Electrooxidation and subcritical water oxidation

hybrid process for pistachio wastewater treatment

Zelal Isik
Mersin University: Mersin Universitesi
Zhalaladdin Jabbiyev
Mersin University: Mersin Universitesi
Raouf Bouchareb
Mersin University: Mersin Universitesi
Yasin Ozay
Tarsus Üniversitesi: Tarsus Universitesi
Erdal Yabalak
Mersin University: Mersin Universitesi
Nadir Dizge (  [email protected] )
Mersin University: Mersin Universitesi

Research Article

Keywords: Pistachio wastewater, electrooxidation, subcritical water oxidation, RSM

Posted Date: May 27th, 2022

DOI: https://doi.org/10.21203/rs.3.rs-1479310/v1

License:   This work is licensed under a Creative Commons Attribution 4.0 International License.
Read Full License

Page 1/23
Abstract
In this study, a hybrid system was used for pistachio wastewater treatment based on the electrooxidation
treatment process and subcritical water oxidation. The effects of current density (50–150 A/m 2 ),
experiments time (0–180 min), and wastewater initial pH (4-8) were optimized for maximum total phenols
and chemical oxygen demand (COD) removal. The experimental study proved that the best conditions of
current density and solution pH for COD and total phenol removal efficiencies were 150 A/m 2 and
original pH of 5 giving removal efficiencies of 50% and 19.02%, respectively. The electrochemically
pretreated wastewater was subjected to subcritical water oxidation (SWO). The effects of temperature
(376–410 K), treatment time (26.4-93.65 min), and concentration of H 2 O 2 (0.08-0.92 M) were examined
for COD and total phenol removal. The experimental and predicted COD and total phenol removal of the
central composite design (CCD) models demonstrated that temperature, treatment time and concentration
of H 2 O 2 have highly affected each of COD and total phenol removals. The highest removal efficiencies
were as much as 54.3% and 76.87% for COD and total phenol at the following optimum conditions:
temperature of 403 K, 40 min of treatment time and 0.75 M of H 2 O 2 concentration. The SWO results
showed that, although treatment time affects the removal yields, it is not as effective as hydrogen
peroxide concentration. In addition, it was found that COD and total phenol removal yields at the lowest
values of all variables remained at 18.36% and 44.58%, respectively. Besides, they increased to 42.17%
and 53.17%, respectively, by increasing only treatment time to its highest level. However, further increasing
the treatment time after a specific level, may not increase the removal yields. Nevertheless, higher removal
rates in both COD and total phenol were achieved even in the lowest level of the treatment time, but the
highest level of the other two variables.

1. Introduction
Pistachio nuts (Pistacia vera, Anacardiaceae family) are valuable food and are largely consumed
worldwide as a result of their dietary quality and health-related benefits (Gür and Demirer 2019).
According to the latest statistics of 2020 reported by the International Nut and Dried Food Council (INC),
over a million metric tons of pistachio was produced worldwide (Fil et al. 2014, Ozay et al. 2018, Isik et al.
2020). Iran and the USA are the leading producers, followed by Turkey as the third biggest pistachio
producer in the world with a yearly production of 240,000 tons. For processing one ton of pistachio,
approximately 06 m3 of water is necessary (01 m3 for spalling, 04 m3 for paring, and 10 m3 for washing),
which is directly discharged to the ecosystem (Bayar et al. 2014, Gür and Demirer 2019, Isik et al. 2020).
Pistachio processing wastewater contains a high level of undesirable toxic contaminants causing high
values of total phenols, chemical oxygen demand (COD), and total organic carbon (TOC) (Bayar et al.
2018, Ozay et al. 2018, Gür and Demirer 2019, Isik et al. 2020).

Until now, different physicochemical and biological technologies have been reported for industrial
wastewater treatment for instance flotation (Elazzouzi et al. 2019), coagulation/flocculation (Bouchareb
et al. 2020), electrocoagulation (Isik et al. 2021), precipitation (Xiao et al. 2018), oxidation (Shan et al.
2019), evaporation (Toth et al. 2018), solvent extraction (Al-Doury 2019), membrane filtration (Bouchareb
Page 2/23
et al. 2020), adsorption (de Caprariis et al. 2017), ion exchange (Wang et al. 2018), and
biodegradation (Bouchareb et al. 2021). However, it is always a challenge for the treatment method
selection since there is no best absolute and/or specific technique of water treatment. Each treatment
method has got its particular advantages and disadvantages not only concerning cost, however mostly
for the chosen method efficiency, practicability and environmental impact (Crini and Lichtfouse 2019).
Due to the complex nature of industrial discharges, a single treatment method is incapable of adequate
treatment and meeting the desired water quality standards. In addition, more biological treatment
approaches remain ineffective to overwhelmed the mineralization of difficult-to-degrade pollutants and
physicochemical methods cause the creation of the hazardous intermediary product (Yabalak 2018a).
Therefore, highly effective hybrid or combined techniques are necessary to degrade the pollutants in the
water and achieve recommended water quality in the most efficient and economical approach (Crini and
Lichtfouse 2019, Bouchareb et al. 2020, Isik et al. 2021).

Lately, there has been pronounced interest in the improvement of useful electrochemical techniques for
the destruction of pollutant contents present in manufacturing wastewater. Electrooxidation (EO) of these
contaminants is achieved through diverse methods (Shan et al. 2019). For instance, the indirect
electrooxidation process uses hypochlorite and chlorine generated anodically to destruct organic
contaminants (Rahmani et al. 2015a). Pollutants are also degraded by electrochemically produced
hydrogen peroxide (Frangos et al. 2016). Contaminants direct anodic oxidation can also take place
straight on anodes by producing physically adsorbed “active oxygen” (oxygen in the oxide lattice, MOx+1)
over a process named direct or anodic oxidation (Frangos et al. 2016). Direct oxidation does not require
further chemicals or oxygen and does not generate secondary contaminants or involve complicated
accessories (Martínez-Huitle and Ferro 2006). The most significant constituent in the anodic oxidation
progression is the anode itself. The most common anode materials for EO are carbon fiber, glassy carbon
(Ti/RuO2, Ti/Pt–Ir), and stainless steel. On the other hand, not any of these materials has the combination
of both stability and activity necessary for efficiency and durability to be applied. In recent years, anodes
made of boron-doped conductive diamond have been used in electrooxidation and seem to be one of the
most capable technologies in industrial wastewater treatment (Chen 2004). Compared to other electrode
substances, conductive diamond has shown higher efficiency, stability and over potential for both
hydrogen and oxygen development. These properties have led to the utilization of diamond electrodes in
decolorization of toxic dyes solutions, electrosynthesis, oxidation of benzoic and carboxylic acids,
oxidation of organic contaminants, degradation of surfactant, breakdown of phenolic aqueous wastes,
and degradation of triazines. It is important to draw attention that this system reaches the total
mineralization of contained organics in wastewater (Linares-Hernández et al. 2010).

The environmentally eco-friendly subcritical water oxidation (SWO) method, which is a thermochemical
technique and delivers resistant degradation and complex organic pollutants, is one of the most efficient
methods in demand (Yabalak 2018a). Exposed elements are degraded to insignificant and non-harmful
constituents, in the end to water and carbon dioxide deprived of any residue by the SWO method (Yabalak
et al. 2018). In this method, the production of hydroxyl and other radical species occurs at high
temperatures and pressures (Yabalak 2018b). Subcritical water is the a hot (373 K–647 K) and
Page 3/23
pressurized water to a degree to preserve it in the liquid phase at a working temperature. H2O2 is acting as
a green oxidizing agent, however, H2O2 oxidation effect alone in the mineralization of target contaminants
stays relatively low. On the other hand, H2O2 synergistically contributes to the free radicals’ formation in
the subcritical water environment. Numerous studies such as oxidation, extraction, solubility and organic
synthesis have been conducted in a subcritical water environment due to its impressive advantages for
instance availability and cheapness, having adaptable polarity as well as being an eco-friendly
solvent (Yabalak 2018a).

This research study engrossed on investigating the feasibility and performance of pistachio wastewater
treatment by electrooxidation and the subcritical water oxidation hybrid process. The effects of current
density (from 50 to 150 A/m2), operating time (from 0 to 180 min), and initial wastewater pH (from 4 to 8)
were examined removal of COD and total phenols, as well as recording the changes in conductivity for the
electrooxidation treatment process. Moreover, electrochemically pre-treated wastewater was subjected to
subcritical water oxidation. The effects of temperature (110–130 °C), treatment time (40–80 min), and
concentration of H2O2 (0.25-0.75 M) were examined removal of COD and total phenol for SWO treatment
process.

2. Material And Methods

2.1. Pistachio wastewater characterization

Pistachio wastewater characterization was carried out by using samples from a full-scale pistachio
production industry located in Gaziantep, Turkey. The generated pistachio wastewater amount is about 10
m3/d. Samples were collected in June 2020 and kept in the refrigerator at +/- 4 °C. Wastewater was not
diluted and was used as received in the further experiments. The characteristics of the studied pistachio
wastewater showed that water quality parameters were higher than the acceptable range restricted by the
standard pollution regulations as summarized in Table 1.

Table 1. Pistachio wastewater characteristics

Sample date pH Conductivity COD Total Phenol (mg/L)

(mS/cm) (mg/L)

20.06.2020 5.15±0.15 8.32±0.22 20700±220 1546±45

2.2. Experimental setup

Electrooxidation (EO) experiments of pistachio wastewater were accomplished in a 500 mL borosilicate

glass reactor with a working volume of 250 mL. A magnetic stirrer (Wisd -Wisestir msh-20A) and a Teflon-
covered magnetic stirring bar were used to mix the wastewater at 300 rpm. The reactor was engaged in

Page 4/23
the temperature-controlled water bath to fund a constant reaction temperature (298 ± 1K). The
experimental arrangement is shown in Fig. 1.

Activated carbon cloth (ACC) was utilized as anode/cathode electrode pairs. The electrodes were supplied
by Norm Technologies, Turkey. Anode/cathode electrode pairs dimensions were arranged as 05 cm
width × 08 cm high × 01 mm thickness with 40 cm2 of total effective area and 02 cm of distance between
the electrodes. In order to set anode and cathode connected to the positive and negative outlets, DC power
source (AATech ADC-3303D, maximum current of 30 A) was used.

SWO experiments were performed using an experimental setup which is previously given in
detail (Yabalak 2018b). Briefly, a homemade stainless steel cylindrical container was used as a reactor,
the reactor was heated and stirred using a heater with a integrated magnetic stirrer (Heidolph-MR.3001)
and the temperature was controlled by a digital thermometer (Elimko, E-2000). 50 mL of EO-pre-treated
pistachio wastewater was placed in the reactor and a certain amount of H2O2, which was determined
according to pre-treatments of SWO, was added. The reactor was closed and its initial inner pressure was
adjusted to 30 bar using nitrogen gas in each case. The reactor was heated to a specific degree and held
constant during the experiment time. The reactor was cooled to room temperature, depressurised and
opened at the end of the treatment time. In each experiment, an amount of 20 mL of SWO-treated sample
was collected and stored at 04 °C for total phenol and COD analyses. The levels of each experimental
variable above-mentioned were constructed by the central composite design (CCD) model given in Table
2.

Table 2. The CCD of the independent variables of the SWO method

Factors Independent variables Coded levels

-1.682 -1 0 1 +1.682

x1 Temperature (K) 376 383 393 403 410

x2 Treatment time (min) 26.4 40 60 80 93.6

x3 Concentration of H2O2 (M) 0.08 0.25 0.5 0.75 0.92

2.3. Analysis

Three studied current densities (50, 100, and 150 A/m2), four altered electrooxidation times (30, 60, 120,
and 180 min), and three different wastewater pH (5, 6, and 8) were investigated for pistachio wastewater
EO treatment. For suitable time intervals, taken samples were centrifuged at 6000 rpm for 5 min and
analysed by measuring COD, total phenols, pH, and conductivity. All electrooxidation experiments were
accomplished in duplicates.

Page 5/23
Removal or elimination efficiencies were calculated using Eq (1):

Ci : The initial concentration and Cf is the concentration after definite reaction time t (min).

3. Results And Discussion

In this study, EO and SWO hybrid process was used for pistachio wastewater treatment. ACC electrode
pairs were used as the anode/cathode electrodes. Current density effects, solution pH, and electrolysis
time were investigated on both COD and total phenols elimination efficiencies. Moreover,
electrochemically pre-treated wastewater was subjected to subcritical water oxidation. The effects of
temperature (376–410 K), treatment time (26.4-93.65 min), and concentration of H2O2 (0.08-0.92 M) were
examined removal of COD and total phenols for the SWO process as explained in the following
subsections.

3.1. Current density effect on removal efficiencies in the EO method

Current density has an important impact on the success of the EO process and hence on COD and total
phenol removals since it is the driving force in the migration of charge. As illustrated in Fig. 2A, COD
removal efficiency increases with the increase of current density from 50 to 150 A/m2. Current density
effect was better when increased up to 150 A/m2, where maximum COD elimination performance was
nearly 50 % after 03 h electrolysis. COD removal efficiency versus time was improving as experiment time
progresses. Previous studies by K. V. Radha ( 2009) and Toth et al. (2018), as well revealed that increases
in current density directed to the increase in COD elimination. The potentials required for the oxidation of
organic material are usually high. This implies that water could be oxidized and the production of oxygen
is the main side reaction, which is a non-desired reaction and it may influence dramatically on the COD
removal efficiency in particular for low current densities. The application of enough high current density,
promotes the formation of stable oxidants, at the same time with the oxidation of other species contained
in the treated pistachio wastewater. This had an additional benefit as these oxidants helped pollutants in
wastewater oxidization. Similar results were known for total phenol removal with lower efficiency as
shown in Fig. 2B. The total phenol concentration was reduced more with the increase of current density.
Within the first 1h of reaction, there was a minor difference in total phenol removal efficiency when
applying 100 and 150 A/m2 of current density. The maximum recorded phenol removal efficiency as
much as 19.02 % was obtained for 150 A/m2 of current density after the total investigated reaction time
of 180 min. In fact, the mechanism of phenol reaction is complicated and not yet entirely understood. It is
agreed that dual oxidation pathways are elaborated in EO process: direct pathway degradation and
indirect pathway (Cañizares et al. 1999). In the two paths, oxidation is achieved via hydroxyl radicals
(OH·), which are formed on the electrode surface by release of water and/or by hydroxyl ions direct
oxidation.
Page 6/23
As soon as hydroxyl radicals are generated, two restrictive behaviours, subject to the electrode nature, can
be notable: chemisorption of hydroxyl radicals, which is established in active electrodes, and
physisorption of hydroxyl radicals, that is established in nonactive electrodes as activated carbon cloth
that was used in this study. Therefore, phenol is changed directly into water and carbon dioxide by
physisorbed OH· radicals. Then, phenoxy radicals are produced, by the electrophilic attack of adsorbed
OH· radicals over phenol. The formed radicals can be coupled to produce diverse molecular weights
polymers.

As EO is processing, solution pH has known a minor increase simultaneously with current density
increase (Fig. 2C) up to maximum pH 6 for 150 A/m2 after 180 min of reaction. On the contrary,
conductivity has known a tinny decrease as a function of current density or as experience processing time
as seen in Fig. 2D.

3.2. pH effect on removal efficiencies in the EO method

pH is a significant operating parameter affecting the efficiency of the EO process. In the EO process,
several free radicals are formed of different ratios dependent on the pH. These products show an
important role in the removal of COD and total phenol in the EO process (Can 2014).

COD and total phenols elimination performances as a function of electrooxidation experiment treatment
time at various pH values while holding constant current density at 150 A/m2 is shown in Fig. 3A and Fig.
3B subsequently. The best mineralization rate by electrooxidation happened at the original pH 5 then
decreased at a higher solution pH of 6 and pH 8. Since the anode is perfectly stable in the acidic solution,
no attempt has been performed to explore COD removal and phenol degradation in a higher alkaline
environment. Similar results were obtained in previous studies, where the EO process well performed for
COD and phenol removals in an acidic medium (Periyasamy and Muthuchamy 2018, Rahmani et al.
2018). The radicals formation from the supporting electrolyte in the acidic medium (pH 5) has possibly
enhanced the elimination rate of organic substances (Periyasamy and Muthuchamy 2018, Rahmani et al.
2018). The acidic pH 5 perhaps inhibits the reaction of oxygen progress, resulting to the improvement of
COD and total phenol degradation efficiency (Babuponnusami and Muthukumar 2012). This can be
explained as at high pH, hydroxyl radical is transformed to O·-, which its oxidation characteristic declines.
On the other hand, at low pH, hydroxyl radicals do react with hydrogen ions performing as a scavenger of
hydroxyl radicals (Godini et al. 2013, Rahmani et al. 2015b). Fig.3B exhibits that the high phenol removal
efficiency is obtained at pH=8 rather than pH=5. It can be explained as the removal of phenolic
compounds on the anode surface based on electropolymerization in an alkaline solution (Belhadj Tahar
and Savall 2009). During electrochemical oxidation of phenol in alkaline aqueous solution, polymer films
form on ACC anodes. Hydroxyl radicals oxidize the polymer film formed on ACC electrodes. As seen in
Fig. 3C, the electrooxidation reaction did not affect the final solution pH much during the whole reaction
duration of 180 min. However, conductivity decreased at acidic pH (Fig. 3D).

3.3. Evaluation of SWO method

Page 7/23
The predicted and experimental COD and total phenol removal of 20 runs of the CCD models were given
in Table 3. x1, x2 and x3, demonstrates temperature (K), treatment time (min) and concentration of H2O2
(M), respectively. The highest experimental and predicted COD removal yields were obtained in the run 11
and 20, respectively, as 54.5% and 54.3%. The highest experimental and predicted total phenol removal
yields were obtained in the run 7 and 20, respectively, as 76.5% and 76.87%. Although run 7 provided
significant total phenol removal, COD removal remained at 33.1%. Considering the experimental
conditions of run 7 and run 11, it can be inferred that the temperature is a more efficacious parameter in
the removal of total phenol than the removal of COD. Furthermore, considerable removal rates of total
phenol compared to COD removal rates were due to the mechanism required for each response. Namely,
to obtain high COD removal, the target pollutant must be degraded to the final oxidation product, H2O and
CO2, where a minor change in the structure of the target molecule provides higher total phenol removal
rates (Yabalak 2021). Therefore, more challenging experimental conditions are required to obtain higher
COD removal rates compared to total phenol removal. In this case, the experimental variables can be
adjusted according to the desired/regulation removal values of each response. Residual (Res.) values
indicating a precision of a model and agreement between actual and predicted values were obtained to be
between -2.1– 1.93 in the COD removal model and -5.76 – 3.34 in the total phenol removal
model (Yabalak 2021). Also, leverages (L.) value close to 1 is not desired as L. values demonstrates the
potential of the design points affecting the fit of the model coefficients, based on its location in the design
space (Yabalak 2021). Herein, although quite reasonable and compatible experimental and predicted
values were obtained in both models, the COD removal model showed better performance in the
derivation of predicted results.

Table 3. Experimental variables, COD and total phenol removal values, residuals and leverages values of
experimental and predicted results.

Page 8/23
 Run Experimental COD removal (%) Total phenolic removal (%)
variables

x1 (K) x2 x3 Exp. Pre. Res. L. Exp. Pre. Res. L.

(min) (M)

1 403 40 0.25 18.70 18.06 0.64 0.670 26.60 23.74 2.86 0.670

2 393 60 0.5 29.90 29.92 -0.024 0.166 52.10 57.86 -5.76 0.166

3 376 60 0.5 28.60 29.60 -1.00 0.607 49.40 50.34 -0.94 0.607

4 393 60 0.5 29.10 29.92 -0.82 0.166 59.60 57.86 1.74 0.166

5 383 80 0.75 33.50 34.31 -0.81 0.670 37.50 40.10 -2.60 0.670

6 393 60 0.5 29.50 29.92 -0.42 0.166 55.50 57.86 -2.36 0.166

7 403 80 0.75 33.10 33.51 -0.41 0.670 76.50 75.07 1.43 0.670

8 383 40 0.75 38.10 37.91 0.19 0.670 42.20 38.96 3.24 0.670

9 393 60 0.5 31.60 29.92 1.68 0.166 58.30 57.86 0.44 0.166

10 393 60 0.08 27.80 29.34 -1.54 0.607 34.80 35.29 -0.49 0.607

11 393 60 0.92 54.50 52.72 1.78 0.607 69.10 68.98 0.12 0.607

12 383 80 0.25 44.70 42.77 1.93 0.670 55.50 53.17 2.33 0.670

13 393 26.4 0.5 27.50 26.97 0.53 0.607 32.10 33.67 -1.57 0.607

14 393 60 0.5 30.50 29.92 0.58 0.166 61.20 57.86 3.34 0.166

15 410 60 0.5 29.90 28.66 1.24 0.607 62.80 62.22 0.58 0.607

16 383 40 0.25 18.60 18.36 0.24 0.670 43.40 44.58 -1.18 0.670

17 393 60 0.5 28.90 29.92 -1.02 0.166 60.50 57.86 2.64 0.166

18 403 80 0.25 24.90 25.26 -0.36 0.670 26.40 29.39 -2.99 0.670

19 393 93.6 0.5 29.70 29.99 -0.29 0.607 40.60 39.39 1.21 0.607

20 403 40 0.75 52.20 54.30 -2.10 0.670 74.80 76.87 -2.07 0.670

The theoretical regression equations in terms of coded factors of COD and total phenol removal models
were given in Eq. 1 and 2, respectively.

Page 9/23
Eq. 2 and 3 are practical equations to predict the removal rates in the given experimental conditions. Also,
the coefficient of each term in the equations demonstrates its influence portion on the response. Y1 and
Y2 indicate the COD and total phenol removal percent in Eq. 2 and 3, respectively. The independent
variables were depicted by x1, x2, and x3, while square effects were symbolized by and x12, x22, and x32,
interaction effects were symbolized by x1x2, x1x3, and x2x3 in both equations. Therefore, the most
influential term of COD removal was x2x3 (-7.0) followed by x3 (6.95), where the most influential term of
total phenol removal was x2x3 (-7.0) x1x3 (14.69) followed by x3 (10.02).

Table 4. ANOVA of the CCD models of COD and total phenol removals

df: Degree of freedom

The accuracy and suitability of the applied CCD models as well as the interactions between the
experimental variables were assessed using the ANOVA test, the results of which were given in Table 4.
Several statistical terms and tests such as F test and lack-of-fit tests and coefficients of determination
(R2, R2adj) were evaluated to determine the adequacy of both models. The higher the F value, the more
noteworthy influence the relevant term or model has on the response [35]. On the contrary, the p-value is
desired to be as low as possible (<0.05). Although p-values of both models were <0.0001, F values were
76.33 and 41.1 in COD and total phenol removal model, respectively. Therefore, the COD removal model is
more compatible to designate the experimental variables compared to the total phenol removal model.
Besides, in the COD removal model, x3, x1x2, x1x3, x2x3 and x32, and in the total phenol removal model,

Page 10/23
x1, x3, x1x3, x2x3, x22 and x32 were the significant model terms. Also, an essential term that measures how
well the model is lack of fit value. Lack of fit value is desired to be >0.05 (Yabalak et al. 2021). In this
case, both performed model fit the data well as p values of COD and total phenol removal model were
obtained as 0.0937 and 0.5535, respectively.

Table 5. Regression and correlation coefficients of the CCD models of COD and total phenol removals

Regression and correlation coefficients COD removal Total phenol removal

Standard deviation 1.52 3.36

Mean 32.06 50.95

C.V. % 4.74 6.60

PRESS 150.47 557.52

R2 0.9857 0.9737

Adjusted R2 0.9727 0.9500

Predicted R2 0.9067 0.8702

Adequate precision 33.695 22.348

The regression and correlation coefficients of the applied CCD method were demonstrated in Table 5. The
fit of a model to each point in the design can be evaluated by the predicted residual error sum of squares
(PRESS) value(Yabalak 2021). Herein, a smaller PRESS value observed for the COD removal model
(150.12) than the total phenol removal model (557.52) indicates a better fit of the COD removal model.
Also, R2 values were 0.9857 and 0.9737 for COD and total phenol removal models, respectively, where
predicted R2 values were 0.9067 and 0.8702 for COD and total phenol removal models, respectively. A
higher predicted R2 value observed for the COD removal model proved its better efficiency in the derivation
of predicted results. Also, the closeness of the adjusted and predicted R2 values (the maximum
acceptable difference is 0.2) indicate a better agreement between predicted and actual values. Therefore,
an adequate correlation between adjusted and predicted R2 values was observed in both models.

Figure 4 depicted the agreement between the actual and predicted results of both models. Also, any point
that cannot be predicted well by the applied model can be easily determined using this graph. In the
desired model with a fit between actual and predicted results, data points split by 45-degree line(Yabalak
et al. 2021). The well-aligned points observed along the above-mentioned line in Figure 4 indicate the
accordance between actual and predicted results in both models. However, the more closeness of the
points in the COD removal model proved a better fit of the model. It is supported by the COD and total
phenol removal standard deviation values model obtained as 1.52 and 3.36, respectively.

Page 11/23
Residuals of actual results versus experimental run order in both CCD model were demonstrated in Figure
5. This plot is practical in the detection of the hidden variables that have the potential to affect the
response. Random distribution was observed and the majority of the points remains in the upper and
lower control limits (-4 – 4) in both methods. However, run 12, and 20 in the COD elimination model and
run 2 in the total phenol removal model were pretty near to the above-mentioned limits.

Experimental variables binary effects on the COD removal yields were demonstrated in Figure 6 (a), (c)
and (e), where Figure 6 (b), (d) and (f) displays the binary effects of the experimental variables on the
total phenol removal yields. According to Figure 6 (a), high COD removal yields can be attained at high
temperature and low treatment time using 0.75 M of H2O2. However, the red area indicating the high-
yielded area is much wider even at 0.65 M of H2O2 in Figure 6 (b). The difference between the red area in
Figure 6 (a) and (b) proved that even high removal rates of total phenol removal can be obtained under
mild conditions, more challenging conditions are required to obtained high COD removal rates. For
instance, at constant 0.75 M of H2O2, 373 K and in 60 min of treatment time, 40.80% and 65.85% of COD
and total phenol removal, respectively, can be increased to 44.42% and 83.51%, respectively, by increasing
the only temperature to 403 K. Although the increase in temperature provides 3.62% increase in COD
removal rates, it provides 17.66% increase in the total phenol removal rates. Therefore, it can be said that
complete mineralisation (COD removal) is less affected by temperature increase compared to phenol
removal. Because, although the temperature increase easily causes deterioration in the structure of
phenolic compounds, the oxidized species in the smaller chain structure still cause the COD value to be
high.

Figure 6 (c) and (d) demonstrates the combined effects of temperature and H2O2 concentration on the
COD and total phenol removal yields, respectively, at a fixed treatment time of 45 minutes. Figure 6 (c)
demonstrates that the red area demonstrating high COD removal yields is much less than the red area
shown in Figure 6 (d), which demonstrates high total phenol removal yields. In constant treatment time of
45 min, the obtained COD removal (20.69%) and total phenol removal (38.90%) rates at 393 K and using
0.25 M of H2O2 can be altered to 28.96% of COD removal and 52.34% of total phenol removal by doubling
the H2O2 concentration. Further, tripling the H2O2 concentration provides 45.09% of COD removal and
61.73% of total phenol removal. These finding proved that H2O2 concentration has a significant effect on
the responses and it must be at a sufficient level for efficient hydroxyl and other radicals’ formation based
on the below-given reactions (Eq. 4-6) (Yabalak 2018a).

H2O2 2•OH (4)

H2O2 + •OH → HO2• + H2O (5)

H2O2 + •OH → O2•- + H+ + H2O (6)

However, hydroxyl and other radical species are very unstable and can be easily affected by the
experimental conditions, especially the hydrogen peroxide concentration, and can quench the radicals
Page 12/23
formed by reacting as seen in Eq. 7-8, thus causing a decrease in the removal yields (Yabalak 2018a).
Therefore, it is of great importance to determine the concentration of hydrogen peroxide precisely.

•OH + •OH → H2O2 (7)

•OH + HO2• → H2O + O2 (8)

The treatment time and H2O2 concentration combined effects on the COD and total phenol removal yields
at a fixed temperature of 400 K were shown in Figure 6 (e) and (f), correspondingly. At a fixed temperature
of 400 K and 0.25 M of H2O2 concentration, COD and total phenol removal yields can be altered from
18.25% and 27.15%, respectively, to 23.65%, and 37.73%, respectively, by increasing the treatment time
from 40 min to 60 min. Further increasing the treatment time to 80 min, provides 28.03% and 33.23% of
COD and total phenol removal yields, respectively. Therefore, although treatment time affects the removal
yields, it is not as effective as hydrogen peroxide concentration. Also, further increasing the treatment
time after a specific level, may not increase the removal yields.

In Figure 7, the cube graph designed based on the predicted responses obtained from CCD models is
given. By setting the experimental variable between the lowest (-1) and the highest (+1) levels, the
optimum conditions for each variable for a given response level can be easily determined (Yabalak 2021).
COD and total phenol removal efficiencies at the lowest values of all variables remain as 18.36% and
44.58%, respectively. However, they can be increased to 42.17% and 53.17%, respectively, by increasing
only treatment time to its highest limit. Also, COD and total phenol removal performances at the highest
values of all variables can be obtained as 33.51% and 75.07%, respectively. Nevertheless, higher removal
rates in both COD and total phenol can be obtained even in the lowermost level of the treatment time, but
the highest level of the other two variables.

4. Conclusion
This study has demonstrated the importance of treating pistachio wastewater using a hybrid system
between electrocoagulation as pre-treatment stage followed by subcritical water oxidation for both COD
and total phenol removal. Current density and initial solution pH were optimised in the first stage of
treatment. In the next stage, temperature, treatment time and concentration of H2O2 were the main
parameters to study the efficiency of the selected treatment process.

The interactions between the experimental variables, the accuracy and suitability of the applied CCD
models were evaluated using the ANOVA test, the results of the F test and lack-of-fit tests, coefficients of
determination (R2, R2adj) and supplementary statistical terms were evaluated to determine the adequacy
of both models. It was found that the COD removal model is more compatible to designate the
experimental variables compared to the total phenol removal model. Although p-values of both models
were <0.0001, F values were 76.33 and 41.1 in COD and total phenol removal model, respectively.

Page 13/23
The regression and correlation coefficients of the applied CCD method were investigated. The fit of a
model to each point in the design was evaluated by the predicted residual error sum of squares (PRESS)
value. Herein, a smaller PRESS value observed for the COD removal model (150.12) than the total phenol
removal model (557.52) indicated a better fit of the COD removal model. Also, R2 values were 0.9857 and
0.9737 for COD and total phenol removal models, respectively, where predicted R2 values were 0.9067 and
0.8702 for COD and total phenol removal models, consecutively. A higher predicted R2 value observed for
the COD removal model proved its better efficiency in the derivation of predicted results.

The CCD was used to study the impact of the different studied parameters and predicted results were
compared to the experimental ones. It was found that the model fits more with COD removal compared to
total phenol removal.

Declarations
Ethical Approval: This article does not contain any studies with human participants or animals performed
by any of the authors.

Consent to Participate: All authors certify that they have no affiliations with or involvement in any
organization or entity with any financial interest or non-financial interest in the subject matter or materials
discussed in this manuscript.

Consent to Publish: All authors agreed with the content and that all gave explicit consent to submit and
that they obtained consent from the responsible authorities at the institute/organization where the work
has been carried out, before the work is submitted.

Authors Contributions: All authors contributed to the study conception and design. Material preparation,
data collection and analysis were performed by Zelal Isik, Zhalaladdin Jabbiyev and Yasin Ozay. The first
draft of the manuscript was written by Raouf Bouchareb, Erdal Yabalak, and Nadir Dizge and all authors
commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Funding: No funding was received to assist with the preparation of this manuscript.

Competing Interests: The authors have no competing interests to declare that are relevant to the content
of this article.

Availability of data and materials: All authors are requested to make sure that all data and materials as
well as software application or custom code support their published claims and comply with field
standards

References
1. Al-Doury MMI (2019) Treatment of oily sludge using solvent extraction. Pet Sci Technol 37:190–196.
https://doi.org/10.1080/10916466.2018.1533859
Page 14/23
 2. Babuponnusami A, Muthukumar K (2012) Advanced oxidation of phenol: A comparison between
Fenton, electro-Fenton, sono-electro-Fenton and photo-electro-Fenton processes. Chem Eng J 183:1–
9. https://doi.org/10.1016/j.cej.2011.12.010
3. Bayar S, Boncukcuoǧlu R, Yilmaz AE, Fil BA (2014) Pre-Treatment of Pistachio Processing Industry
Wastewaters (PPIW) by Electrocoagulation using Al Plate Electrode. Sep Sci Technol 49:1008–1018.
https://doi.org/10.1080/01496395.2013.878847
4. Bayar S, Massara TM, Boncukcuoglu R, et al (2018) Advanced treatment of industrial wastewater
from pistachio processing by Fenton process. Desalin Water Treat 112:106–111.
https://doi.org/10.5004/dwt.2018.22197
5. Belhadj Tahar N, Savall A (2009) Electrochemical removal of phenol in alkaline solution. Contribution
of the anodic polymerization on different electrode materials. Electrochim Acta 54:4809–4816.
https://doi.org/10.1016/j.electacta.2009.03.086
6. Bouchareb EM, Kerroum D, Bezirhan Arikan E, et al (2021) Production of bio-hydrogen from bulgur
processing industry wastewater. Energy Sources, Part A Recover Util Environ Eff 1–14.
https://doi.org/10.1080/15567036.2021.1877853
7. Bouchareb R, Derbal K, Özay Y, et al (2020) Combined natural/chemical coagulation and membrane
filtration for wood processing wastewater treatment. J Water Process Eng 37:101521.
https://doi.org/10.1016/j.jwpe.2020.101521
8. Can OT (2014) COD removal from fruit-juice production wastewater by electrooxidation
electrocoagulation and electro-Fenton processes. Desalin Water Treat 52:65–73.
https://doi.org/10.1080/19443994.2013.781545
9. Cañizares P, Domínguez JA, Rodrigo MA, et al (1999) Effect of the current intensity in the
electrochemical oxidation of aqueous phenol wastes at an activated carbon and steel anode. Ind Eng
Chem Res 38:3779–3785. https://doi.org/10.1021/ie9901574
10. Chen G (2004) Electrochemical technologies in wastewater treatment. Sep Purif Technol 38:11–41.
https://doi.org/10.1016/j.seppur.2003.10.006
11. Crini G, Lichtfouse E (2019) Advantages and disadvantages of techniques used for wastewater
treatment. Environ Chem Lett 17:145–155. https://doi.org/10.1007/s10311-018-0785-9
12. de Caprariis B, De Filippis P, Hernandez AD, et al (2017) Pyrolysis wastewater treatment by adsorption
on biochars produced by poplar biomass. J Environ Manage 197:231–238.
https://doi.org/10.1016/j.jenvman.2017.04.007
13. Elazzouzi M, Haboubi K, Elyoubi MS (2019) Enhancement of electrocoagulation-flotation process for
urban wastewater treatment using Al and Fe electrodes: Techno-economic study. Mater Today Proc
13:549–555. https://doi.org/10.1016/j.matpr.2019.04.012
14. Fil BA, Boncukcuoglu R, Yilmaz AE, Bayar S (2014) Electro-oxidation of pistachio processing industry
wastewater using graphite anode. Clean - Soil, Air, Water 42:1232–1238.
https://doi.org/10.1002/clen.201300560

Page 15/23
15. Frangos P, Shen W, Wang H, et al (2016) Improvement of the degradation of pesticide deethylatrazine
by combining UV photolysis with electrochemical generation of hydrogen peroxide. Chem Eng J
291:215–224. https://doi.org/10.1016/j.cej.2016.01.089
16. Godini K, Azarian G, Rahmani AR, Zolghadrnasab H (2013) Treatment of waste sludge: A comparison
between anodic oxidation and electro-fenton processes. J Res Health Sci 13:188–193.
https://doi.org/10.34172/jrhs13876
17. Gür E, Demirer GN (2019) Anaerobic Digestability and Biogas Production Capacity of Pistachio
Processing Wastewater in UASB Reactors. J Environ Eng 145:04019042.
https://doi.org/10.1061/(asce)ee.1943-7870.0001559
18. Isik Z, Arikan EB, Ozay Y, et al (2020) Electrocoagulation and electrooxidation pre-treatment effect on
fungal treatment of pistachio processing wastewater. Chemosphere 244:125383.
https://doi.org/10.1016/j.chemosphere.2019.125383
19. Isik Z, Bouchareb R, Saleh M, Dizge N (2021) Investigation of sesame processing wastewater
treatment with combined electrochemical and membrane processes. Water Sci Technol 84:2652–
2660. https://doi.org/10.2166/wst.2021.152
20. Linares-Hernández I, Barrera-Díaz C, Bilyeu B, et al (2010) A combined electrocoagulation-
electrooxidation treatment for industrial wastewater. J Hazard Mater 175:688–694.
https://doi.org/10.1016/j.jhazmat.2009.10.064
21. Martínez-Huitle CA, Ferro S (2006) Electrochemical oxidation of organic pollutants for the wastewater
treatment: Direct and indirect processes. Chem Soc Rev 35:1324–1340.
https://doi.org/10.1039/b517632h
22. Ozay Y, Ünşar EK, Işık Z, et al (2018) Optimization of electrocoagulation process and combination of
anaerobic digestion for the treatment of pistachio processing wastewater. J Clean Prod 196:42–50.
https://doi.org/10.1016/j.jclepro.2018.05.242
23. Periyasamy S, Muthuchamy M (2018) Electrochemical oxidation of paracetamol in water by graphite
anode: Effect of pH, electrolyte concentration and current density. J Environ Chem Eng 6:7358–7367.
https://doi.org/10.1016/j.jece.2018.08.036
24. Radha K V., Sridevi V, Kalaivani K (2009) Electrochemical oxidation for the treatment of textile
industry wastewater. Bioresour Technol 100:987–990.
https://doi.org/10.1016/j.biortech.2008.06.048
25. Rahmani AR, Godini K, Nematollahi D, Azarian G (2015a) Electrochemical oxidation of activated
sludge by using direct and indirect anodic oxidation. Desalin Water Treat 56:2234–2245.
https://doi.org/10.1080/19443994.2014.958761
26. Rahmani AR, Nematollahi D, Azarian G, et al (2015b) Activated sludge treatment by electro-Fenton
process: Parameter optimization and degradation mechanism. Korean J Chem Eng 32:1570–1577.
https://doi.org/10.1007/s11814-014-0362-2
27. Rahmani AR, Nematollahi D, Samarghandi MR, et al (2018) A combined advanced oxidation process:
Electrooxidation-ozonation for antibiotic ciprofloxacin removal from aqueous solution. J Electroanal

Page 16/23
 Chem 808:82–89. https://doi.org/10.1016/j.jelechem.2017.11.067
28. Shan J, Zheng Y, Shi B, et al (2019) Regulating electrocatalysts via surface and interface engineering
for acidic water electrooxidation. ACS Energy Lett 4:2719–2730.
https://doi.org/10.1021/acsenergylett.9b01758
29. Toth AJ, Haaz E, Szilagyi B, et al (2018) COD reduction of process wastewater with vacuum
evaporation. Waste Treat Recover 3:1–7. https://doi.org/10.1515/wtr-2018-0001
30. Wang M, Payne KA, Tong S, Ergas SJ (2018) Hybrid algal photosynthesis and ion exchange (HAPIX)
process for high ammonium strength wastewater treatment. Water Res 142:65–74.
https://doi.org/10.1016/j.watres.2018.05.043
31. Xiao W, Ke S, Quan N, et al (2018) The Role of Nanobubbles in the Precipitation and Recovery of
Organic-Phosphine-Containing Beneficiation Wastewater. Langmuir 34:6217–6224.
https://doi.org/10.1021/acs.langmuir.8b01123
32. Yabalak E (2018a) An approach to apply eco-friendly subcritical water oxidation method in the
mineralization of the antibiotic ampicillin. J Environ Chem Eng 6:7132–7137.
https://doi.org/10.1016/j.jece.2018.10.010
33. Yabalak E (2018b) Degradation of ticarcillin by subcritical water oxidation method: Application of
response surface methodology and artificial neural network modeling. J Environ Sci Heal - Part A
Toxic/Hazardous Subst Environ Eng 53:975–985. https://doi.org/10.1080/10934529.2018.1471023
34. Yabalak E (2021) Treatment of agrochemical wastewater by thermally activated persulfate oxidation
method: Evaluation of energy and reagent consumption. J Environ Chem Eng 9:105201.
https://doi.org/10.1016/j.jece.2021.105201
35. Yabalak E, Görmez Ö, Gizir AM (2018) Subcritical water oxidation of propham by H2O2 using
response surface methodology (RSM). J Environ Sci Heal - Part B Pestic Food Contam Agric Wastes
53:334–339. https://doi.org/10.1080/03601234.2018.1431468
36. Yabalak E, Ozay Y, Vatanpour V, Dizge N (2021) Membrane concentrate management for textile
wastewater with thermally activated persulfate oxidation method. Water Environ J 35:1281–1292.
https://doi.org/10.1111/wej.12718

Figures

Page 17/23
Figure 1

Electrooxidation experimental set-up (1. Magnetic stirrer, 2. Magnetic bar, 3. Storage tank for wastewater,
4. Electrodes, 5. Water Bath, 6. DC power source).

Page 18/23
Figure 2

Current density effect on (A) COD removal performance, (B) total phenol removal performance, (C) change
of solution pH, (D) change of conductivity versus time during the reaction (Experimental conditions:
Solution pH=5.15, volume=250 mL, stirrer speed=500 rpm).

Page 19/23
Figure 3

Solution pH effect on (A) COD removal efficiency, (B) total phenol removal efficiency, (C) change of
solution pH, (D) change of conductivity versus time during the reaction (Experimental conditions: Current
density=150 A/m2, volume=250 mL, stirrer speed=500 rpm).

Page 20/23
Figure 4

The actual and predicted values of COD and total phenol removal models correlation.

Figure 5

Page 21/23
Residuals versus the experimental run in COD and total phenol removal models.

Figure 6

Binary effects of experimental variables on the COD removal (a, c and e) and total phenol removal (b, d
and f) yields.

Page 22/23
Figure 7

Cube plot of COD removal and total phenol removal models

Page 23/23