflotation as a combined process:
remediation methods, 2021, pp. 212–231. process optimization through response surface methodology, Environ. Sci. Pollut.
[6] K. Thirugnanasambandham, V. Sivakumar, J. Prakash Maran, Modeling and Res. 28 (2021) 34473–34488.
optimization of biogas production from rice mill effluent using up flow anaerobic [24] J. Lu, P. Zhang, J. Li, Electrocoagulation technology for water purification: an
sludge blanket reactor, J.Renew.Sustain.Energy 6 (2014), 023129. update review on reactor design and some newly concerned pollutants removal,
[7] S. Kumar, S. Deswal, A review on current techniques used in India for rice mill J. Environ. Manag. 296 (2021), 113259.
wastewater treatment and emerging techniques with valuable by-products, [25] M. Negarestani, M. Motamedi, A. Kashtiaray, A. Khadir, M. Sillanpää,
Environ. Sci. Pollut. Res. 28 (2021) 7652–7668. Simultaneous removal of acetaminophen and ibuprofen from underground water
[8] A.K. Badawi, K. Zaher, Hybrid treatment system for real textile wastewater by an electrocoagulation unit: operational parameters and kinetics, Groundw.
remediation based on coagulation/flocculation, adsorption and filtration processes: Sustain. Dev. 11 (2020), 100474.
performance and economic evaluation, J.Water Process Eng. 40 (2021), 101963. [26] D. Bhagawan, S. Poodari, T. Pothuraju, D. Srinivasulu, G. Shankaraiah, M.Y. Rani,
[9] G. Jing, S. Ren, S. Pooley, W. Sun, P.B. Kowalczuk, Z. Gao, Electrocoagulation for V. Himabindu, S. Vidyavathi, Effect of operational parameters on heavy metal
industrial wastewater treatment: an updated review, Environ.Sci.Water Res. removal by electrocoagulation, Environ. Sci. Pollut. Res. 21 (2014) 14166–14173.
Technol. 7 (2021) 1177–1196. [27] S. Aoudj, A. Khelifa, N. Drouiche, M. Hecini, H. Hamitouche, Electrocoagulation
[10] M. Mousazadeh, E.K. Niaragh, M. Usman, S.U. Khan, M.A. Sandoval, Z. Al-Qodah, process applied to wastewater containing dyes from textile industry, Chem. Eng.
Z.B. Khalid, V. Gilhotra, M.M. Emamjomeh, A critical review of state-of-the-art Process. Process Intensif. 49 (2010) 1176–1182.
electrocoagulation technique applied to COD-rich industrial wastewaters, Environ. [28] R. Niazmand, M. Jahani, S. Kalantarian, Treatment of olive processing wastewater
Sci. Pollut. Res. 28 (2021) 43143–43172. by electrocoagulation: an effectiveness and economic assessment, J. Environ.
[11] Zakaria Al-Qodah, Mohammad Al-Shannag, On the performance of free radicals Manag. 248 (2014), 109262.
combined electrocoagulation treatment processes, Sep.Purif.Rev. 48 (2019) [29] R. Sridhar, S. Venkatachala, I. Prince, J. Prakash, J. Maran, Treatment of pulp and
143–158. paper industry bleaching effluent by electrocoagulant process, J.Hazard.Mater. 86
[12] S.D.U. Islam, Electrocoagulation (EC) technology for wastewater treatment and (2011) 1495–1502.
pollutants removal, Sustain.Water Resour.Manag. 5 (2019) 359–380.

12
 Update
Journal of Water Process Engineering
Volume , Issue , , Page

DOI: https://doi.org/10.1016/j.jwpe.2024.105436
 Journal of Water Process Engineering xxx (xxxx) xxx

Contents lists available at ScienceDirect

Journal of Water Process Engineering

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

Corrigendum

Corrigendum to “Optimization of electrocoagulation process for treatment

of rice mill effluent using response surface methodology” [J. Water Process
Eng., volume 49 (October 2022) 103074]
A. Ronaldo Anuf a, *, K. Ramaraj a, Vishnu Sankar Sivasankarapillai b,
Ragupathy Dhanusuraman b, J. Prakash Maran c, G. Rajeshkumar d, Abbas Rahdar e, *,
Ana M. Díez-Pascual f, *
a
Department of Biotechnology, Centre for Research, Kamaraj College of Engineering and Technology, S.P.G.C. Nagar, Virudhunagar, Tamil Nadu, India
b
Nano Electrochemistry Lab (NEL), Department of Chemistry, National Institute of Technology Puducherry, Karaikal 609609, India
c
Department of Food Science and Nutrition, Periyar University, Salem 636 011, Tamil Nadu, India
d
Department of Mechanical Engineering, PSG Institute of Technology and Applied Research, Coimbatore, Tamil Nadu, India
e
Department of Physics, Faculty of Science, University of Zabol, 538-98615 Zabol, Iran
f
Universidad de Alcalá, Facultad de Ciencias, Departamento de Química Analítica, Química Física e Ingeniería Química, Ctra. Madrid-Barcelona, Km. 33.6, 28805
Alcalá de Henares, Madrid, Spain

The authors regret that there was a mistake in the funding. This “The authors declare that they have no known competing financial
research received no funding. interests or personal relationships that could have appeared to influence
Additionally, there was a mistake in section “Declaration of the work reported in this paper”.
competing interest”. The authors would like to apologise for any inconvenience caused.
It should be as follows:

DOI of original article: https://doi.org/10.1016/j.jwpe.2022.103074.

* Corresponding authors.
E-mail addresses: [email protected] (A.R. Anuf), [email protected] (A. Rahdar), [email protected] (A.M. Díez-Pascual).

https://doi.org/10.1016/j.jwpe.2024.105436

2214-7144/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).

Please cite this article as: A. Ronaldo Anuf et al., Journal of Water Process Engineering, https://doi.org/10.1016/j.jwpe.2024.105436