* Corresponding author.
1944-3994/1944-3986 © 2017 Desalination Publications. All rights reserved.
Desalination and Water Treatment
80 (2017) 89–99
Optimization of metal component, characterization, and stability
of Cu/Mg/Al–chitosan catalyst in catalytic ozonation of a landll leachate
Dariush Ranjbar Vakilabadia, Bahman Ramavandib,*, Amir Hessam Hassanic,
aDepartment of Environmental Science, Faculty of Environment and Energy, Science and Research Branch, Islamic Azad University,
Poonak sq. Ashrafi Esfahani Blvd., Hesarak, Tehran 14515-775, Iran, email: firstname.lastname@example.org
bDepartment of Environmental Health Engineering, Faculty of Health, Bushehr University of Medical Sciences, Mobaraki Street,
7518759577 Bushehr, Iran, Tel. +989363311903; Fax: +987733450134; emails: email@example.com, firstname.lastname@example.org
cDepartment of Environmental Engineering, Faculty of Environment and Energy, Science and Research Branch,
Islamic Azad University, Poonak sq. Ashrafi Esfahani Blvd., Hesarak, Tehran 14515-775, Iran,
emails: email@example.com (A.H. Hassani), firstname.lastname@example.org (G. Omrani)
Received 14 November 2016; Accepted 14 May 2017
The aim of this work was to optimize the metal components of the Cu/Mg/Al–chitosan catalyst for
the treatment of a landfill leachate during a catalytic ozonation process. A central composite design
with response surface methodology was applied to assess the relationships between the value of Mg,
Cu, and Al in the Cu/Mg/Al–chitosan catalyst and chemical oxygen demand (COD) removal from the
landfill leachate and identify the optimum conditions. Quadratic model for three variables proved to
be significant with very low probabilities (<0.0001). The optimum metal content for synthesis of the
Cu/Mg/Al–chitosan was determined as Mg = 4 mmol/L, Cu = 0.89 mmol/L, and Al = 2 mmol/L. A con-
firmation run gave 81.35% of COD removal compared with 79.89% of predicted value. Results showed
that the magnesium metal in the catalyst was more effective in COD removal than other metals. After a
50-min reaction time, the COD removal percentage of 49, 61, and 78 was attained for the landfill leach-
ate with initial pH of 5.5, 7.3, and 9.1, respectively. The recyclability test indicated that the optimized
catalyst could be efficiently utilized three times with COD removal efficiency of 81.35%, 66%, and
50%. The X-ray powder diffraction and electron dispersive spectroscopy tests confirmed the successful
modification of chitosan with Mg/Cu/Al. The findings of this study demonstrate the applicability of
Cu/Mg/Al–chitosan for eliminating COD from the leachate of sanitary landfill.
Keywords: Landfill leachate; Response surface methodology; Cu/Mg/Al–chitosan; Optimization;
Chemical oxygen demand
Among landfill, composting, recycling in agriculture,
dumping into the sea, and incineration for municipal solid
waste (MSW), the landfill method is the predominant one
in most countries because of its low immediate costs .
Percolation of rainfall through landfill cells in combination
with the biological, chemical, and physical decomposition
of MSW lead to the generation of a highly contaminated
liquid, namely ‘leachate’ . Owing to the mixture of
MSW deposited in landfills, leachate contains a variety of
harmful pollutants such as high levels of chemical oxygen
demand (COD), nitrogenous substances, inorganic salts,
xenobiotics, heavy metals, and other toxicants . Leaching
the untreated leachate to the environment and water bod-
ies is very dangerous for humans and animals. Therefore,
many researchers have tried to explore potentially efficient
techniques to treat landfill leachate before it enters the
D.R. Vakilabadi et al. / Desalination and Water Treatment 80 (2017) 89–9990
surrounding environment. For instance, chemical coagu-
lation–nanofiltration , combined sequence batch reactor
reaction , Fenton reaction , membrane filtration [7,8],
electrochemical oxidaton , single ozonation and preo-
zonation [10,11], coagulation/flocculation and ozonation
, adsorption , electrocoagulation processes , and
photocatalytic , techniques have been studied to treat
sanitary landfill leachate. These techniques have different
advantages and limitations. For example, chemical coagula-
tion, adsorption, and Fenton reaction techniques might have
sludge production problems . Membrane technique is
expensive. The single-ozonation technology can remove the
colour of landfill leachate effectively , but it might be
restricted for COD removal. Therefore, more efficient and
economical techniques are required for treating landfill
leachate. A combination of ozone and an appropriate cat-
alyst could overcome single ozonation limitations. In this
context, many catalysts such as granular activated carbon
, metallic ions , and TiO2 nanotube  have been
used by researchers. Among the catalysts used for remedi-
ation, the metal-based one has shown great potential and
high activity for oxidizing organic pollutants from aqueous
solutions . Therefore, the investigation to find a novel
metal-based catalyst is still continuing.
Bimetallic catalysts, in particular, are considered as the
most promising catalysts due to the synergistic effect .
Nevertheless, the transition metal oxides and chalcogenides
still suffer from some pulverization problems such as severe
particle aggregation, surface passivation due to metal oxide
formation, and poor stabilities [22,23]. To overcome these lim-
itations, the use of three metallic catalysts like ZnAlFe ,
ZnAlLa , MgZnAl , ZnAlTi , and MgZnIn  is
one of the promising and suggested strategies in the literature.
In this research, a three-metal catalyst, namely Cu/Mg/Al
was coated on the surface of a chitosan to make Cu/Mg/
Al–chitosan and used for COD removal from a sanitary
landfill leachate. Furthermore, chitosan was employed as a
supporting agent for the catalyst. Chitosan has three func-
tional groups (two hydroxyl groups and one amino group)
per glucosamine unit, which easily bind to Cu/Mg/Al . A
major obstacle of using metal or metal containing catalysts
is the optimization of their components to achieve the best
efficiency and avoid an excess of catalyst, which delays the
reaction and increases the overall cost of the technique. In
light of this, and to optimize metal components of the cat-
alyst and evaluate the combined effects and interactions
of catalyst metal components, that is, Cu, Mg, and Al, the
experimental design of response surface methodology (RSM)
Thus, the specific aims of the research are to (a) optimize
the metal content of Cu/Mg/Al–chitosan for the ozonation
of landfill leachate, (b) assess the stability of the catalyst, (c)
evaluate the initial effect of pHs of different landfill leachates
on the catalyst performance, and (d) study the characteris-
tics of the catalyst, which, to the best of author’s knowledge,
has not been reported to date. We noted that other aspects of
the catalytic ozonation of landfill leachate including mecha-
nism of leachate degradation, metal leaching from Cu/Mg/
Al–chitosan, and leachate treatment by single ozonation and
adsorption onto Cu/Mg/Al–chitosan were presented else-
2. Materials and methods
Philocheras lowisi shrimp was directly collected from
the Persian Gulf and shipped to laboratory within 1 h. The
shrimp was deshelled and shell wastes were deacetylated
using a method described in literature . The achieved
chitosan passed through an American Society for Testing
and Materials (ASTM) sieve (mesh no. 14 and 18) to obtain
particles with size in the range of 1–1.41 mm. The reagents
and chemicals used in this study were of analytical grade and
used during the experiments without any purification.
2.2. Cu/Mg/Al–chitosan synthesis
Cu/Mg/Al–chitosan was synthesized by employing cop-
per, magnesium, and aluminium nitrates and Na2CO3/NaOH.
The catalyst synthesis was performed in a 500-mL flask con-
taining 0–4 mmol/L metal nitrates of Cu2+, Mg2+, and Al3+ to
achieve the favoured Cu/Mg/Al molar ratio with high effi-
ciency in COD removal from landfill leachate. The optimiza-
tion of the metal content of the Cu/Mg/Al–chitosan catalyst
was done based on the RSM design. An amount of 10 g of
chitosan particles were poured into to the solution contain-
ing the desired amount of metal (based on the RSM design)
and then a 250-mL base solution with Na2CO3 (0.05 mol) and
NaOH (0.8 mol) were added dropwise for 4 h into the flask
and agitated vigorously by a magnetic heater-stirrer at 45°C.
Then, the temperature of the mixture solution was lowered
to around 25°C and the solution filtered. The obtained sol-
ids were washed several times by double-distilled water
until the supernatant was nitrate-free. This was then dried at
105°C for 24 h.
2.3. Landfill leachate sampling
Leachate samples were taken in polyethylene bottles from
a municipal landfill site situated in the Kahrizak area, Tehran,
Iran. This landfill came into being in 1967 on a total area of
1,300 ha. The annual rainfall of the Kahrizak area is 232.8 mm.
Tehran city solid waste amounting to approximately 6,400 ton-
ne/d enter the landfill site. In this site, several leachate ponds
of different ages were found. Three samples from different
leachate ponds with various initial pHs (5.5, 7.3, and 9.1)
and other characteristics were taken. Leachate samples were
transferred to the laboratory in closed containers at 4°C in
accordance with the standard methods . In the labora-
tory, the raw landfill leachate samples were pre-filtered via a
0.45 µm glass fibre filter to remove large particles (suspended
solids) and debris, and maintain uniformity of tested sam-
ples. The main properties of the pre-filtered leachates (which
were used for the experiments) are presented in Table 1.
2.4. Batch adsorption experiments
A catalytic ozonation of the landfill leachate was done in a
batch reactor with a total volume of 300 mL. The reactor was
equipped with a glass sparger, an ozone generator, a sintered
glass diffuser to distribute the ozone stream to the leachate, an
ozone off-gas removal system, valves, and tubing. All tests were
carried out in a batch mode, with a constant leachate volume
91D.R. Vakilabadi et al. / Desalination and Water Treatment 80 (2017) 89–99
of 150 mL, at a temperature of 24°C ± 1°C. The known amount
of the catalyst (dcat = 1.67 g/cm3) was then poured in the reactor
and immediately after that the ozone was injected to initiate the
reaction. Ozone was generated by a generator (Model 3S-A3,
Tonglin Technology, Beijing) with maximum ozone capacity of
5 g/h. The ozone dose was adjusted at 3.5 mg/min during the
tests. The ozone in the off-gas of the reactor was neutralized by
a concentrated KI solution. Due to analysis requirement and
the small volume of the reactor, sample was not used during
a test but instead the whole volume was withdrawn at the end
of each test scheduled for a determined time. The catalyst was
kept suspended in the reactor by using the ozone gas bub-
bling and before the sample was taken, the ozone gas flow was
stopped, and thus, the catalyst was deposited. The experiments
of this study were carried out in three sections and under con-
ditions specified in Table 2. The amount of parameters (catalyst
dose and ozone flow rate) was chosen based on the pre-test
and published researches . Once the reactions were com-
plete, the content in the reactor was centrifuged for 3 min at
10,000 rpm using a centrifuge (TDL-5Z, Hunan Xingke, China).
The COD value of the supernatant showed the residual COD in
the effluent after the overall process [30,33]. The COD removal
was calculated by the ‘(CODinitial – CODfinal/CODinitial) × 100’ for-
mula. Section 2.2 of the experiments was designed by the RSM.
Section 2.4, the tests were done in a triplicate mode and the
mean of measurements were reported.
The leachate pH measurements were done using a pH
meter equipped with a specific electrode (Jenway 3505). The
elemental composition of the Cu/Mg/Al–chitosan was deter-
mined by using an electron dispersive spectroscopy (EDAX,
QUANTA 200 FEG). The Brunauer–Emmett–Teller (BET) sur-
face area and the pore structure of the Cu/Mg/Al–chitosan
were evaluated using a Builder SSA-420 instrument at –196°C.
The pore area and volume for mesopores (2–50 nm) were
calculated by the Barrett–Joyner–Halenda (BJH) method.
By using a Fourier transform infrared (FTIR) spectroscopy
spectrometer, the functional groups on the surface of fresh
and used Cu/Mg/Al–chitosan were recognized (NICOLET
5700-FTIR) in the range of 400–4,000 cm–1. The examination
of surface morphologies of chitosan and Cu/Mg/Al–chitosan
samples were performed by scanning electron microscopy
(SEM, Sirion from FEI). The crystallinity and phase structure
of the Cu/Mg/Al–chitosan were assessed by X-ray power dif-
fraction (XRD, XD-3A, Shimadzu) recorded from 20° to 80°
at 5° cm−1 by using Cu Kα radiation (λ = 0.15418 nm) under
40 kV and 30 mA. The pHpzc of the catalyst was attained
according to the previous study  using the pH drift
method and the batch equilibrium technique with 1:250 solid
mass to liquid ratio in 0.01 M NaCl solution as an inert elec-
trolyte. The initial pH of the NaCl solution was set at 2–12 by
adding NaOH or HCl (0.1 N). The suspensions were allowed
to equilibrate for 24 h at 24°C ± 1°C with 130 rpm mixing.
After that, the suspensions were filtered through 0.42 µm
filter, and the pH values of filtrates were measured using
a pH meter. The iodine number was measured by using a
0.1 N standardized iodine solution and titration using 0.1 N
sodium thiosulphate . The leachate COD was measured
by the potassium dichromate oxidation according to method
described in Standard Methods for Examination of water and
wastewater . The BOD5 parameter was analyzed follow-
ing the 5210-D test using an OxiTop (manometric respirom-
etry) . Measurement of metals ions was performed using
a Varian AA240 atomic absorption spectrophotometer. Other
analyses such as ammonia-nitrogen (NH4+), nitrate-nitrogen
(NO3–), nitrite-nitrogen (NO2–), and chloride were also done
according to the standard methods .
2.6. Response surface methodology
The RSM is the main branch of an experimental design,
which is used to evaluate the effect of several factors and their
interaction on the response. The RSM is a combination of sta-
tistical and mathematical methods. This method is useful
for developing and optimizing the variables and response,
and providing a fewer number of design points, thus reduc-
ing the overall cost of the experiment. In recent years, the
attempts have increased to better understand the catalytic
processes and the effects of different parameters on catalyst
performance . The maximum application of the RSM is
in cases where several variables affect the system response
. The RSM consists of three steps: design of experiments,
response surface modelling, and optimization . In design
and analysis, the optimization of the metal components of
the catalyst was done by using the ‘Design Expert software’
(Ver.8.2, Minneapolis, USA). In the present study, the central
Properties of leachate samples in this study
Property (unit) Sample 1 Sample 2 Sample 3
COD (mg/L) 52,870 35,070 40,700
BOD5 (mg/L) 10,580 4,100 2,100
TOC (mg/L) 39,877 23,455 34,040
NO3– (mg/L) 178 319 332
NO2– (mg/L) 72 149 113
NH4+ (mg/L) 121 189 652
Cl– (mg/L) 2,137 1,947 728
pH (unit of pH) 5.5 7.3 9.1
Colour Brown Dark brown Very dark brown
The experimental sections of the study and their conditions
Section Experimental runs Test conditions
2.2 Optimization of
on the RSM
Ozone dose: 3.5 mg/min,
catalyst dose: 20 mg/L,
reaction time: 50 min,
leachate pH: 9.1
2.4 Stability test of the
Ozone dose: 3.5 mg/min,
catalyst dose: 20 mg/L,
reaction time: 50 min,
leachate pH: 9.1
2.4 Treatment of
different initial pHs
Ozone dose: 3.5 mg/min,
catalyst dose: 20 mg/L,
reaction time: 5–50 min,
leachate pH: 5.5, 7.3, and 9.1
D.R. Vakilabadi et al. / Desalination and Water Treatment 80 (2017) 89–9992
composite of the RSM statistical design was chosen to study
the effect of metal components of the catalyst (Cu/Mg/Al–
chitosan) on the COD removal from the landfill leachate by
means of Minitab 14 software. For this purpose, three main
elements in the catalyst were chosen: Cu (mmol/L), Mg
(mmol/L), and Al (mmol/L). The design matrix contained a
‘23 = 8’ factorial design added to six replications and six axial
points (α = 1.6) at the midpoints (all factors at zero level) to
estimate the residual error. Thus, the total number of exper-
iments required for this study was 20. Experimental data
acquired by the RSM were also analyzed using the Minitab
14 software. The variables were coded according to Eq. (1):
where Xi and xi are the coded and uncoded (original) value
of the i-th test variable, respectively. The x
∗ parameter is the
value of xi at the midpoint of the considered range of the fac-
tor, and Δxi is the step size. The correlation between response
and the selected factors can be expressed by a quadratic
equation that is given as :
YB BX BX BXX
=+ ++ +
== ≠01 1
where Y is the predicted response (predicted COD removal),
Xi, Xj are the independent variables in coded levels, Bi, Bii, Bij
are the coefficients of linear, quadratic, and interaction effect,
respectively, B0 is the model coefficient, n is factor number (or
independent variables), and ε parameter is the model error.
In the catalytic ozonation system, three components of the
catalyst, that is, Mg, Cu, and Al were selected as indepen-
dent variables for the central composite design (CCD) as X1,
X2, and X3, respectively. The mathematical relationship of the
response on these variables can be estimated by the following
quadratic polynomial equation :
YC CX CX CX CXXCXX
23 23 11 1
where C0 is a constant value, C1, C2, and C3 denotes linear coef-
ficients, C12, C13, and C23 denotes cross-product coefficients,
and C11, C22, and C33 denotes quadratic coefficients. The qual-
ity of fit of the polynomial model equation was expressed
by the determination coefficient (R2) and the responses were
completely analyzed using analysis of variance (ANOVA).
The factor interaction was evaluated by constructing the
response surface and contour plots based on the level effects
of corresponding factors. With the aim of the maximum COD
removal and Mg and minimum level of Al and Cu, the opti-
mum condition was obtained by using the response opti-
mizer function in the Minitab software.
3. Results and discussion
3.1. Characteristics of optimized catalyst
The FTIR spectra of a simple chitosan and optimized
Cu/Mg/Al–chitosan are shown in Fig. 1. It shows the chitosan
was successfully modified by Cu/Mg/Al and there was inter-
action between chitosan and Cu/Mg/Al crystal surfaces. The
crystalline phases such as green rust and ferrihydrite were
identified by the FTIR in the literature . The spectral fea-
tures similarity affirmed the successful modification of the
chitosan particles by Cu/Mg/Al. The absorption bands of –OH
and –NH2 stretching modes at 3,393 and 1,647 cm–1 undergo
discernible shifts when compared with simple chitosan on its
own with the Cu/Mg/Al–chitosan, indicating an interaction
between metals and the chitosan surface. A complex forma-
tion between Cu/Mg/Al and an amino group (–NH2) is most
likely to take place in monodentate mode, which will expect-
edly leave more space on the surface of Cu/Mg/Al. Both –
NH2 and –OH groups of chitosan may be involved in inter-
actions with Cu/Mg/Al. However, the –NH2 group behaviour
has more to do with the particle stabilization aspect derived
from stronger binding strength with metals. Generally, there
was evidence of chitosan being modified by Cu/Mg/Al.
XRD is commonly used for phase identification. The
XRD pattern of the simple chitosan and Cu/Mg/Al–chitosan
is given in Fig. 2. The XRD analysis confirms the results of
the FTIR measurements and show further evidence for the
formation of crystalline material in the chitosan sample.
According to Fig. 2, the crystallinity of Cu/Mg/Al–chitosan
shows many sharp diffraction peaks between 2Ө = 6°–45°,
while no such peaks are visible in the XRD of simple chi-
tosan, perhaps because of the trapping of Cu/Mg/Al by
chitosan. An XRD analysis confirmed that the Cu/Mg/Al–
chitosan particles contained Mg (zero-valent magnesium),
Fig. 1. FTIR spectra of (a) chitosan and (b) Cu/Mg/Al–chitosan.
Fig. 2. XRD patterns of (a) Cu/Mg/Al–chitosan and (b) simple
93D.R. Vakilabadi et al. / Desalination and Water Treatment 80 (2017) 89–99
aluminium oxide, and copper oxide (Fig. 2). The charac-
teristic peaks of Mg appeared at 44.93° and 51.01° (2θ)
(JCPDS 89-2838), the peak of aluminium oxide was at 82.17°
(2θ) ( JCPDS 05-0667), and the peak of copper oxide was at
64.38° (2θ) (JCPDS 89-2838). An absence of peaks at 35° and
42° (2θ) in the XRD diffractogram revealed that no oxides of
magnesium were formed.
The elemental composition of the prepared catalyst, that
is, Cu/Mg/Al–chitosan both fresh and used for the leach-
ate treatment as well fresh chitosan, was determined using
the EDAX technique listed in Table 3. Many notes could be
implied from this table: first, the presence of the elements
Cu, Mg, and Al in the fresh catalyst confirmed the success-
ful modification of chitosan; however, low levels of these
elements naturally existed in the chitosan alone. Second, the
metal leaching from the Cu/Mg/Al–chitosan is not consider-
able; this result will be important from practical and toxicol-
ogy point of view. Finally, an increase in the oxygen content
of used Cu/Mg/Al–chitosan could be a sign of oxidation con-
ditions in the solution.
SEM images of the chitosan and the Cu/Mg/Al–chitosan
samples are illustrated in Fig. 3. A comparison of the two
SEM micrographs makes it clear that the crystalline com-
pound particles (with the mean crystal domain of 80–100 nm)
with uniform size and roughly hexagonal plates were formed
on the surface of prepared catalyst .
Basic characteristics of Cu/Mg/Al–chitosan are presented
in Table 4. Surface analysis showed that Cu/Mg/Al–chitosan
particles had a BET multipoint surface area of 64.5 m2/g and
a total pore volume of 0.37 cm3/g. The relatively high BET
surface area and pore volume of the catalyst was provided
appropriate space for reacting with organic matter in the
leachate . The pHpzc of the catalyst was obtained to be 7.8,
signifying a negative surface charge for a working solution
pH greater than 7.8 and a positive surface charge for a solu-
tion pH below 7.8. According to Table 4, the pore sizes of the
modified chitosan falls within the range of 2–50 nm, indicat-
ing that the Cu/Mg/Al–chitosan catalyst was a mesoporous
type. Moreover, in Table 4, the high value of the C constant
of BET and iodine number imply that the catalyst is effective
in treating solutions with a high content of organic and inor-
ganic matter [42,43].
3.2. Optimization of metal content of Cu/Mg/Al–chitosan catalyst
3.2.1. Central composite design
The optimization of the amount of Cu, Mg, and Al ele-
ments as components of Cu/Mg/Al–chitosan catalyst in the
COD removal from the landfill leachate was performed using
CCD. The levels of Cu, Mg, and Al elements in the catalyst as
selected preparation parameters were presented in Table 5 in
EDAX analysis of chitosan and fresh and used Cu/Mg/
Element Value (wt%)
O 35 37 53
C 22 23 29
N 8 9 12
S 1.5 1.3 1.5
Mg 16 14.3 1.2
Cu 11 10 1
Al 6 5.1 0.25
Na 0.5 0.3 0.7
Characteristics of Cu/Mg/Al–chitosan catalyst which had highest
efficiency based on RSM
BET, m2/g 64.5
C constant of BET 498
BJH, m2/g 96.32
Iodine number, mg/g 132.12
Pore volume, cm3/g 0.37
Pore size, nm 17.1
Catalyst type Mesoporous
Size distribution, mm 1–1.41
Fig. 3. SEM images of (a) chitosan and (b) Cu/Mg/Al–chitosan.
D.R. Vakilabadi et al. / Desalination and Water Treatment 80 (2017) 89–9994
coded and actual values. The critical ranges of selected param-
eters were determined by preliminary tests based on the litera-
ture review [30,42]. The designed experiments and their results
for COD removal from the landfill leachate are listed in Table 6.
3.2.2. The statistical analysis and regression model
ANOVA is required to test the significance and adequacy
of the model. The results of ANOVA of this model are tabu-
lated in Table 7. A very low probability value (pmodel < 0.0001)
and a very high F value (Fmodel = 15.75, much greater than unity)
showed that the model created by the software was highly sig-
nificant and ensured an accurate representation of the exper-
imental data. The coefficient of determination (R2 = 0.942)
indicated that the regression model represented 94.2% of the
experimental results and only about 5.8% of the variability in
the response could not be explained by this model. The qual-
ity of the developed model was evaluated, based on the cor-
relation coefficient value. The ‘adjusted R2’ (a measure of the
variation amount of the mean explained by the model) and
the predicted R2 (a measure of goodness the model predicts a
response data) should be within about 0.20 of each other to be
in reasonable agreement. If they are not, the data or the model
may be faced with a problem. Here, the ‘predicted R2= 0.942’
and ‘adjusted R2= 0.957’ have a satisfactory agreement, as the
difference is less than 0.2. The approximating equation of the
fitted response surface (COD removal) was also checked for
the model adequacy using ANOVA results and the diagnos-
tic plots of Figs. 4–6. As shown in Fig. 6, the predicted vs.
observed points are placed relatively near a straight line illus-
trating low discrepancies between them.
Variables and their levels in the experimental design for synthe-
sis of Mg/Cu/Al–chitosan
Factor Name Type Low
A Mg, mmol/L Numeric 0 4 –1 1
B Al, mmol/L Numeric 0 4 –1 1
C Cu, mmol/L Numeric 0 4 –1 1
Experimental design matrix and experimental results for catalyst
synthesis based on the RSM
Std Run A: Mg
6 1 3.18 0.81 3.18 96
19 2 2 2 2 86
20 3 2 2 2 84
16 4 2 2 2 83
14 5 2 2 4 89
8 6 3.18 3.18 3.18 84
12 7 2 4 2 88
4 8 3.18 3.18 0.81 58
5 9 0.81 0.81 3.18 33
11 10 2 0 2 50
1 11 0.81 0.81 0.81 33
15 12 2 2 2 88
7 13 0.81 3.18 3.18 65
10 14 4 2 2 98
9 15 0 2 2 22
17 16 2 2 2 88
13 17 2 2 2 88
2 18 3.18 0.81 0.81 47
18 19 2 2 2 88
3 20 0.81 3.18 0.81 52
ANOVA for the fit of the experimental data to response surface quadratic model for catalyst synthesis
Source Sum of squares Degree of freedom Mean square F Value p Value
Model 9,996.806 9 1,110.75 15.75 <0.0001 Significant
A-Mg 4,319.972 1 4,319.97 61.25 <0.0001 Significant
B-Al 713.8644 1 713.86 10.12 0.0098 Significant
C-Cu 393.404 1 393.40 5.57 0.0398 Significant
AB 338.0008 1 338.00 4.79 0.0534 Significant
AC 480.4992 1 480.49 6.81 0.0260 Significant
BC 12.50008 1 12.50 0.17 0.6826 Insignificant
A21,544.012 1 1,544.01 21.89 0.0009 Significant
B2744.0382 1 744.03 10.55 0.0087 Significant
C2609.1249 1 609.12 8.63 0.0148 Significant
Residual 705.1945 10 70.51 – – –
Lack of fit 677.4802 4 169.37 136.66 0.0522 Insignificant
Pure error 27.71429 6 4.61 – – –
Corrected total 10,702 19 – – – –
R2 = 94.2%, R2(adj) = 95.7%.
95D.R. Vakilabadi et al. / Desalination and Water Treatment 80 (2017) 89–99
The coefficient of the model equation and their statisti-
cal significance are evaluated using Design-expert 8.2 soft-
ware. The quadratic regression model for the COD removal
in terms of coded factors is given by Eq. (4):
COD removal = +79.75 + 32.74A + 13.31B + 8.67C – 18.38AB
+ 17.48AC – 2.82BC – 29.43A2 – 20.43B2
– 15.14C2 (4)
where A, B, and C are the coded values of the process vari-
ables Mg, Al, and Cu content of the catalyst, respectively. The
negative sign ahead of the terms indicates antagonistic effect,
while positive sign indicates synergistic effect.
The amount of ‘F value’ of ‘Lack of fit’ signifies the resid-
ual sum of squares resulting from the model not fitting the
data. It is the sum of squared deviations between the mean
response at each factor level and the corresponding fitted
value. As shown in Table 7, the ‘Lack of fit F value= 36.66’ indi-
cates lack of fit is not significant in relation to the pure error.
Non-significant lack of fit is suitable and indicates that the
above model is appropriate to predict the COD removal from
the landfill leachate within the range of variables studied.
3.2.3. 3D surface plot analysis for COD removal
The resultant surface response 3D plots of COD removal
as a function of two independent variables (A, B, and C) have
been presented in Figs. 7(a)–(c). The perturbation plot helps
to compare the effect of all the factors at a particular point in
the design space. The response (COD removal) is plotted by
Fig. 4. Normal percentage probability and studentized residual
plot for COD removal from the landfill leachate.
Fig. 5. The studentized residuals and predicted response plot for
COD removal from the landfill leachate.
Fig. 6. The actual and predicted plot for COD removal from the
Fig. 7. Response surface plot and its contour plot for COD
removal from landfill leachate: (a) effects of Mg–Al, (b) effects of
Cu–Al, and (c) effect of Cu–Mg.
D.R. Vakilabadi et al. / Desalination and Water Treatment 80 (2017) 89–9996
changing only two factors over its range, while another fac-
tor is considered in fixed value. Design-Expert sets the default
reference point at the midpoint (coded 0) of all factors. A steep
curvature or slope in a factor shows that the response is sensi-
tive to that factor. A relatively flat line shows unresponsiveness
to the given factor—in this case, the plot is like ‘one-factor-at-
a-time’ testing and does not show the effects of interactions.
Generally, from Fig. 7(a), the COD removal is significantly
increased with an increase of Mg (A) value but a moderate
value of the Cu (C) factor was better than high and low lev-
els. Fig. 7(b) shows that B and C play a different role in the
reaction progress. In the high range of the data (coded), the
most important role in COD removal was observed for Cu (C),
while, in the middle range of data, Al (B) was more important.
According to Fig. 7(c), it is clear that the two parameters (A
and B) behaved similarly in the COD removal. In other words,
increasing the Mg (A) and Al (B) contents of the catalyst accel-
erated the COD removal. As shown in Fig. 7, increasing the
Mg concentration significantly improved the oxidation of the
landfill leachate as the amount of COD residual seemed to be
the lowest due to the presence of a significant amount of sur-
face oxygen species [30,44] on the catalyst surface, whereas the
catalyst activity was enhanced to the middle levels of Cu (here,
2 mmol/L). At low Al concentrations, Al did not play a criti-
cal role and, thus, Cu/Mg/Al became a bimetallic of Cu/Mg.
In the Cu/Mg bimetal, as reported in previous studies [23,45],
the COD removal could be justified by the formation of the
metal–hydride (M–H) structure on copper particles and the
dissociation of molecular hydrogen or other hydrogen sources
on the surface of copper element serving as a direct agent for
oxidizing the leachate COD. No information could be found
in literature on landfill leachate oxidation by Cu/Mg/Al cata-
lyst to compare with our results. However, Tien Thao and Kim
Huyen  reported that the Cu content in a hydrotalcite-like
material affected the activity during the conversion of styrene
to benzaldehyde but did not affect catalyst selectivity. Thus,
the textural feature of the catalyst and regulation of the cata-
lyst metal content is vital to the oxidation process, and the sta-
bilized metal sites act as active sites for the selective oxidation.
3.2.4. Optimization response and verification test
Response optimizer of the software helps to identify the
combination of input variable settings that jointly optimize
a single response or a set of responses . The software
was used to predict the best condition for the conversion by
pre-setting certain criteria, as shown in Table 8. The optimum
values of the independent variables, that is, metal components
of the Cu/Mg/Al–chitosan catalyst are selected by consider-
ing the toxicology and health point of view of the effluent.
Thus, in the software optimization section, the desired goal
of each operational condition was chosen as ‘maximize’ for
Mg and percentage COD removal (as response) and ‘mini-
mize’ for Al and Cu due to the possible toxicity in the treated
solution. However, the element of Cu and Al are used in
water works during the world [47,48]. According to Table 8,
the predicted response for the loading of Mg = 4 mmol/L,
Cu = 0.89 mmol/L, and Al = 2 mmol/L on the chitosan to syn-
thesize the catalyst was that the COD removal from landfill
leachate would be 79.89%, while the actual result for the COD
removal was 81.35% in identical test conditions. This revealed
an excellent agreement between the model-predicted and
experimental values. Thus, the obtained data allowed us to
suggest the superior operational conditions for synthesizing
the catalyst in order to optimize the response. These studies
confirm that the predicted results are fully corroborated by
the experimental values. Accordingly, the RSM successfully
optimized the metal content of the catalyst to maximize the
COD removal from landfill leachate.
3.3. Catalyst stability
The stability of a catalyst is important from an environmen-
tal and economic perspective. The first four solutions (from 30
solutions) for the synthesis of the catalyst were provided as
point predictions by the RSM. These predicted solutions were
tested to assess their stability. For this test, after the completion
the reaction of the catalyst and landfill leachate, the reacted
catalyst was separated by the Whatman filter paper, and dried
at 105°C for 6 h in order to reuse the catalyst and study its
effective lifespan. Mg/Cu/Al–chitosan was successfully recy-
cled three times. The results, in Fig. 8, show that the COD
removal percentage was in excess of 50% for all options in the
third cycle. The colour removal by the catalyst (first solution
by the RSM) is shown in Fig. 9 for visual comparison of the
treated leachate after each cycle. The Mg/Cu/Al–chitosan cat-
alyst maintained sustained performance even after being used
for four cycles, and the COD removal decreased only slightly.
Therefore, the Mg/Cu/Al–chitosan catalyst has a good stabil-
ity and long lifespan as a heterogeneous solid-based catalyst
for treating the landfill leachate.
The initial and final concentration of Al, Mg, and Cu
ions for three cycles in the aquatic environment was also
The preset criteria for optimization of the maximum COD
removal for catalyst synthesis based on the RSM
Name Goal Level Low
A Mg Maximize 4 0 4
B Al Minimize 0.89 0 4
C Cu Minimize 2 0 4
R1 COD removal Maximize 79.89%
Desirability = 0.89.
COD removal (%)
Soluon 1 Soluon 2
Soluon 3 Soluon 4
Fig. 8. The stability of catalyst (leachate pH: 9.1; catalyst dose:
20 mg/L; reaction time: 50 min, stirring rate: 150).
97D.R. Vakilabadi et al. / Desalination and Water Treatment 80 (2017) 89–99
measured. The initial concentration of all metal ions was not
detectable in all samples. Nevertheless, after leachate treat-
ment by both fresh and recycled Mg/Cu/Al–chitosan catalyst,
the relatively low level of Mg2+ ions (64–140 µg/L) was found
in the samples. The amount of Cu2+ leached to the treated
leachate during the experiment was lower than the maxi-
mum contaminant level of copper ions in drinking water
(1.3 mg/L) . The amount of Al3+ was not detectable in the
leachate samples treated by both fresh and recycled cata-
lyst. The partial leaching of these metal ions via the repeated
use of Cu/Mg/Al(CMA)–chitosan led to a reduction in COD
removal especially in third round of catalyst recycling.
3.4. Treatment of different landfill leachates with various
Three landfill leachates with different initial pH (9.1, 7.3,
and 5.5) were sampled to assess the applicability of the opti-
mized catalyst in the COD removal of landfill leachates with
various initial pH. Fig. 10 presents the oxidation of the leach-
ate by CMA–chitosan–O3 as a function of the reaction time,
showing the considerable effect of the leachate pH on the
leachate COD removal. Two notes are derived from Fig. 10:
(I) the leachate COD removal for all studied pH increased
by increasing the time of reaction, and (II), for each reac-
tion time, the COD removal increased by increasing the pH
value. For instance, in a 50-min reaction time under test con-
ditions, the amount of COD removal increased from 49% at
a pH of 5.5 to 78% at pH of 9.1. The increase in the leachate
COD removal as a function of pH can be explained by the
ozone transfer from the gas to the leachate, ozone decompo-
sition reaction, and the characteristics of the catalyst surface
 which interpreted in details elsewhere . Briefly, an
increase in pH enhanced the ozone decomposition through
both homogeneous (due to an increase in the quantity of the
hydroxyl anions, which appropriate the decay of ozone in
the leachate) and heterogeneous (catalytic) reactions .
These reactions led to an improvement of the ozone mass
transfer rate from the gas to the liquid phase, and thus lead-
ing to an increase in the reactive oxidizing radical species.
Therefore, a higher COD removal was attained at the alka-
line pH of 9.
The effects of liquid phase pH on the degradation of
organics in the catalytic oxidation process have been stated in
the literature. For instance, a reduction in the mineralization
of phenolic acids in a catalytic ozonation using Mn–Ce–O cat-
alyst as a function of the solution pH between 3 and 10 has
been reported by Martins and Quinta-Ferreira . Moussavi
and Khosravi  observed the decolourization of an azo
dye by catalytic oxidation with a pistachio hull biomass at an
alkaline pH . Accordingly, it can be deduced that the way
the leachate pH affects COD removal completely depends on
both the type and structure of the reacting compound and
the type of the catalyst. Thus, the optimal pH of the catalytic
oxidation must be identified for each specific condition and
herein, based on the above discussion, it is concluded that the
Cu/Mg/Al–chitosan is a favoured approach for the treatment
of mature landfill leachate, which usually have an alkaline pH.
The RSM was used to optimize the Mg, Cu, and Al
content of the Cu/Mg/Al–chitosan for COD removal from
landfill leachate. The adequacy of the quadratic model was
adequately verified by the validation of experimental data.
Process optimization was done and the experimental val-
ues obtained for the COD removal were found to agree
sufficiently with the predicted values. The optimal finding
for the maximization of COD removal by the RSM based
on CCD found a 79.89% under optimized metal value of
Mg = 4 mmol/L, Cu = 0.89 mmol/L, and Al = 2 mmol/L in the
Cu/Mg/Al–chitosan. Moreover, 3D surface plots showed that
the Mg content of the Cu/Mg/Al–chitosan played a signifi-
cant role in maximizing COD removal. Four predicted solu-
tions for the synthesis of the catalyst by the RSM were tested
to assess its stability. The optimized catalyst was evaluated
for the treatment of three types of leachate landfill with dif-
ferent initial pH (9.1, 7.3, and 5.5). The results showed that
the alkaline leachate was easily treated by the catalyst. The
characteristics of the optimized catalyst were fully studied,
confirming that the impregnation of the chitosan by Mg,
Cu, and Al was successful. Overall, results demonstrate that
Cu/Mg/Al–chitosan–ozone process can decrease COD from
the landfill leachate, offering a promising option to eliminate
COD from aqueous media.
The authors express their appreciation to Bushehr
University of Medical Sciences, Iran for the instrumental
Fig. 9. The landfill leachate: (a) cycle 3, (b) cycle 2, (c) cycle 1, and
(d) raw landfill leachate.
COD removal (%)
Reacon me (min)
pH 9.1pH 7.3pH 5.5
Fig. 10. Treatment of landfill leachate with different initial pHs
(CMA–chitosan dose: 20 mg/L, reaction time: 50 min, ozone
flow: 3.5 mg/min).
D.R. Vakilabadi et al. / Desalination and Water Treatment 80 (2017) 89–9998
 M. Hassan, Y. Zhao, B. Xie, Employing TiO2 photocatalysis to
deal with landfill leachate: current status and development,
Chem. Eng. J., 285 (2016) 264–275.
 F.C. Moreira, J. Soler, A. Fonseca, I. Saraiva, R.A. Boaventura,
E. Brillas, V.J. Vilar, Incorporation of electrochemical advanced
oxidation processes in a multistage treatment system for
sanitary landfill leachate, Water Res., 81 (2015) 375–387.
 P. Oulego, S. Collado, A. Laca, M. Díaz, Tertiary treatment of
biologically pre-treated landfill leachates by non-catalytic wet
oxidation, Chem. Eng. J., 273 (2015) 647–655.
 T. Mariam, L.D. Nghiem, Landfill leachate treatment using
hybrid coagulation-nanofiltration processes, Desalination, 250
 R.C. Contrera, K.C. da Cruz Silva, D.M. Morita, J.A. Domingues
Rodrigues, M. Zaiat, V. Schalch, First-order kinetics of landfill
leachate treatment in a pilot-scale anaerobic sequence batch
biofilm reactor, J. Environ. Manage., 145 (2014) 385–393.
 S. Ismail, A. Tawfik, Performance of passive aerated immobilized
biomass reactor coupled with Fenton process for treatment of
landfill leachate, Int. Biodeterior. Biodegrad., 111 (2016) 22–30.
 W.-Y. Ahn, M.-S. Kang, S.-K. Yim, K.-H. Choi, Advanced landfill
leachate treatment using an integrated membrane process,
Desalination, 149 (2002) 109–114.
 A.I. Zouboulis, M.D. Petala, Performance of VSEP vibratory
membrane filtration system during the treatment of landfill
leachates, Desalination, 222 (2008) 165–175.
 B. Zhou, Z. Yu, Q. Wei, H. Long, Y. Xie, Y. Wang, Electrochemical
oxidation of biological pretreated and membrane separated
landfill leachate concentrates on boron doped diamond anode,
Appl. Surf. Sci., 377 (2016) 406–415.
 D.M. Bila, A.F. Montalvao, A.C. Silva, M. Dezotti, Ozonation
of a landfill leachate: evaluation of toxicity removal and
biodegradability improvement, J. Hazard. Mater., 117 (2005)
 N. Amaral-Silva, R.C. Martins, S. Castro-Silva, R.M. Quinta-
Ferreira, Ozonation and perozonation on the biodegradability
improvement of a landfill leachate, J. Environ. Chem. Eng., 4
 V. Oloibiri, I. Ufomba, M. Chys, W.T.M. Audenaert, K.
Demeestere, S.W.H. Van Hulle, A comparative study on
the efficiency of ozonation and coagulation–flocculation as
pretreatment to activated carbon adsorption of biologically
stabilized landfill leachate, Waste Manage., 43 (2015) 335–342.
 S. Lee, J. Hur, Heterogeneous adsorption behavior of landfill
leachate on granular activated carbon revealed by fluorescence
excitation emission matrix (EEM)-parallel factor analysis
(PARAFAC), Chemosphere, 149 (2016) 41–48.
 M. Morozesk, M.M. Bonomo, L.D. Rocha, I.D. Duarte,
E.R.L. Zanezi, H.C. Jesus, M.N. Fernandes, S.T. Matsumoto,
Landfill leachate sludge use as soil additive prior and after
electrocoagulation treatment: a cytological assessment using
CHO-k1 cells, Chemosphere, 158 (2016) 66–71.
 O. Rojviroon, T. Rojviroon, S. Sirivithayapakorn, Removal of
color and chemical oxygen demand from landfill leachate by
photocatalytic process with AC/TiO2, Energy Procedia, 79
 J.-M. Wang, C.-S. Lu, Y.-Y. Chen, D.T. Chang, H.-J. Fan, Landfill
leachate treatment with Mn and Ce oxides impregnated GAC–
ozone treatment process, Colloids Surf., A, 482 (2015) 536–543.
 B. Calli, B. Mertoglu, B. Inanc, Landfill leachate management in
Istanbul: applications and alternatives, Chemosphere, 59 (2005)
 F.J. Rivas, F. Beltrán, O. Gimeno, B. Acedo, F. Carvalho,
Stabilized leachates: ozone-activated carbon treatment and
kinetics, Water Res., 37 (2003) 4823–4834.
 A.L.C. Peixoto, M.B. Silva, H.J. Izário Filho, Leachate treatment
process at a municipal stabilized landfill by catalytic ozonation:
an exploratory study from Taguchi orthogonal array, Braz. J.
Chem. Eng., 26 (2009) 481–492.
 L. Pan, M. Ji, X. Wang, L. Zhao, Influence of calcination
temperature on TiO2 nanotubes’ catalysis for TiO2/UV/O3
in landfill leachate solution, Trans. Tianjin Univ., 16 (2010)
 S. Zhang, X. Zhao, H. Niu, Y. Shi, Y. Cai, G. Jiang,
Superparamagnetic Fe3O4 nanoparticles as catalysts for the
catalytic oxidation of phenolic and aniline compounds, J.
Hazard. Mater., 167 (2009) 560–566.
 B. Chen, G. Ma, Y. Zhu, J. Wang, W. Xiong, Y. Xia, Metal-organic-
framework-derived bi-metallic sulfide on N, S-codoped porous
carbon nanocomposites as multifunctional electrocatalysts, J.
Power Sources, 334 (2016) 112–119.
 S. Mortazavi, B. Ramavandi, G. Moussavi, Chemical reduction
kinetics of nitrate in aqueous solution by Mg/Cu bimetallic
particles, Environ. Technol., 32 (2011) 251–260.
 A. Mantilla, F. Tzompantzi, J. Fernández, J.D.
Góngora, G. Mendoza, R. Gomez, Photodegradation of
2,4-dichlorophenoxyacetic acid using ZnAlFe layered double
hydroxides as photocatalysts, Catal. Today, 148 (2009) 119–123.
 F. Tzompantzi, G. Mendoza-Damián, J. Rico, A. Mantilla,
Enhanced photoactivity for the phenol mineralization on
ZnAlLa mixed oxides prepared from calcined LDHs, Catal.
Today, 220 (2014) 56–60.
 J.S. Valente, F. Tzompantzi, J. Prince, J.G. Cortez, R. Gomez,
Adsorption and photocatalytic degradation of phenol and
2,4 dichlorophenoxiacetic acid by Mg–Zn–Al layered double
hydroxides, Appl. Catal., B, 90 (2009) 330–338.
 R. Sahu, B. Mohanta, N. Das, Synthesis, characterization and
photocatalytic activity of mixed oxides derived from ZnAlTi
ternary layered double hydroxides, J. Phys. Chem. Solids, 74
 L. Huang, S. Chu, J. Wang, F. Kong, L. Luo, Y. Wang, Z. Zou,
Novel visible light driven Mg–Zn–In ternary layered materials
for photocatalytic degradation of methylene blue, Catal. Today,
212 (2013) 81–88.
 G. Crini, Non-conventional low-cost adsorbents for dye
removal: a review, Bioresour. Technol., 97 (2006) 1061–1085.
 D. Ranjbar Vakilabadi, A.H. Hassani, G. Omrani, B. Ramavandi,
Catalytic potential of Cu/Mg/Al-chitosan for ozonation of real
landfill leachate, Process Saf. Environ., 107 (2017) 227–237.
 E. Polhovski, V. Soldatov, Non-exchange sorption of electrolytes
by ion exchangers: II. Sorption of sulfuric acid and lithium
sulfate by Dowex 1 × 8 resin, React. Funct. Polym., 60 (2004)
 A.D. Eaton, M.A.H. Franson, APHA, AWWA, WEF, Standard
Methods for the Examination of Water and Wastewater,
American Public Health Association, D.C., USA, 2005.
 H. Zhang, Z. Wang, C. Liu, Y. Guo, N. Shan, C. Meng, L. Sun,
Removal of COD from landfill leachate by an electro/Fe2+/
peroxydisulfate process, Chem. Eng. J., 250 (2014) 76–82.
 G. Asgari, A. Seid Mohammadi, S.B. Mortazavi, B. Ramavandi,
Investigation on the pyrolysis of cow bone as a catalyst for
ozone aqueous decomposition: kinetic approach, J. Anal. Appl.
Pyrolysis, 99 (2013) 149–154.
 T. Shojaeimehr, F. Rahimpour, M.A. Khadivi, M. Sadeghi, A
modeling study by RSM and artificial neural network (ANN)
on Cu2+ adsorption optimization using light expended clay
aggregate (LECA), J. Ind. Eng. Chem., 20 (2014) 870–880.
 P. Kundu, V. Paul, V. Kumar, I.M. Mishra, Formulation
development, modeling and optimization of emulsification
process using evolving RSM coupled hybrid ANN-GA
framework, Chem. Eng. Res. Des., 104 (2015) 773–790.
 A. Özer, G. Gürbüz, A. Çalimli, B.K. Körbahti, Biosorption of
copper(II) ions on Enteromorpha prolifera: application of response
surface methodology (RSM), Chem. Eng. J., 146 (2009) 377–387.
 A. Murugesan, T. Vidhyadevi, S.S. Kalaivani, K.V.
Thiruvengadaravi, L. Ravikumar, C.D. Anuradha, S. Sivanesan,
Modelling of lead(II) ion adsorption onto poly(thiourea
imine) functionalized chelating resin using response surface
methodology (RSM), J. Water Process Eng., 3 (2014) 132–143.
 S. Heuss-Aßbichler, M. John, D. Klapper, U.W. Bläß, G.
Kochetov, Recovery of copper as zero-valent phase and/or
copper oxide nanoparticles from wastewater by ferritization, J.
Environ. Manage., 181 (2016) 1–7.
 N. Tien Thao, L.T. Kim Huyen, Catalytic oxidation of styrene
over Cu-doped hydrotalcites, Chem. Eng. J., 279 (2015) 840–850.
99D.R. Vakilabadi et al. / Desalination and Water Treatment 80 (2017) 89–99
 B. Ramavandi, A. Rahbar, S. Sahebi, Effective removal of Hg2+
from aqueous solutions and seawater by Malva sylvestris, Desal.
Wat. Treat., 57 (2016) 23814–23826.
 B. Ramavandi, M. Jafarzadeh, S. Sahebi, Removal of phenol
from hyper-saline wastewater using Cu/Mg/Al–chitosan–H2O2
in a fluidized catalytic bed reactor, React. Kinet., Mech. Catal.,
111 (2014) 605–620.
 G. Asgari, B. Ramavandi, S. Farjadfard, Abatement of azo dye
from wastewater using bimetal-chitosan, Sci. World J., 2013
 H. Wang, X. Xiang, F. Li, D.G. Evans, X. Duan, Investigation
of the structure and surface characteristics of Cu–Ni–M(III)
mixed oxides (M = Al, Cr and In) prepared from layered double
hydroxide precursors, Appl. Surf. Sci., 255 (2009) 6945–6952.
 B. Ramavandi, S. Mortazavi, G. Moussavi, A. Khoshgard, M.
Jahangiri, Experimental investigation of the chemical reduction
of nitrate ion in aqueous solution by Mg/Cu bimetallic particles,
React. Kinet., Mech. Catal., 102 (2011) 313–329.
 V. Mahdavi, A. Monajemi, Optimization of operational
conditions for biodiesel production from cottonseed oil on
CaO–MgO/Al2O3 solid base catalysts, J. Taiwan Inst. Chem.
Eng., 45 (2014) 2286–2292.
 D. Cortés-Arriagada, A. Toro-Labbé, Aluminum and iron doped
graphene for adsorption of methylated arsenic pollutants, Appl.
Surf. Sci., 386 (2016) 84–95.
 B. Ramavandi, S.B. Mortazavi, G. Moussavi, B. Ranjbar, S.
Mamisaheby, Experimental investigation of the chemical
reduction of nitrate in water by Mg and Cu/Mg bimetallic
particles in the absence of any pH-control mechanism, Fresenius
Environ. Bull., 20 (2011) 2475–2484.
 A. Ebrahimi, S. Hashemi, S. Akbarzadeh, B. Ramavandi,
Modification of green algae harvested from the Persian Gulf by
L-cysteine for enhancing copper adsorption from wastewater:
experimental data, Chem. Data Collect., 2 (2016) 36–42.
 N. Leitner, H. Fu, pH effects on catalytic ozonation of carboxylic
acids with metal on metal oxides catalysts, Top. Catal., 33 (2005)
 F.J. Beltrán, J. Rivas, P. Álvarez, R. Montero-de-Espinosa,
Kinetics of heterogeneous catalytic ozone decomposition
in water on an activated carbon, Ozone Sci. Eng., 24 (2002)
 R.C. Martins, R.M. Quinta-Ferreira, Catalytic ozonation of
phenolic acids over a Mn–Ce–O catalyst, Appl. Catal., B, 90
 G. Moussavi, R. Khosravi, Preparation and characterization of
a biochar from pistachio hull biomass and its catalytic potential
for ozonation of water recalcitrant contaminants, Bioresour.
Technol., 119 (2012) 66–71.