Environmental Technology

ISSN: 0959-3330 (Print) 1479-487X (Online) Journal homepage: http://www.tandfonline.com/loi/tent20

Evaluation of a magnetic coagulant based on

Fe3O4 nanoparticles and Moringa oleifera extract
on tartrazine removal: Coagulation-adsorption and
kinetics studies

Gustavo Affonso Pisano Mateus, Tássia Rhuna Tonial dos Santos, Isabela
Sperandino Sanches, Marcela Fernandes Silva, Murilo Barbosa de Andrade,
Michele Putti Paludo, Raquel Guttierres Gomes & Rosângela Bergamasco

To cite this article: Gustavo Affonso Pisano Mateus, Tássia Rhuna Tonial dos Santos, Isabela
Sperandino Sanches, Marcela Fernandes Silva, Murilo Barbosa de Andrade, Michele Putti Paludo,
Raquel Guttierres Gomes & Rosângela Bergamasco (2018): Evaluation of a magnetic coagulant
based on Fe3O4 nanoparticles and Moringa�oleifera extract on tartrazine removal: Coagulation-
adsorption and kinetics studies, Environmental Technology, DOI: 10.1080/09593330.2018.1543358

To link to this article: https://doi.org/10.1080/09593330.2018.1543358

Accepted author version posted online: 01

Nov 2018.

Submit your article to this journal

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=tent20
Publisher: Taylor & Francis & Informa UK Limited, trading as Taylor & Francis Group
Journal: Environmental Technology
DOI: 10.1080/09593330.2018.1543358

Evaluation of a magnetic coagulant based on Fe3O4 nanoparticles and

Moringa oleifera extract on tartrazine removal: Coagulation-
adsorption and kinetics studies

Gustavo Affonso Pisano Mateusa, Tássia Rhuna Tonial dos Santosb, Isabela
Sperandino Sanchesb, Marcela Fernandes Silvab, Murilo Barbosa de
Andradeb, Michele Putti Paludob, Raquel Guttierres Gomesc, Rosângela
Bergamascob*
a
Department of Biotechnology, Genetics and Cell Biology, Universidade Estadual de
Maringá, Maringá, Paraná, Brazil
b
Department of Chemical Engineering, Universidade Estadual de Maringá, Maringá,
Paraná, Brazil
c
Department of Food Engineering, Universidade Estadual de Maringá, Maringá,
Paraná, Brazil

* Corresponding author
Rosângela Bergamasco
Postal address: Universidade Estadual de Maringá. Avenida Colombo, 5790. Bloco
D90. CEP: 87020-900. Maringá, Paraná, Brazil.
Phone number: 55+44+30114782.
E-mail address: ro.bergamasco@hotmail.com
 ABSTRACT

The lack of data regarding the mechanisms at work in the coagulation processes of different
substances using magnetic coagulants makes it difficult to understand the phenomena involved
and, consequently, makes it difficult to elucidate the mechanisms involved in the coagulation
process. Thus, the present study aimed at evaluating the performance of a magnetic coagulant
composed of iron oxide (Fe3O4) functionalised with Moringa oleifera (MO) salt extract in the
treatment of a synthetic food industry wastewater simulated by the addition of dye to distilled
water. From the data obtained in the coagulation/flocculation assays followed by magnetic
sedimentation, the different mechanisms involved were evaluated for their fit to pseudo-first
order, pseudo-second order, Langmuir and Freundlich theoretical models. The adjustments to
the models were evaluated from the kinetic data and indicated that at pH 3 the best fit was to the
pseudo-second order model, whereas for pH 6 and 9 the best fit was for the pseudo-first order
model. The isothermal data were adjusted to the Langmuir model, suggesting adsorption of a
monolayer, characterising chemical processes with selective adsorption. In relation to the
mechanisms involved in the process, it is suggested that the neutralisation of charges was the
predominant mechanism in the removal of tartrazine at pH 3, whereas at the other pH values
evaluated the mechanism that prevailed was monolayer adsorption. Thus, the proposed
magnetic coagulant was found to be an efficient alternative material for tartrazine removal,
allowing easy separation in the sedimentation stage while also being compatible with
environmental issues.

Keywords: tartrazine, magnetic nanoparticles, Moringa oleifera, charge neutralisation

adsorption.

1 INTRODUCTION

The food industry is growing, and with this growth comes the increased
generation of waste which requires more attention [1]. The use of chemical additives,
such as dyes, is a common practice in the food industry, although its use in many foods
is justified only by food habits associated with appearance, acceptance and
consequently consumption of the product [1, 2]. The residues produced in such
processes have a diverse composition and may contain toxic and recalcitrant pollutants
resistant to commonly used treatment systems such as
coagulation/flocculation/sedimentation (CFS), adsorption, and biological degradation
[3].
Several impacts can be caused by the release of food dyes to the environment
without due treatment, such as pollution of water resources. These impacts occur
because these compounds hinder the penetration of solar rays into the aquatic
environment, negatively influencing the photosynthetic metabolism of some species
besides possessing carcinogenic potential [4]. In relation to their toxicological
potential, studies have been carried out to verify their harmful effects on living beings,
as these additives may be harmful to health [2], presenting mutagenic and genotoxic
potential [5].
Among the common dyes, tartrazine stands out for its wide use in the food
industry. Tartrazine is a water soluble azo dye that is synthetic, anionic, yellow in color
and which contains two sulphonic groups, one azo (N=N) and one carboxylic
functional group. Tartrazine is widely used in various food products (sweets, cakes,
sodas, alcoholic beverages, chips, chewing gum and jellies), cosmetics and
pharmaceuticals (capsules, pills and gels). This dye is known to cause allergic reactions
and hyperactivity in children, as well as several other side effects such as lethal asthma,
DNA damage, attention deficit disorder and hyperactivity [6].
Usually, most of the industry uses conventional treatments, for example the
application of aluminium salts, to treat the waste generated. However, the disadvantage
to these is that aluminium residue may be left in food and drinking water, and this
element may still cause undesirable damage to the nervous system related to
Alzheimer's disease [7].
To solve these problems, research has sought to develop alternative techniques
for the treatment of water and wastewater, replacing conventional methods with natural
substances [8-10] such as the natural coagulant present in the seeds of Moringa oleifera
(MO) [11]. There are a number of positive impacts related to the use of this seed
among which four stand out: the technological impact as regards the discovery of new
technologies for wastewater treatment, the scientific impact in relation to the
professional accomplishment of works and theoretical–practical studies, the economic
impact in relation to the reduction in production costs due to the possible reuse of water
and finally the environmental impact of reducing the volume of wastewater discarded
in treatment plants.
Another promising technology in water treatment is the use of magnetic
nanoparticles functionalised with the natural coagulants extracted from plant seeds.
Such combinations of treatments are able to reduce the sedimentation time of surface
water, making the process more efficient and requiring less time for accomplishment
[12]. Such magnetic coagulants have shown promising results in the removal of dyes
from real wastewaters, demonstrating that they have the potential for application to the
wastewaters of a variety of industrial segments [13, 14].
However, little data on the mechanisms at work in the coagulation processes of
different substances using magnetic coagulants is available in the literature, which
makes it difficult to understand the phenomena involved, and consequently making it
difficult to elucidate the mechanisms involved in the coagulation process. According to
Mangale Sapana, Chonde Sonal et al. [15], adsorption and load neutralisation
correspond to the mechanisms involved in the coagulation flocculation process when
using MO-based coagulants. Most of the studies done so far elucidate only the
mechanism of load neutralisation through the analysis of the pH or zeta potential [16-
18] throughout the process, and little is discussed about the adsorption mechanism.
With this understanding, research has been carried out using concepts and kinetic
models of adsorption in order to increase the amount of information available about the
mechanisms involved in the coagulation process [19].
In this sense, the objectives of the present study were (I) to obtain a primary
process for high efficiency treatment using a magnetic coagulant capable of treating a
synthetic food industry wastewater, (II) to evaluate the influence of the magnetic field
on the reduction of sedimentation time and (III) to analyse the zeta potential and to
apply kinetic models of adsorption in order to elucidate the mechanisms involved in
this process.

2 Materials and methods

2.1 Synthetic wastewater from the food industry

The tartrazine yellow dye was supplied by an textile industry located in the
northeast part of the state of Santa Catarina, Brazil. The synthetic wastewater was
simulated by the addition of the dye to distilled water resulting in an initial solution with
a concentration of 50±0.36 mg L-1 measured by UV/Vis spectroscopy in a Spectrum DR
5000 at a wavelength of 426 nm, based on a previously constructed calibration curve
with dye concentrations ranging from 1 to 70 mg L-1. After preparation, the synthetic
food industry wastewater was further characterised by its apparent color (DR 5000
HACH), pH (Thermo-Scientific Orion Versastar pHmeter) and zeta potential.

2.2 Synthesis of nanoparticles

 The Fe3O4 nanoparticles were synthesised by the co-precipitation method in
aqueous solution from ferric precursors, based on the methodology proposed by Mateus,
Paludo et al. [20].

2.3 MO saline extract preparation

The MO seeds were kept under refrigeration until use. The seeds were then
peeled and crushed to prepare the saline extract from the mixture of 1.0 g of crushed
seed in 100 mL of saline solution (NaCl 1 M). The extraction of the soluble compounds
which exhibit the coagulant properties of the seeds was carried out in a blender by
turbolisis for 3 minutes, followed by intense stirring (220 RPM) with a magnetic stirrer
for 30 minutes. After stirring, the resulting solution was subjected to filtration using
qualitative filter paper (0,45 µm) [21].

2.4 Characterisation of the material used: magnetic nanoparticles, MO saline extract

and synthetic wastewater from the food industry

Mӧssbauer spectrometry and transmission electron microscopy (TEM) were

performed only for the magnetic nanoparticles to investigate the iron morphology
present in the sample. The functional groups of the magnetic nanoparticles and the MO
saline extract were investigated by Fourier Transform infrared (FT-IR) spectroscopy.
The magnetic nanoparticles and synthetic food industry wastewater were characterised
by scanning electron microscopy (SEM), energy dispersive X-ray spectroscopy (EDX)
and zeta potential.

2.4.1 Mӧssbauer Spectrometry

The Mӧssbauer spectrometry analyses were recorded using a constant

acceleration spectrophotometer with a Co(Rh) source to investigate the chemical forms
of iron present in the sample. The analysis was performed using the NORMOS
commercial adjustment program, which is executed in a DOS environment and applies
the least squares criteria. The equipment was calibrated with metallic iron (α-Fe).
2.4.2 Transmission electron microscopy (TEM)

A transmission electron microscope (TEM JEOL JEM-1400) operating at 120

kV was used. The nanoparticles were dispersed in ethanol by sonication for 5 minutes,
then 5 μL of suspension were deposited in copper grit (KOCH Electron Microscopy
Ltda - FCF200-CU) and then observed under a microscope.

2.4.3 Scanning electron microscopy (SEM) and Energy dispersive X-ray spectroscopy
(EDX)

The micrographs of the tartrazine yellow dye and the magnetic coagulant were
obtained using an electronic microscope Quanta 250 (FEI) coupled with the Spectra
AZtec 3.0 EDS software (Oxford Instruments Nanotechnology Tools Ltd). For X-ray
spectroscopy by energy dispersion, the samples were deposited on a double carbon tape,
fixed on a metal sample holder and covered with gold.

2.4.4 Zeta Potential

The zeta potential of the samples was evaluated as a function of pH, varying
from 2 to 12, to elucidate the possible mechanisms involved in the coagulation process
of synthetic wastewater containing tartrazine yellow dye. The equipment used for the
zeta potential analyses was a Beckman Coulter Delsa (TM) Nano Zeta Potential
apparatus with the Particle Analyzer Delsa (TM) Nano UI 3.73 software. The pH
adjustments were performed using 0.1 and 0.01 M HCl and NaOH solutions.

2.4.5 FT-IR spectroscopy

The FT-IR spectrum was obtained in the wavelength region ranging from 4000
to 400 cm-1 using Avatar 360-FTIR (Thermo Nicolet Instrument) equipment with the
KBr pellet method containing approximately 1% sample.
2.5 Preparation of magnetic coagulant

The functionalisation of the nanoparticles with the MO extract was performed

according to the methodology proposed by Mateus, Paludo et al. [20] in which the
saline extract of MO seeds (20 mL) along with the nanoparticles (20 mg) were sonicated
(Ultronique Q5.9/40A) followed by intense shaking (220 RPM) (GOstirrer MS-H-Pro)
for 1 h at room temperature (25 ± 2 ºC), resulting in the Fe3O4-MO coagulant.

2.6 Evaluation of process efficiency

Synthetic wastewater samples were also characterised after the

coagulation/flocculation and sedimentation tests in the presence and absence of a
magnetic field using the proposed magnetic coagulant in order to evaluate its efficiency.
In this sense, the concentration parameters of tartrazine yellow dye and the apparent
color were analysed. All analyses were performed in triplicate and followed the
methodologies of Standard Methods for Water and Wastewater Examination [22].
The removal efficiency for each analysed parameter was calculated according to
equation 1.

% 𝑅𝑒𝑚𝑜𝑣𝑎𝑙 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 = ((Ci − Cf)/Ci) × 100 (1)

in which Ci and Cf are the initial and final values of each parameter.

2.7 Coagulation/flocculation/sedimentation assays

CFS tests were performed on jar test equipment in beakers containing 0.3 L of
synthetic wastewater with the wastewater pH adjusted to 3, 6 and 9. The operational
conditions of the jar test included rapid mixing at 100 rpm, coagulation time of 2
minutes, slow mixing at 30 rpm and flocculation time of 30 minutes based on the
conditions described by Lee, Choi et al. [23]. The concentration of iron oxide in the
magnetic coagulant was 600 mg L-1, chosen based on preliminary tests of the
concentrations previously tested in wastewaters by dos Santos, Mateus et al. [13].
 Sedimentation was carried out for 10, 20 and 30 minutes in the absence and
presence of a magnetic field of 600 kAm-1, measured by a Gaussmeter. For the tests
with the presence of the magnetic field, a magnet was placed below the containers
containing the synthetic food wastewater. To evaluate the influence of the process,
aliquots of the sample were collected to evaluate the performance of the coagulant in
reducing the apparent color and concentration of the dye.

2.8 Study of the adsorption mechanism

The possible mechanisms of the adsorption of tartrazine yellow dye were

investigated by analysing the data obtained in the jar test, in which the reduction of the
concentration by treatment with the magnetic coagulant based on MO was measured
considering all of the steps in the process (coagulation, flocculation, followed by
magnetic settling). For this, the adsorption capacity (qe) was calculated according to the
following equation:

(𝐶𝑖 − 𝐶𝑒 ) ∗ 𝑉
𝑞𝑒 = (2)
𝑊

where qe is given in mg g-1, Ci and Ce correspond to the initial and final

concentrations of dye, respectively (mg L-1); V corresponds to the wastewater volume
(L) and W corresponds to the dose of the coagulant (g), according Hussain, van
Leeuwen et al. [19], with modifications.
The concentration of the coagulant in terms of the amount of Fe3O4 corresponds
to 0.02 g L-1, and the MO has a protein content of 85,643 mg L-1, according to data
obtained by Baptista, Coldebella et al. [17], both for the jar test in a volume of 2000
mL.
The experimental data were evaluated using two known pseudo-first order and
pseudo-second order kinetic models aiming to express the kinetic characteristics of the
process of adsorption of tartrazine yellow dye in Fe3O4-MO. The kinetic models are
represented by equations (3) and (4).
The nonlinear form of pseudo-first order [24] is described as:
 𝑞𝑡 = − exp(−𝐾1 . 𝑡) . 𝑞𝑒 + 𝑞𝑒 (3)

in which qt and qe represent the amount of dye adsorbed at time t (min) and at
the equilibrium time (mg g-1), and k1 (min-1) is equivalent to the pseudo-first order
adsorption rate constant.
The non-linear form of the pseudo-second order [25] model is described as:

𝑘2 . 𝑞𝑒2
𝑞𝑡 = (4)
1 + 𝑘2 . 𝑞𝑒 . 𝑡

in which qe and qt (mg g-1) refer to the amount of dye adsorbed at equilibrium time and
time t (min), and k2 (g mg-1 min-1) is the pseudo-second order adsorption rate constant.
The applicability of each kinetic model was calculated by the linear regression
method (comparing the coefficients of R2) with the experimental data at different pH
values.
The adsorption mechanisms were investigated by considering the adjustments
of the models to the data obtained in the jar test (removal of dye) using Langmuir [26]
e Freundlich [27].

3 Results and Discussion

3.1 Characterisation of the synthetic food industry wastewater

The characterisation of the synthetic food industry wastewater is presented in
Table 1.

Table 1. Characterisation of the synthetic food industry wastewater.

Parameter Results Unit

Concentration 50,00 ± 0,02 mg L-1
Apparent Color 5176 ± 0,9 mg Pt-Co L-1
pH 6,01 ± 0,03 -
Zeta Potential -22,19 ± 0,11 mV
 Similar concentrations of tartrazine were evaluated in adsorption studies
performed by Gautam, Gautam et al. [28] and GoscianskaPietrzak [29] who concluded
that in concentrations higher than this, significant removals could not be observed.
The SEM micrograph of tartrazine and the elemental composition obtained by
EDX are shown in Figure 1.

Figure 1. SEM micrographs of tartrazine and EDX analysis of synthetic tartrazine dye.

The micrograph of the tartrazine dye indicated the presence of agglomerated

particles with heterogeneous sizes and formats. The EDX results shows that the
approximate amount of each of the elements present is 49.2% C, 26.7% O, 7.1% Na,
6.5% S, 4.6% Au and 4.4 % N; most of the elements detected are common to the
molecular formula of tartrazine (C16H9N4Na3O9S2), and the Au element was present due
to the sample preparation methodology.

3.2 Characterization of nanoparticles

In order to verify the effectiveness of the suggested methodology for the

synthesis of the iron oxide nanoparticles, the Mössbauer spectrometry was performed,
which analyzed the chemical forms of iron present in the sample and the results are
described in Figure 2.

Figure 2. Mӧssbauer spectrometry of iron oxide nanoparticles.

The results indicate that the synthesis carried out by the proposed methodology
leads to the effective formation of Fe3O4, according to the characteristic sextets of the
hyperfine magnetite parameters (Table 2) which were obtained in agreement with data
in the literature [30].

Table 2. Hyperfine parameters of magnetite at room temperature.

Fe sites δ (mm/s) Bhf (T) A (%)

Fe (3+)(tetrahedric) 0,46 48,5 40,8

Fe (2+)(octahedral) 0,56 54,6 54,6
δ: isomeric displacement; Bhf: hyperfine magnetic field, A: sub-spectral relative area

The size and morphology of the Fe3O4 particles were observed in the TEM
micrographs which are shown in Figure 3.

Figure 3. TEM micrograph of the synthesised iron oxide nanoparticles.

The micrographs obtained (Figure 3) indicate that the nanoparticles have an

estimated size of 20 nm with a predominantly spherical shape and are uniformly
distributed as described in a previous study by Mateus, Paludo et al. [20]. The
nanoparticles synthesised by the same method presented similar sizes and morphology,
with superparamagnetic characteristics and a saturation magnetisation of 63 emu g-1.
The elemental composition obtained by EDX and the SEM micrograph of Fe3O4
are shown in Figure 4.

Figure 4. Elemental composition obtained by EDX and SEM micrograph of iron oxide
nanoparticles.

According to the figure, it is possible observe a characteristic agglomeration of

the magnetic nanoparticles, and their composition is shown to be free of the presence of
elements foreign to the iron oxide (Fe3O4), agreeing with the Mӧssbauer spectrum. The
element composition estimated by EDX was approximately 70% Fe and 25% O, with
the presence of C (<5%) attributed to the carbon tape used to couple the sample to the
surface of the sample holder.
The major functional groups and bonds of the Fe3O4 and MO were identified as
shown in Figure 5.

Figure 5. FT-IR spectra of iron oxide nanoparticles and MO.

The FT-IR spectrum of the Fe3O4 indicated the presence of peaks at 3400 cm-1
and 1640 cm-1 corresponding to O-H bonds from H2O adsorbed on the surface of the
iron oxide [31], and characteristic peaks in the 565 cm-1 and 590 cm-1 regions
corresponding to the Fe-O bond present in iron oxide [32]. Also, vibrations between the
582 cm-1 and 634 cm-1 bands are associated with vibrations of the functional groups
Fe2+-O2 and the symmetric deformation of the octahedral sites [33]. These results
corroborate with those indicated by the Mӧssbauer spectrum indicating that the iron
oxide synthesised presents magnetite conformation.
The spectra of the saline extract of MO showed peaks in the region of the 2900
cm-1 (corresponding to the regions of C-H bonds), 1540 cm-1 and 1670 cm-1 bands
characteristic of protein amide I and II. The peak at 1587 cm-1 can be attributed to the
elongation of the C-N bond and the deformation of the N-H bonds present in the seeds
[34]. In addition, the peak at approximately 1100 cm-1 refers to the presence of
polysaccharides (C-O) [18, 35].
As suggested by dos Santos, Silva et al. [36], the association of magnetic
nanoparticles with the compounds present in the MO saline extract results in a material
with the characteristics of both materials used in its synthesis, indicating that there is a
wide variety of functional groups available in the coagulant, increasing its efficiency in
the CFS process. Furthermore, the magnetic coagulant used in the present study was
synthesised using the methodology proposed by Mateus, Paludo et al. [20], in which the
authors suggest that the resulting nanoparticles have an estimated crystallite size of 35
nm, with superparamagnetic characteristics and a saturation magnetisation of 27 emu g-
1
.

3.3 Evaluation of the influence of the magnetic field and sedimentation time

Control tests were performed in the presence and absence of an external

magnetic field using only synthetic wastewater solution (control) and coagulant based
on MO and Fe3O4-MO at the natural pH of the synthetic wastewater solution (pH
6.01). To evaluate the efficiency of the process, aliquots of the supernatant were
collected and analysed for 10, 20 and 30 minutes of sedimentation time; the results are
expressed in Figures 6 and 7.

Figure 6. Removal of tartrazine (%) observed after 10, 20 and 30 minutes of

sedimentation for the control group and other assays performed with MO and Fe3O4-
MO in the presence and absence of a magnetic field.
 The results obtained in relation to the control assay demonstrate that there is no
natural degradation of the dye as a function of time. This fact can be justified due to the
complexity of the molecular structures and functional aromatic rings of azo dyes that
keep them stable in solution and make their removal difficult [37], requiring the
addition of an auxiliary agent [38]. On the other hand, the tests performed using the
MO showed positive results in relation to the reduction of the dye concentration with a
slight increase when exposed to the presence of an external magnetic field. However,
Reck, Paixão et al. [39] demonstrated in their adsorption study that tests carried out at
pH 6 may not be favourable for the removal of tartrazine dye in an aqueous medium,
which justifies the performance of tests with pH variations as will be demonstrated
during the course of the present study.
Finally, for the magnetic coagulant, limited removals in the pH range used at
this stage of the study (pH 6.01) were found because, according to Dos Santos (2018),
the ideal pH range for the removal of synthetic dyes using magnetic coagulants is 3.27.

Figure 7. Apparent color removal (%) observed after 10, 20 and 30 minutes of
sedimentation for the control group and tests performed with MO and Fe3O4-MO in the
presence and absence of a magnetic field.

In relation to the apparent color removal, the tests performed using only the MO
coagulant under the influence of a magnetic field presented considerable results having
a significant increase in the efficiency, especially for the sedimentation time of 10
minutes, which can be explained by the amount of iron and other elements with
metallic characteristics present in the seeds of MO [40]. Also, according to the same
authors, the amount of iron present in 100 g of MO seed is 44.8 mg g -1.
For the tests using Fe3O4-MO, it is observed that the presence of a magnetic
field favoured increased removal in the shortest sedimentation time proposed, and the
results obtained in 10 minutes of magnetic sedimentation (22.5% removal, residual
value of 4011.4 mg Pt-Co L-1) were superior to the results obtained in 30 minutes of
sedimentation without the influence of the magnetic field (20% removal, residual value
of 4140.8 mg Pt-Co L-1).
In their studies dos Santos, Mateus et al. [13], in evaluating the influence of
magnetic nanoparticles (α-Fe2O3) on the removal of apparent color from textile
wastewater, observed that the nanoparticles alone without addition of the protein
extract of MO were able to remove about 35% of the apparent color under the
influence of a magnetic field, suggesting that the nanoparticles help in the process of
CFS when associated with the MO extract. dos Santos, Mateus et al. [13] evaluated the
capacity of γ-Fe2O3 isolated and functionalised with MO extract in the reduction of the
apparent color, turbidity and concentration of compounds, determined using the
absorption of UV254nm in surface water (UV254nm), and concluded that the nanoparticles
alone under an external magnetic field were able to reduce these parameters by 40%
(residual value of 424.4 mg Pt-Co L-1), 50% (residual value of 64 NTU) and 25%
(residual value of 0.480 cm-1), respectively, whereas after functionalization with the
MO extract obtained by an extraction method using organic solvents, the reduction
values were 85.5% for the apparent color (residual value of 102.8 mg Pt-Co L-1), 90.6%
for turbidity (residual value of 12.03 NTU) and 60.6% for UV254nm (residual value of
0.252 cm-1), reinforcing the idea that nanoparticles improve the performance of the
magnetic coagulant.

3.4 pH effect

pH variation tests were performed using the proposed synthetic coagulant in the
presence and absence of a magnetic field in order to determine if pH changes affect the
apparent color and contaminant removal in the synthetic food wastewater. The results
are shown in Figures 8 and 9.

Figure 8. Removal of tartrazine (%) observed after 10, 20 and 30 minutes of

sedimentation at different pH values of 3, 6 and 9 in the presence and absence of a
magnetic field.

Figure 9. Removal of apparent color (%) observed after 10, 20 and 30 minutes of
sedimentation at different pH values of 3, 6 and 9 in the presence and absence of a
magnetic field.

As can be observed, removal results for both concentration and apparent color
were better at pH 3 in the presence of a magnetic field with removals of 70.16%
(residual value of 14.9 mg L-1) and 68.33% (residual value of 1639.3 mg Pt-Co L-1),
respectively, in only 10 minutes of magnetic sedimentation, whereas the results after
30 minutes were minimally higher with values of 71.8% (residual value of 14.1 mg L-1)
for the concentration and 68.85% (residual value of 1612.3 mg Pt-Co L-1) for the
apparent color, therefore indicating that 10 minutes is enough time for the magnetic
sedimentation to occur with high efficiency. These experiments confirm the influence
of pH on the process as already reported in previous studies [6, 29, 39]. The decrease in
the removal efficiency at pH 6 and 9 can be explained by the slightly cationic
characteristic of the coagulant in media with an alkaline tendency [20].
Sánchez-Martín, Beltrán-Heredia et al. [41] carried out coagulation/flocculation
tests using a natural tannin coagulant in the removal of azo dyes, and among the dyes
tested, the authors did not obtain removals of tartrazine yellow dye whose absence of
removal was justified by the high molecular weight and high specificity of the dye in
relation to the coagulant used, which is not observed in relation to the magnetic
coagulant used in the present study. Shirmardi, Mahvi et al. [42] in their studies using
functionalized carbon nanotubes for malachite green dye adsorption observed that the
increase of pH, adsorbent dosage and contact time, increased dye removal,
contradicting what was observed in the present study.

3.4 Elucidation of the mechanisms involved

Coagulation mechanisms are very complex and difficult to model [41], which
makes it difficult to determine the mechanisms involved. The available literature
indicates that the mechanisms involved in the process of coagulation/flocculation in the
presence of a coagulant based on MO are charge neutralisation and adsorption [15].
In order to elucidate the mechanisms involved, zeta potential analyses were
performed to investigate the participation of the load neutralization mechanism in the
coagulation/flocculation process. In addition, kinetic models were applied to determine
the involvement of the adsorption process; the results are presented below.

3.4.1 Charge neutralization

Zeta potential analysis of Fe3O4-MO and synthetic wastewater was performed

in aqueous solutions with pH variations from 2 to 12, and the results are shown in
Figure 10.
Figure 10. Zeta potential analysis of Fe3O4-MO and synthetic wastewater.

As reported in item 3.4, a decrease in the removal efficiency was observed as

the pH increased. This occurs because according to Beltrán-Heredia, Sánchez-Martín et
al. [43] and Beltrán-Heredia, Sánchez-Martín et al. [44] electrical and chemical
interactions between the anima and sulphate groups of the coagulant and the dye,
respectively, tend to be lower in alkaline media. Al-Shabib, Khan et al. [6], when
studying the aggregation of tartrazine by the action of a cationic protein (myoglobin),
concluded that the best results are observed at acidic pH values, leading to greater
aggregation of particles. The same author attributes the possible cause of this
aggregation to electrostatic and hydrophobic interactions, since tartrazine is negatively
charged as a function of the presence of two sulphonated groups, whereas myoglobin is
positively charged due to the protonation of amino acids at acidic pH values; thus, the
negatively charged sulphate groups of tartrazine interact with the positive charges of
the protonated amino acids and consequently neutralise all surface charges, resulting in
the formation of aggregates [6]. The same was observed in the present study since the
coagulant had a positive charge of 16.83 mV at pH 3 while the synthetic wastewater
showed a negative charge of -0.44 mV in the same pH range.
At pH 6, in spite of the positive charge presented by the coagulant (4.17 mV),
the small amount of it added to the solution of synthetic wastewater with a load of -
22.19 mV was not positive enough to neutralise the superficial loads, and there were no
electrostatic interactions with the negative dye, which possibly indicates that there is
another mechanism involved in the removals observed in item 3.4.
For pH 9, both coagulant and synthetic wastewater loads presented negative
values of -6.54 and -18.59 mV, respectively, resulting in an electrostatic repulsion
between the molecules present, again indicating that the removals observed in item 3.4
in this pH range were not the result of neutralisation of charges but that another
mechanism was possibly involved. According to Gautam, Gautam et al. [28], at high
pH values the number of protonated functional groups decreases and more OH are
available to compete (repulsion effect) with the sulphonated anionic groups of the dye
molecules, and as a result, the tartrazine removal capacity decreases with increasing
pH.
 At a pH below the isoelectric point (IEP) of 10–11, MO is positively charged
due to the protonation of arginine, lysine and histidine, amino acids present in the
composition of MO seeds [40]. Non-protonation of these same amino acids was
responsible for the lack of interaction between myoglobin and tartrazine in the study of
Al-Shabib, Khan et al. [6], which was performed at an extremely acidic pH.

3.4.2 Study of the adsorption mechanism

3.4.2.1 Adsorption kinetics

In order to investigate the kinetic characteristics of the adsorption process of the

tartrazine dye in Fe3O4-MO, the experimental data were evaluated using the kinetic
models previously described in the methodology. The kinetic models were applied in
relation to the correlation coefficient, and those with the highest linearity values (R2 >
0.90) were considered to be adequate. The parameters obtained are presented in Table 3,
and the kinetic data adjusted to the described models are shown in Figure 11.

Figure 11. Pseudo-first and second order kinetic models applied to the adsorption of
tartrazine in Fe3O4-MO at pH 3 (A), 6 (B) and 9 (C).

As shown in Figure 11, the highest values of qe observed occur in the initial
stages of the CFS process, and considerably higher removals are not observed after
approximately 40 minutes. This coincides with the removal values presented in item 3.4
in which the highest removals observed occurred within 10 minutes of sedimentation,
which corresponds to approximately 40 minutes in the kinetic study of the
coagulation/flocculation process (32 minutes) and sedimentation on the influence of
magnetic field (10 minutes).
Table 3. Parameters of the kinetic adsorption models of tartrazine in Fe3O4-MO at
different pH values.
Experimental pH 3
pseudo-first order
K1 (min-1) 19.9421
qecal (mg g -1) 73.0959
R2 0.7840
pseudo-second order
K2 (g mg-1 min-1) 0.0333
qecal (mg g -1) 74.2719
R2 0.8583
Experimental pH 6
pseudo-first order
K1 (min-1) 0.0244
qecal (mg g -1) 44.2900
R2 0.9737
pseudo-second order
K2 (g mg-1 min-1) 0.0004
qecal (mg g -1) 55.4542
R2 0.9610
Experimental pH 9
pseudo-first order
K1 (min-1) 0.0218
qecal (mg g -1) 42.6612
R2 0.9116
pseudo-second order
K2 (g mg-1 min-1) 0.3358
qecal (mg g -1) 56.2488
R2 0.8853

According to the data expressed in Figure 11 and Table 3, the pseudo-second

order model shows the best convergence only for pH 3, which suggests that in this pH
range the process occurs in several steps involving external surface sorption and
diffusion in the interior of the adsorbent. Dalvand, Nabizadeh et al. [14] in their
adsorptive studies observed similar results at pH 3 for RB 19 azo dye removal using
Fe3O4 functionalized with L-arginine, the authors also concluded in their kinetic studies
that the data obtained were adequate to the pseudo-second-order model, similar to the
present study.
This model also postulates that there is chemical sorption involving valence
forces during the sharing or exchange of electrons, which is the reason for the lower
values of qe observed in this model at a pH of 6 and 9.
 According to the zeta potential values, at pH 6 and 9, the electrostatic charges of
both the synthetic wastewater and the coagulant are similar, causing repulsion and
preventing the sharing and exchange of electrons [45, 46]. This fact is consistent with
the discussion presented previously in relation to the load neutralisation mechanism,
since the values of R2 for the pseudo-first and second order models at pH 3 are the
lowest of all the correlation coefficients presented in this study, which possibly
indicates that the mechanism that predominates at pH 3 is neutralisation of charges and
not adsorption.
In contrast, the correlation coefficients obtained for pH 6 and 9 were higher for
the pseudo-first order model with values of 0.9737 and 0.9116, respectively, thus
validating its applicability to the description of the kinetic data of adsorption of
tartrazine in Fe3O4-MO. This model assumes that the rate at which the adsorbate is
removed as a function of time is directly proportional to the difference in the saturation
concentration, the number of active sites in the adsorbent [24], and in the case of the
present study, the number of active sites present in the coagulant.

3.4.2.2 Adsorption isotherms

The isothermal models correspond to mathematical approaches that describe the

adsorption mechanisms involved in the interaction between coagulant and adsorbate
[19]. The experimental data were analysed using Langmuir [26] and Freundlich [27]
isothermal models. The concentration of Fe3O4 in the isothermal experiments was 0.02
g L-1, and the concentration of soluble proteins in MO was 85.643 mg L-1.
The nonlinear form of the Langmuir equation can be written as:

𝑞𝑚 . 𝑘𝐿 . 𝐶𝑒
𝑞𝑒 = (5)
(1 + 𝑘𝐿 . 𝐶𝑒 )

where qe (mg g-1) is the amount of tartrazine adsorbed at equilibrium, KL (mg g-

1
) is the equilibrium adsorption constant of Langmuir, Ce is the equilibrium
concentration in the liquid phase (mg L-1) and qm corresponds to the maximum coverage
capacity of the monolayer (mg mg-1).
The nonlinear form of the Freundlich equation can be written as:
 1
𝑞𝑒 = 𝐾𝐹 . 𝐶𝑒 𝑛 (6)

where qe (mg g-1) is the amount of atrazine adsorbed at equilibrium, Ce is the

equilibrium concentration in the liquid phase (mg L-1), KF ((mg g-1) (L mg-1)1/n) and n
are Freundlich constants related to the adsorption capacity and intensity, respectively.
The parameters of the adsorption isotherms were calculated and listed in Table 4
and the adjustment curves for the isothermal models are shown in Figure 12.

Figure 12. Adsorption isotherms of tartrazine in Fe3O4-MO at pH 3 (A), 6 (B) and 9

(C).

Table 4. Langmuir and Freundlich model parameters.

Models
Langmuir pH 3 pH 6 pH 9
qm(mg g-1) 82.9963 71.3604 65.6457
KL 0.1472 0.0592 0.1290
R2 0.7841 0.9007 0.9055
Freundlich
kf 40.6462 19.4660 30.5055
1/n 0.1317 0.2382 0.1431
R2 0.6825 0.8040 0.8238

With the experimental data presented, it is possible to observe that the Langmuir
model presented a better fit (>R2), suggesting homogeneous and monolayer adsorption
characterising a more selective adsorption process [47]. This model also assumes that
the surface of the coagulant has similar energy due to its homogeneity and that the
adsorption sites adsorb similar amounts of the adsorbate, in the case of the present
study, similar amounts of tartrazine [19].
The data in Table 4 allowed us to conclude that the models better fit the data
collected at pH 6 and 9 which suggests that the predominant mechanism is monolayer
adsorption. The data collected for pH 3 presented the lowest correlation in relation to
the kinetic and isothermal models, providing evidence for the idea that the predominant
mechanism in the process of tartrazine removal in this pH range is not adsorption but
charge neutralization.

3.5 Proposed mechanism

 Figure 13 illustrates the possible interactions between the Fe3O4-MO coagulant
and the tartrazine yellow dye whose removal efficiency is influenced by pH variation.

Figure 13. Proposed mechanism of tartrazine removal by the coagulant Fe3O4-MO at

pH values of 3, 6 and 9.

In this study, the adsorption kinetics and isothermal models, as well as the zeta
potential analysis, have suggested that the mechanisms involved in the coagulation of
the tartrazine dye by Fe3O4-MO is pH-dependent: at pH 3 the predominant mechanism
is charge neutralisation, and at higher pH values, it is suggested that the involved
mechanism is different, as evidenced by the low improvement in the quality parameters
evaluated.
At pH 3, charge neutralisation is facilitated by the interaction between the
positive Fe3O4-MO charges and negative charges of the synthetic wastewater as
evidenced by the zeta potential (Fig. 10); the negatively charged groups (SO3Na,
NaOOC, OH e NaO3S) in tartrazine interact with the highly positive charges (16.83
mV) of Fe3O4-MO, neutralising the surface charges and resulting in the formation of
flocs that will be easily attracted by the external magnetic field. Another observation
that shows this behaviour is that the value resulting from the zeta potential after the
treatment remains neutral (0.76 mV).
At pH 6, the tartrazine molecule showed positively and negatively charged
functional groups capable of interacting with the functional groups of the Fe3O4-MO
(O-H, FeOH2+, C-H+, C-N, C=O-, N-H), justifying the small degree of removal
observed in the coagulation process. Furthermore, the presence of oxygen-containing
functional groups, such as the carboxylic acids present in tartrazine, facilitates
adsorption through the H bond regardless of the pH [48].
Finally, at pH 9, according Saleh, Hashem et al. [49], in extremely alkaline
media only the negatively charged functional groups of the tartrazine molecule (COO-
and SO3-) are susceptible to interactions with the components of the magnetic coagulant,
which also undergo ionisation at alkaline pH by removing the available H+ ions in the
coagulant, resulting in an excess of negative charges in the solution of both the
wastewater and the coagulant and possibly causing electrostatic repulsion between the
groups present, which would lead to a decrease in the removal efficiency. In addition,
polar functional groups, such as the hydroxyl and amine groups present in Fe3O4-MO,
exhibit the effect of electron withdrawal at alkaline pH values and can interact with
aromatic rings (π electron receptors), which justifies the small degree of removal
observed [50].

4 Conclusion
The present study aimed to characterize and apply a new magnetic coagulant compound
by iron oxide nanoparticles and functional molecules from MO seed. The
characterization of nanoparticles by TEM, Mossbauer spectroscopy, SEM and FT-IR
allows to confirm the achievement of spherical particles of magnetite phase with 20 nm
of diameter, which could effectively functionalized by the compounds present in MO
saline extract, as presented by FT-IR spectroscopy and zeta potential analysis. This
successfully obtained coagulant was used on the removal of the dye tartrazine from
synthetic wastewater in different pH, and the mechanisms involved in the process were
investigated. From the data obtained in the coagulation/flocculation assays followed by
magnetic sedimentation, the different adsorption mechanisms were evaluated for their
adjustment to the pseudo-first order, pseudo-second order, Langmuir and Freundlich
theoretical models. The results of the CFS tests at pH 3 showed that after 10 minutes of
magnetic sedimentation (40 minutes of adsorption kinetics), the adsorption equilibrium
was sufficient to allow a removal of 70.16% of the tartrazine and 68.33% of the
apparent color, with a maximum adsorption capacity of 82.9963 mg g-1. The kinetic
data presented a better fit to the pseudo-second order model at pH 3 and pseudo-first
order model at pH 6 and 9, making it possible to conclude that the adsorption process of
tartrazine in Fe3O4-MO is pH dependent. The isothermal data were adapted to the
Langmuir model, suggesting an adsorption of a homogeneous and monolayer nature,
characterising chemical processes with a more selective adsorption. In relation to the
mechanisms involved in the process, it is suggested that the neutralisation of charges
was the predominant mechanism in the removal of tartrazine at pH 3, whereas at the
other pH values evaluated, the mechanism that prevailed was monolayer adsorption.
Thus, the proposed magnetic coagulant was shown to be an efficient alternative material
for tartrazine removal, allowing an easy separation in the sedimentation stage and also
being compatible with environmental issues. Thus, this innovative coagulant appears as
an excellent alternative for the treatment of wastewater.
Acknowledgments
The authors thank the financial support of CAPES and Fundação Araucária.

References

[1] Leite BZ, Pawlowsky U. Alternativas de minimização de resíduos em uma

indústria de alimentos da região metropolitana de Curitiba. Eng. Sanit.
Ambiental. 2005; 10: 96-105.

[2] Prado MA, Godoy HT. Corantes artificiais em alimentos. Aliment. and Nutr
2003; 14: 237-250.

[3] Salgado BCB, Nogueira MIC, Rodrigues KA, et al. Descoloração de efluentes
aquosos sintéticos e têxtil contendo corantes índigo e azo via processos Fenton e
foto-assistidos (UV e UV/H2O2). Eng. Sanit. Ambiental. 2009; 14: 1-8.

[4] Peternel I, Koprivanac N, Kusic H. UV-based processes for reactive azo dye
mineralization. Water Res. 2006; 40: 525-532.

[5] Demir E, Kocaoğlu S, Kaya B. Genotoxicity testing of four benzyl derivatives in

the Drosophila wing spot test. Food Chem. Toxicol. 2008; 46: 1034-1041.

[6] Al-Shabib NA, Khan JM, Khan MS, et al. Synthetic food additive dye
“Tartrazine” triggers amorphous aggregation in cationic myoglobin. Int. J. Biol.
Macromol. 2017; 98: 277-286.

[7] Wang B, Wang W, Han H, et al. Nitrogen removal and simultaneous

nitrification and denitrification in a fluidized bed step-feed process. J. Environ.
Sci. 2012; 24: 303-308.

[8] Dalvand A, Gholibegloo E, Ganjali MR, et al. Comparison of Moringa

stenopetala seed extract as a clean coagulant with Alum and Moringa
stenopetala-Alum hybrid coagulant to remove direct dye from Textile
Wastewater. Environ. Sci. Pollut. Res. 2016; 23: 16396-16405.
[9] Gholami-Borujeni F, Mahvi AH, Nasseri S, et al. Enzymatic treatment and
detoxification of acid orange 7 from textile wastewater. Appl. Biochem.
Biotechnol.. 2011; 165: 1274-1284.

[10] Maleki A, Mahvi AH, Ebrahimi R, et al. Study of photochemical and

sonochemical processes efficiency for degradation of dyes in aqueous solution.
Korean J. Chem. Eng. 2010; 27: 1805-1810.

[11] Formentini-Schmitt DM, Alves ÁCD, Veit MT, et al. Ultrafiltration Combined
with Coagulation/Flocculation/Sedimentation Using Moringa oleifera as
Coagulant to Treat Dairy Industry Wastewater. Water Air Soil Pollut. 2013; 224:
1682.

[12] Santos TR, Silva MF, Nishi L, et al. Development of a magnetic coagulant based
on Moringa oleifera seed extract for water treatment. Environ. Sci. and Pollut.
Res. 2016; 23: 7692-7700.

[13] dos Santos TRT, Mateus GAP, Silva MF, et al. Evaluation of Magnetic
Coagulant (α-Fe2O3-MO) and its Reuse in Textile Wastewater Treatment.
Water Air Soil Pollut. 2018; 229: 92.

[14] Dalvand A, Nabizadeh R, Reza Ganjali M, et al. Modeling of Reactive Blue 19

azo dye removal from colored textile wastewater using L-arginine-
functionalized Fe3O4 nanoparticles: Optimization, reusability, kinetic and
equilibrium studies. J. Magn. Magn. Mater. 2016; 404: 179-189.

[15] Mangale Sapana M, Chonde Sonal G, Raut P. Use of Moringa oleifera

(drumstick) seed as natural absorbent and an antimicrobial agent for ground
water treatment. Res. J. Recent Sci. 2012; 3: 31-40.

[16] Ndabigengesere A, Narasiah KS, Talbot BG. Active agents and mechanism of
coagulation of turbid waters using Moringa oleifera. Water Res. 1995; 29: 703-
710.
[17] Baptista ATA, Coldebella PF, Cardines PHF, et al. Coagulation–flocculation
process with ultrafiltered saline extract of Moringa oleifera for the treatment of
surface water. Chem. Eng. J. 2015; 276: 166-173.

[18] Baptista ATA, Silva MO, Gomes RG, et al. Protein fractionation of seeds of
Moringa oleifera lam and its application in superficial water treatment. Sep.
Purif. Technol. 2017; 180: 114-124.

[19] Hussain S, van Leeuwen J, Chow C, et al. Removal of organic contaminants

from river and reservoir waters by three different aluminum-based metal salts:
Coagulation adsorption and kinetics studies. Chem. Eng. J. 2013; 225: 394-405.

[20] Mateus GAP, Paludo MP, dos Santos TRT, et al. Obtaining drinking water using
a magnetic coagulant composed of magnetite nanoparticles functionalized with
Moringa oleifera seed extract. J. Environ. Chem. Eng. 2018; 6: 4084-4092.

[21] Madrona, Serpelloni, Vieira AMS, et al. Study of the Effect of Saline Solution
on the Extraction of the Moringa oleifera Seed's Active Component for Water
Treatment. Water Air Soil Pollut. 2010; 211: 409-415.

[22] APHA, Standard methods for the examination of water and wastewater, in
American Public Health Association (APHA): Washington, DC, USA2005.

[23] Lee J-W, Choi S-P, Thiruvenkatachari R, et al. Evaluation of the performance of
adsorption and coagulation processes for the maximum removal of reactive
dyes. Dyes Pigm. 2006; 69: 196-203.

[24] Lagergren S. Zur theorie der sogenannten absorption gelöster stoffe. Kungliga
Svenska Vetenskapsakademiens Handlingar: PA Norstedt & söner. 1898.

[25] Ho YS, McKay G. Pseudo-second order model for sorption processes. Process
Biochemistry. 1999; 34: 451-465.
[26] Langmuir I. The constitution and fundamental properties of solids and liquids.
Part i. Solids. J. Am. Chem. Soc. 1916; 38: 2221-2295.

[27] Freundlich H. Over the adsorption in solution. J. Phys. Chem. 1906; 57: 385-
470.

[28] Gautam RK, Gautam PK, Banerjee S, et al. Removal of tartrazine by activated
carbon biosorbents of Lantana camara: kinetics, equilibrium modeling and
spectroscopic analysis. J.Environ. Chem. Eng. 2015; 3: 79-88.

[29] Goscianska J, Pietrzak R. Removal of tartrazine from aqueous solution by

carbon nanotubes decorated with silver nanoparticles. Catal. Today. 2015; 249:
259-264.

[30] Roth H-C, Schwaminger SP, Schindler M, et al. Influencing factors in the CO-
precipitation process of superparamagnetic iron oxide nano particles: a model
based study. J. Magn. Mag. Mater. 2015; 377: 81-89.

[31] Maity D, Agrawal DC. Synthesis of iron oxide nanoparticles under oxidizing
environment and their stabilization in aqueous and non-aqueous media. J. Magn.
Mag. Mater. 2007; 308: 46-55.

[32] Aghababaie M, Beheshti M, Razmjou A, et al. Covalent immobilization of

Candida rugosa lipase on a novel functionalized Fe3O4@SiO2 dip-coated
nanocomposite membrane. Food Bioprod. Process. 2016; 100: 351-360.

[33] Zheng L, Su W, Qi Z, et al. First-order metal–insulator transition and infrared

identification of shape-controlled magnetite nanocrystals. Nanotechnology.
2011; 22: 485706.

[34] Araújo CST, Melo EI, Alves VN, et al. Moringa oleifera Lam. seeds as a natural
solid adsorbent for removal of AgI in aqueous solutions. J. Braz. Chem. Soc.
2010; 21: 1727-1732.
[35] Kwaambwa HM, Maikokera R. Infrared and circular dichroism spectroscopic
characterisation of secondary structure components of a water treatment
coagulant protein extracted from Moringa oleifera seeds. Colloids Surf. B. 2008;
64: 118-125.

[36] dos Santos TRT, Silva MF, de Andrade MB, et al. Magnetic coagulant based on
Moringa oleifera seeds extract and super paramagnetic nanoparticles:
optimization of operational conditions and reuse evaluation. Desalin. Water
Treat. 2018; 106: 226–237.

[37] Bonetto LR, Ferrarini F, de Marco C, et al. Removal of methyl violet 2B dye
from aqueous solution using a magnetic composite as an adsorbent. J. Water
Process Eng. 2015; 6: 11-20.

[38] Verma AK, Dash RR, Bhunia P. A review on chemical coagulation/flocculation

technologies for removal of colour from textile wastewaters. J. Environ Manage.
2012; 93: 154-168.

[39] Reck IM, Paixão RM, Bergamasco R, et al. Removal of tartrazine from aqueous
solutions using adsorbents based on activated carbon and Moringa oleifera
seeds. J. Cleaner Prod. 2018; 171: 85-97.

[40] Brilhante RSN, Sales JA, Pereira VS, et al. Research advances on the multiple
uses of Moringa oleifera: A sustainable alternative for socially neglected
population. Asian Pac. J. Trop. Med. 2017; 10: 621-630.

[41] Sánchez-Martín J, Beltrán-Heredia J, Coco-Rivero B, Quebracho colorado: A

New Source for Water Coagulant—Performance on Dye Removal, in The Role
of Colloidal Systems in Environmental Protection2014, Elsevier. p. 347-362.

[42] Shirmardi M, Mahvi AH, Hashemzadeh B, et al. The adsorption of malachite

green (MG) as a cationic dye onto functionalized multi walled carbon
nanotubes. Korean J. Chem. Eng. 2013; 30: 1603-1608.
[43] Beltrán-Heredia J, Sánchez-Martín J, Delgado-Regalado A. Removal of
Carmine Indigo Dye with Moringa oleifera Seed Extract. Ind. Eng. Chem. Res.
2009; 48: 6512-6520.

[44] Beltrán-Heredia J, Sánchez-Martín J, Gómez-Muñoz C. Performance and

characterization of a new tannin-based coagulant. Appl. Water Sci. 2012; 2: 199-
208.

[45] Han X, Wang W, Ma X. Adsorption characteristics of methylene blue onto low

cost biomass material lotus leaf. Chem. Eng. J. 2011; 171: 1-8.

[46] Al-Ghouti M, Khraisheh MAM, Ahmad MNM, et al. Thermodynamic behaviour

and the effect of temperature on the removal of dyes from aqueous solution
using modified diatomite: A kinetic study. J. Colloid Interface Sci. 2005; 287: 6-
13.

[47] Santos TRT, Andrade MB, Silva MF, et al. Development of α and γ-Fe2O3
decorated graphene oxides for glyphosate removal from water. Environ.
Technol. 2017: 1-21.

[48] Fang Q, Chen B, Lin Y, et al. Aromatic and hydrophobic surfaces of wood-
derived biochar enhance perchlorate adsorption via hydrogen bonding to
oxygen-containing organic groups. Environ Sci Technol. 2014; 48: 279-88.

[49] Saleh MM, Hashem EY, Al-Salahi NO. Oxidation and Complexation-Based
Spectrophotometric Methods for Sensitive Determination of Tartrazine E102 in
Some Commercial Food Samples. Computational Chemistry. 2016; 4: 51.

[50] Lonappan L, Rouissi T, Kaur Brar S, et al. An insight into the adsorption of
diclofenac on different biochars: Mechanisms, surface chemistry, and
thermodynamics. Bioresour Technol. 2018; 249: 386-394.