Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

Contents lists available at ScienceDirect

Journal of Photochemistry & Photobiology, A: Chemistry

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

Removal of Rhodamine B dye by adsorption onto an eco-friendly zeolite

and machine learning modeling
M.E.S. Pedebos a, D.M. Druzian a, L.R. Oviedo a, Y.P.M. Ruiz b, A. Galembeck b, G. Pavoski c,
D.C.R. Espinosa c, W.L. da Silva a, *
a
Applied Nanomaterials Research Group (GPNAp), Nanoscience Graduate Program Franciscan University (UFN), Santa Maria, Rio Grande do Sul, Brazil
b
Department of Fundamental Chemistry (DQF), Academic Center of Vitória (CAV), Federal University of Pernambuco (UFPE), Recife, Pernambuco, Brazil
c
Polytechnical School of Chemical Engineering, University of the Sao Paulo (USP), São Paulo, São Paulo, Brazil

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

Keywords: The present work aims to synthesize and characterize a zeolite doped with cobalt (ZO-Co) from such as rice husk,
Clay minerals residual sludge and pumice stone for application in the removal of RhB dye by adsorption. Experimental design
Ecotoxicity and machine learning were used to verify the ideal condition of RhB adsorption, as well, as ecotoxicity tests
ZO-Co nanoparticles
against Lactuca sativa L. seeds were carried out. ZO-Co showed diffraction peaks of the tsaregorodsevite phase,
Textile effluents
groups containing hydroxyl, silicates and aluminosilicates. The ideal condition by machine learning (ML) was
Wastewater treatment
[RhB] of = 130 mg/L, [ZO-Co] = 0.6 g/L, pH ≈ 7, and T = 298.15±2 K. The best experimental fit for the
adsorption data was the isotherm of Khan (equilibrium) and Avrami (kinetic model), where the maximum
adsorption capacity was 259.17 mg g− 1 with 83.66 % of the RhB removal. It is noteworthy that in ecotoxicity
tests ZO-Co showed root radicular growth of 29.74 mm in 7 days. Therefore, cobalt-doped ZO can be effectively
used for the adsorption by organic pollutant RhB correlating topics of environment and nanotechnology.

1. Introduction reproductive and chronic toxicity in both animals and humans. For
example, Sigma Aldrich’s data sheet corresponds to level 3 of toxicity in
The expansion of industrial activities has caused an increase in the aquatic bodies, being harmful in a concentration range of 10–100 mg
emission of contaminated wastewater [1]. Among the industrial sectors around 96 h in contact with water [10]. Moreover, RhB demonstrates
responsible for the emission of organic pollutants, the textile industry applications in ballpoint pens, stamp inks, leather, paper coloring and
stands out due to the high volume of water consumed in the dyeing and stone dyeing [11].
finishing processes of products. For example, 700,000 tons of dyes are Alternatives have been widely applied for RhB removal, as they
produced per year, of which 200,000 tons are generated as waste [2,3]. achieve a greater ability to remove persistent organic pollutants from
The dyes consist of organic pollutants of considerable chemical sta­ alternative materials [12–15], such as membrane separation [16],
bility, high aqueous solubility and low biodegradability, making it ozonation [17], sono-photocatalysis [18,19] and adsorption [20,21].
difficult to remove these substances in wastewater by conventional The adsorption process has advantages due to its versatility, easy
processes, such as physicochemical (coagulation and chemical precipi­ operation and the possibility of using different materials from agro-
tation) and biological (aerobic and anaerobic systems) [4–6]. Moreover, industrial waste (denominated biosorbents) [22]. Thus, activated ma­
the contamination caused by these compounds is very harmful to the terials have been used to increase the adsorption capacity of these ma­
environment due to their high toxicity, carcinogenic and mutagenic terials by physical or chemical activation processes [23]. Chemical
potential, which can cause eye and skin irritation and respiratory activation allows the incorporation of functional groups onto the surface
problems [7,8]. It is worth mentioning that in the environment, organic of the adsorbent, which can increase the interactions with dye molecules
dyes affect the physiological processes and the metabolism of plants, [24]. For example, cobalt chloride (CoCl2) has been standing out as an
leading to an imbalance of loads, causing the inhibition of the plant [9]. activating agent, due to the Co2+ ions providing changes in surface
Among these dyes, Rhodamine B (RhB) has carcinogenicity, promoting greater electrostatic interactions with organic pollutant

* Corresponding author.
E-mail address: w.silva@ufn.edu.br (W.L. da Silva).

https://doi.org/10.1016/j.jphotochem.2023.115404
Received 14 August 2023; Received in revised form 3 December 2023; Accepted 10 December 2023
Available online 11 December 2023
1010-6030/© 2023 Elsevier B.V. All rights reserved.
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

molecules with a high percentage of dye removal (>90 %) [25]. It is 2.3. Zeolite synthesis and chemical activation
worth mentioning that cobalt chloride is a product of the recycling of
metals demonstrated in the recycling of batteries and catalysts, which is ZO was synthesized by the hydrothermal method (Fig. S1a) in a
an environmental passive [26]. sealed round bottom flask connected to a reflux system [41], in which RS
Parallelly, (agro)industrial residue has been used as precursors of (alum sludge) and PS were used as a source of alumina (Al2O3) and silica
adsorbents for the removal of organic pollutants, such as rice husk (RH), (SiO2) and, respectively. Thus, 2.15 g of Al2O3 and 0.65 g of SiO2 were
residual sludge (RS) from water supply treatment plants (WSTPs) and mixed in a NaOH solution (2 mol/L, Synth®) under magnetic stirring
ash from thermoelectric plants, due to their high availability and (120 rpm for 20 min). The solution was placed in a reflux system heated
chemical composition rich in SiO2 and Al2O3 [27–29]. Pumice stone (PS) at 10 K min− 1 for 3 h. After, it was cooled (298.15±2 K for 2 h) and
is an aggregate of metallic oxides (mainly silica, alumina, calcium and filtered (ϕ = 0.45 µm with distilled water and alcohol until pH ≈ 7.5).
titanium) that is formed during the rapid cooling/solidification of vol­ For chemical activation (Fig. S1b), the impregnation/activation method
canic matter and can be used in wastewater treatment by advanced was used [42], where 0.003 mol/L of CoCl2 was mixed with 2 g of ZO in
processes, such as adsorption and heterogeneous photocatalysis, 100 mL of distilled water [43] under magnetic stirring (120 rpm for 20
allowing the formation of porous materials with high adsorption ca­ min). Subsequently, the sample was dried (338.15±2 K for 12 h) and
pacity and photoactivity [30]. calcined (623.15±2 K for 2 h with 303.15 K min− 1). The final sample
Porous materials have been used as adsorbents for the removal of was named zeolite activated with cobalt chloride (ZO-Co).
organic pollutants, such as bentonite, montmorillonite, and zeolite (ZO),
due to their high removal capacity, easy separation and the possibility of 2.4. Characterization techniques
reuse [31,32]. It is noteworthy that the class of materials such as zeolite
is denoted by the homogeneity of their structure, porosity, and use of To identify the crystalline phases, X-ray diffraction (XRD) was used
waste as raw materials allowing greater control for the adsorption in a Bruker diffractometer (model D2 Advance) with copper tube (λ =
process [33]. 1.5418 Å) ranging from 5 − 70◦ with an acceleration voltage of 30 kV
Despite the good performance of zeolites in the adsorption of organic and applied current of 30 mA. The crystalline structures were analyzed
dyes, experimental data can be time and cost-consuming. Computational using the database COD (Crystallography Open Database) calculating
methods have been used to develop predictive models such as machine the average crystallite size (dc) (Eq. (1) and interplanar distance (d) (Eq.
learning (ML) [34]. For example, Artificial Neural Networks (ANN) (2) [44,45].
[35], Xtreme Gradient Boosting (XGB) [36] and Random Forest (RF)
[37] have been attracting researchers around the world. These models dc =
0.94*λ
(1)
are based on algorithms that recognize and understand patterns in the β*cos(θ)
dataset, performing predictions and using optimization methods to
λ
prevent overfitting [38]. Then, it can be used alternatively or comple­ d= (2)
mentary to the experimental design, such as central composite rotatable 2*sen(θ)
design (CCRD), Central Composite Designs (CCD) and factorial plan­
Where: β is the total width at half maximum (FWHM) and θ (◦ ) is the
ning. Furthermore, the high precision of the ML predictive regressor
Bragg diffraction angle.
models in the adsorption of organic dyes onto nanoadsorbents is re­
The identification of the functional groups was obtained by the
ported in the literature [39].
Fourier Transform Infrared Spectrum (FTIR) in a Perkin Elmer Spectrum
In this context, the present work aims to synthesize and characterize
Frontier with ATR (attenuated total reflectance) accessory in trans­
a zeolite doped with cobalt (ZO-Co) from RH, RS and PS for application
mittance mode ranging from 4000 and 500 cm− 1 (with 45 scans and 4 %
in the RhB removal by adsorption (equilibrium, kinetic and thermody­
resolution). Scanning electron microscopy (SEM) images were acquired
namic study). Moreover, to apply ML models as a tool for predicting the
by the equipment Phenom Pro X microscope (Thermo Fisher Scientific,
experimental of the adsorption data. The novelty of this work is the use
United States) with a dispersive energy analyzer attached, allowing
of the residues to produce the ZO (high silica composition) correlating
identification of the elements present. The operating conditions for
the experimental tests of the experimental design with machine learning
microscopy were BSD FULL and operated at 15 kV. For these analyses,
to determine the best operational condition by adsorption process,
the samples were placed in stubs of the equipment under a carbon rib­
reducing costs and time. Furthermore, it seeks to meet the sustainable
bon. Field Emission Gun Scanning Electron Microscopy (FEG-SEM) was
development goals, specifically goal 6 (ensure access to water and
used to determine the morphological characteristic in a MIRA3 (TES­
sanitation) improves water quality by reducing pollution, eliminating
CAN, Czech Republic) with 15 kV acceleration with 7500x magnifica­
dumping, and minimizing the release of hazardous chemicals.
tion. To determine the textural properties (e.g., surface area, pore
diameter and pore volume), adsorption/desorption of N2 with relative
2. Materials and methods
pressure ranging from 0 to 1 using the Brunauer-Emmett-Teller (BET)
and Barrett Joyner-Halenda (BJH) methods [46] in a Micromeritics®
2.1. Materials
equipment, model ASAP 2020, according to Eq. (3).
( )
For the ZO synthesis were used the RS (from the WSTPs, São Gabriel, p
c− 1 p 1
Brazil), PS commercial (exfoliating Pumice Stone) and ethyl alcohol
[ p0( ) ] = . + (3)
p vm .c p0 vm .c
(C2H6O – Synth®). For the chemical activation, cobalt(II) chloride v. 1 − p0
hexahydrate (CoCl2⋅6H2O, ≥97 %, Sigma-Aldrich®) was used. RhB dye
(C28H31ClN2O3, ≥95 %, Sigma-Aldrich®) was used as a target organic Where: p and p0 are the equilibrium and the saturation pressure of ad­
pollutant. sorbates at the temperature of adsorption, respectively; v is the adsorbed
gas quantity; vm is the monolayer adsorbed gas quantity and c is the BET
2.2. Pre-treatment of precursors constant.
The surface charge was obtained by zeta potential (ZP) in a NANO­
RS was pre-treated by the calcination process (673.15 K for 2 h) with BROOK OMNI) using the Electrophoretic Light Scattering method
a heating rate of 278.15 K min− 1 [40]. Initially, PS was ground in a ball (PALS) and an electrode (BI-SREL or BI-ZEL) under a scattering angle of
mill (SERVITECH brand, model CT-241) for 25 min at room temperature 15◦ at room temperature (298.15 K). Dynamic light scattering (DLS)
(298.15±2 K). measurements were carried out to determine the hydrodynamic

2
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

Table 1 Table 2
Coded units used for the variable parameters for the statistical analysis by CCRD Kinetic, equilibrium and thermodynamic equations.
to (a) adsorption tests and (b) ecotoxicity tests. Models Parameters Eqs
(a) − 1
qt = q1 *(1 − exp( − k1 *t) ) k1 (min ): rate constant of pseudo first- (6)
Order [RhB] (mg L− 1) [ZO-Co] (g L− 1) pH order;
q1 (mg g− 1): theoretical value of adsorption
− 1.68 16 0.13 2 capacity;
− 1 50 0.3 4 k2 (g mg− 1 min− 1): rate constant of pseudo
0 100 0.55 7 second-order;
+1 150 0.8 10 t q2 (mg g− 1): theoretical value of adsorption (7)
184 1 12 qt = ( )
+1.68 1 t capacity;
+
k2 *q2
2 q2
(b) kd (mg g− 1 min− 0,5): intraparticle diffusion
Order − 1
[ZO-Co] (g L ) pH rate;
− 1.41 0.196 2 t (min): agitation time (min);
− 1 0.3 4 C (mg g− 1): constant related to resistance to
0 0.55 7 diffusion;
+1 0.8 10 qt = Kdif *t0.5 + C kAv: (min− 1): Avrami’s kinetic constant; (8)
+1.41 0.904 11.2 qt = qe + (9)
[1 − exp( − kAv .tnAv ) ]
nAv:: Avrami exponent.
diameter of the ZO and ZO-Co using NanoBrook Omni equipment qmax *KL *Ce qe (mg. g− 1): mass of adsorbed solute per mass (10)
qe =
1 + KL *Ce of adsorbent;
equipped with a red laser diode (35 mV and λ = 640 nm) at a detection qmax (mg g− 1): maximum adsorption capacity
angle of 90◦ . For the pHZCP tests, the method of 11 tests was used [47] of the material;
ranging from 2 to 12, being the point of zero charge defined as the pH at 1 (11)
which the surface of the solid has a neutral charge. Furthermore, ther­ qe = KF *Cen
mal properties were investigated by thermogravimetric analysis (TGA) KL (L mg− 1): Langmuir constant;
Ce (mg L− 1): equilibrium concentration of the
and Derived Thermogravimetry (DTG) in a thermogravimetric analyzer
solute;
(Shimadzu, model TGA-60/60H) using a platinum sample port and air qmax *(KS *Ce )m (12)
atmosphere with a gas flow of 100 mL min− 1, over a temperature range qe =
1 + (KS *Ce )m
of 298.15 – 1273.15 K with a heating rate at 20 K min− 1. KF (mg g− 1) (L mg− 1)1/n: Freundlich
adsorption capacity constant;
n: constant related to surface heterogeneity;
2.5. Adsorption test KS (mg L− 1): Sips constant;
m: fractional exponent related to the
Adsorption tests were carried out with RhB dye (synthetic solution) adsorption mechanism;
in batch mode, according to the literature [48]. Thus, samples were KRP *Ce KRP (L g− 1): Redlich-Peterson constant; (13)
qe =
1 + aRP *Cbe RP
collected at predetermined times (0, 5, 15, 30, 45, 60, 75, 90, 120, 150 − 1
aRP (L mg ): Redlich-Peterson constant;
and 180 min), filtrated (33 mm diameter sterile syringe filter with a bRP: Exponent of the Redlich-Peterson model;
0.45 µm pore size hydrophilic polyethersulfone membrane) and the QK *BK *Ce (14)
qe =
percentage of RhB dye removal was determined (R, %) (Eq. (4) and (1 + BK *Ce )AK
adsorption capacity (qt, mg g− 1) (Eq. (5) using a UV–vis spectropho­ QK (g mg− 1): Khan model maximum
adsorption capacity;
tometer (Varian Cary 100) at λ = 553 nm.
BK (L mg− 1): adsorption parameter for the
C0 − Ce Khan model;
R% = *100 (4) b*Ce (15)
C0 qe = qmax * 1/a
[1 + (b*Ce )a ]
AK: specific adsorbent-adsorbate parameter;
(C0 − Ci )*V
qi = (5) b: Toth’s isotherm constant;
m α: Toth’s heterogeneity parameter.
ΔG = − R* T* ln (ρ*KL ) ρ (g L–1): solution density; (16)
Where: C0 is the initial concentration of the RhB (mg L− 1) (t = 0 min), Ci KL (L mg− 1): Langmuir constant;
is the concentration of the RhB (mg L− 1) (t = t), V is the volume of the ln(ρ*KL ) =
ΔS◦ ΔH◦
−
(17)
solution (L) and m is the mass of the adsorbent (g). R R* T
T (K): temperature (K);
R = 8.314 J mol− 1 K− 1
2.6. Phytotoxicity assays

Lactuca sativa L. seeds were used for the phytotoxicity assay [49]. pH, [ZO-Co] and [RhB], and as response variable was maximum
The samples were sown with germinating paper (Germitest®) with ZO- adsorption capacity (qmax) (Table 1a). Regarding the phytotoxicity
Co (0.196 – 0.904 g/L) and pH of the solution (2 – 11.2) for 168 h at assay, the independent variables were [ZO-Co] and pH of the solution
room temperature (298.15±2 K) and controlled humidity (around of 70 and the dependent variable was seed root growth (Rg) (Table 1b).
%). The samples were analyzed according to root radicular with a digital
caliper (Mitutoyo®) and compared with the negative control (NC – 2.8. Adsorption kinetics, equilibrium and thermodynamic studies
distilled water and seeds).
Adsorption kinetic study was used to evaluate the efficiency of the
2.7. Experimental design liquid phase process determining the residence time using pseudo first-
order (PFO) (Eq. 6), pseudo second-order (PSO) (Eq. 7), intraparticle
To determine the ideal condition of the RhB removal by adsorption diffusion (IDP) (Eq. 8) and Avrami (Eq. 9) [50] using the ideal condition
process and phytotoxicity assays, a central composite rotatable design by CCRD 23. Equilibrium adsorption was used to describe the number of
(CCRD) was used through the response surface and of the Pareto graphic adsorbate molecules per unit mass of adsorbent using the isotherms of
(Statistic10®). Independent variables of the adsorption process were the Langmuir (Eq. 10), Freundlich (Eq. 11), Sips (Eq. 12), Redlich-

3
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

Table 3
Machine learning model configurations.
Model Configuration Equation Reference

Random Forest Maximum depth trees: 1–50; Considering 2-splits: [54]

(RF) Number of decision trees: 1–50; ni,j = wj • bj + wj,left • bj,left − wj,right • bj,right (18)
Evaluated by node importance (with k-splits) Where: ni,j = the importance of node j; wj = weighted number of samples reaching node j;
(Eq.17). bj,left = the impurity value of node j; bj,right = the impurity value of node j; wj,right =
weighted number of samples reaching node j; wj,left = weighted number of samples
reaching node j; a node equals a decision tree.
∑
Artificial Neural Neural Network framework: 3:n:1, where n is the φ = wi • xi + bk (19)
SPS:refid::
Network number of nodes in the hidden layer;
(ANN) Activation function: summation/transfer function 1 bib55[55]
f(φ) = (20)
(Eq. 18) associated to logistic (Eq. 19) and ReLu 1 − e− φ

function (Eq. 20);

Weight updating function: Adam* and lbfgs**. f(φ) = (0, φmax )(21)
Where: φ = transfer function; xi = i-th node of the hidden layer; wi = weight of the i-th
neuron, wi, bk = bias (error associated to the algorithm), φmax = maximum value
predicted from ANN using ReLu activation function.
( ∑n )2
Xtreme Gradient Maximum depth trees: 1–50; Residualsi
SimilarityScore = ∑n 1 [56]
Boosting (XGB) Sampling: bootstrapping; (22)
1 [Pi • (1 − Pi )] + λ
Regression/patter recognition evaluated by Residualsi = Yobs − Ypred (23)
Similarity Score (Eq.21–22). Where: Pi = previous probability of the outcome predicted by the XGB model of the i-th
node (decision tree) is true; λ = a regularization parameter***; Yobs = observed/
experimental value for the response variable; Ypred = predicted value for the response
variable.

cv = 7 for all algorithms. | * Adaptive Movement Estimation algorithm. | ** Limited-memory Broyden–Fletcher–Goldfarb–Shanno algorithm. | *** Increasing lambda
disproportionately reduces the influence of small leaves (the ones with few observations) while having only a minor impact on larger leaves (the ones with many
observations).

Peterson (Eq. 13), Khan (Eq. 14) and Toth (Eq. 15) [51] by CCRD 22. The adsorption of RhB by the ideal condition of the CCRD were performed, in
thermodynamic study was carried out to evaluate the difference be­ three stages; (a) adsorption test; (b) separation (filtration – ϕ = 12.5 cm)
tween physisorption and chemisorption according to the magnitude of and (c) the adsorbent was reinserted in a new adsorption test.
the Gibbs free energy variation (ΔG◦ ) (Eq. 16), enthalpy variation (ΔH◦ )
and entropy variation (ΔS◦ ) (Eq. 17) [52]. Table 2 shows the kinetics 2.11. Statistical analysis
models, equilibrium isotherms and thermodynamic equation.
All experiments were conducted in triplicates as independent ex­
2.9. Machine learning periments the results are presented as arithmetic mean and standard
deviation. Furthermore, statistical comparisons were performed from
Machine learning algorithms are generally black-box models used to analysis of variance (ANOVA), also, post hoc evaluations of differences
develop predictive models. Moreover, to guarantee the accuracy of the between with p < 0.05*, p < 0.01**, p < 0.001*** and p < 0.0001****
developed model, a cross validation method was used, labeled k-fold using GraphPad Prism 7.0® software to the ecotoxicology tests. To
cross-validation (cv), where cv employed are generally 3, 5 and 7. Also, verify the prediction of the isotherms and kinetic models tested, the
cross-validation can be considered a suitable tool to prevent overfitting following metric error functions (Eqs. (26)–(29) were used [58] using
[53]. In this work, cv = 5 was used. Table 3 shows all descriptions of the the software Statistic® 10.0.
machine learning algorithms used in the computational study. [( )
1 − R2 (n − p)
]
The performance of previous models was evaluated from Eqs. (24) R2adj = 1 − (26)
(n − p − 1)
and (25), in which values greater than 0.90 for R2 and lower than 5 for
RMSE were used as a criterion of model selection [57].
1∑n
( )2
∑N ( )2 SSE = qexp − qpred (27)
yi,exp − yi,pred n 1
R2 = 1 − ∑Ni=1( )2 (24)
i=1 yi,exp − ̂
y i,pred [ ]
1∑n
qexp − qpred
√̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅ ARE = (28)
√N ( n 1 qexp
√∑ )2
√
√i=1 yi,exp − ̂y i,pred √̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅̅
RMSE = 1 − (25) √
√ 1 ∑ n
[ ]2
N qexp − qpred
MPSD = 100√ (29)
n− p 1 qexp
Where: R2 is the determination coefficient; RMSE is the root mean
squared error (mg g− 1); yi,exp and ŷi,pred are the actual and the predicted
Where: qexp, qpred is the experimental and the predicted adsorption ca­
value of the response (qmax); N is the data size.
pacity (mg g− 1), qpred is the mean of predicted values (mg g− 1), n is the
In parallel, a feature importance study of the parameters with high
number of data, p is the number of model parameters, R2 is the coeffi­
contribution to the generalized model was performed. Thus, the per­
cient of determination (%), R2adj is the adjusted coefficient of determi­
mutation score value ranges from 0 to infinity. The greater the permu­
nation (%), SSE is the sum of squared errors, ARE is the mean relative
tation score higher the contribution/influence of that parameter on the
error (%), and MPSD is the Marquardt percent standard deviation.
response variable qe.

2.10. Zo-co desorption

To investigate the effects of adsorption in different cycles, 4 cycles of

4
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

Fig. 1. Diffractograms of raw materials and samples (ZO and ZO-Co).

Fig. 2. ATR-FTIR spectra of the ZO and ZO-Co.

3. Results and discussion (kaolinite, quartz low, berlinite, and calcite) indicating a potential ma­
terial to produce zeolites. The ZO and ZO-Co presented phases of a
3.1. Characterization of the ZO and ZO-Co tsaregorodsevite zeolite constituted of aluminum, silicon, and titanium.
Moreover, ZO-Co presented a larger dc (57.77 nm) compared to ZO
Fig. 1 and Table S2 showed the diffractograms of the raw materials (54.41 nm) suggesting the entry of cobalt oxide in the crystalline
used and the materials synthesized. structure of the zeolite confirming the synthesis of ZO and ZO-Co. It is
According to Fig. 1 and Table S2, the raw materials (RS and PS) noteworthy that the quart low, kaolinite, berlinite, and calcite crystal­
presented phases consisting of aluminum, silicon, and calcium line phases were not identified in ZO and ZO-Co possibly due to the

5
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

Table 4 adsorbed layer of N2 indicating the magnitude of the adsorbent/adsor­

Surface area (SBET), pore volume (Vp), pore diameter (Dp), BET constant (c) and bate interactions [74]. Thus, for ZO and ZO-Co they showed a BET
zeta potential (ZP) of the ZO and ZO-Co. constant of 287.77 and 261.29, respectively, suggesting that both sur­
Samples SBET (m2/g) Vp (cm3 g− 1) Dp (nm) c ZP (mV) faces demonstrated a strong interaction in the adsorption of N2, being
RS —— —— —— —— − 5.66 ± 0.2
indicative of other types of organic pollutants. It is noteworthy that
PS —— —— —— —— 12.34±0.4 textural properties such as porosity and surface area are fundamental for
ZO 60±0.06 0.123±0.002 11.2±0.2 287.77 − 5.57 ± 0.2 adsorption applications involving RhB removal due to the increase in
ZO-Co 53±0.04 0.121±0.003 11.1±0.1 261.29 − 8.69 ± 0.3 interactions between active sites on ZO-Co and RhB molecules for the
removal process [75].
ZO-Co showed a higher surface charge by ZP of − 8.69 ± 0.3 mV
coalescence of aluminum, silicon, calcium and phosphorus ions pro­
(Table 4) compared to the ZO (− 5.57±0.2 mV) due to the interactions of
vided by the hydrothermal method (temperature and pressure) [59,60].
the Si4+ and Al3+ ions with the oxygen caused by the spherical imped­
Fig. 2 shows the ATR-FTIR spectra of ZO and ZO-Co.
iments of the atoms [76]. It is noteworthy that a negative ZP indicates a
According to Fig. 2, the ATR-FTIR spectra showed the following
good of electrostatic interactions with RhB molecules (cationic) result­
stretching (a) at 2311 cm− 1: CO2 band of the background (asymmetric
ing in a greater percentage of removal [77,78].
stretch) [61,62]; (b) at 1697.2 cm− 1: O-H vibration of physical adsorbed
Fig. 3 shows the effect of the pHZCP where for the ZO the pHZCP ≈
water and structural water [63]; (c) at 1517.7 cm− 1: correspond to the
8.06 and to the ZO-Co was 7.54, indicating that when pH < pHZCP the
carboxylic (C = O) group’s stretching vibration, that may also originate
surface of the adsorbent is protonated (positive surface liquid charge),
from the RS or PS [64]; (d) at 1366.7 cm− 1: vibration of the benzene ring
while pH > pHZCP the surface is deprotonated (negative surface liquid
and partial vibration of C–H in methyl group [65]; (e) at 985.3 cm− 1: Si-
charge). It is noteworthy that due to the cationic nature of the dye (RhB),
O-Al vibrations [66]; (f) at 722.15 cm− 1: O-Si-O bending vibration of the
a pH around > 8 will favor the removal of dye molecules, due to the
crystal lattice [67]; and (g) at 634.4 cm− 1: –CH of the external vibration
increase in electrostatic interactions between the surface of ZO and ZO-
of the double rings of the ZO structure [68]. However, ZO-Co showed a
Co with RhB [79]. The decrease in pHZCP of ZO-Co is possibly due to the
Co(II)-O functional group at 846.5 cm− 1 indicating the formation of the
entry of cobalt oxide into the crystalline structure leading to increased
Co3O4 phase [69] according to claimed in the XRD analysis. It was
electrostatic interactions such as protonation of Co2+ ions at alkaline pH
possible to confirm the synthesis of ZO and ZO-Co containing functional
[80,81].
groups consisting of silicates, aluminates and cobalt. Fig. S3 shows the
Fig. S4 shows the hydrodynamic particle size distribution (RH) and
adsorption/desorption isotherms and textural properties for ZO and ZO-
polydispersity (PD) by DLS. ZO and ZO-Co demonstrated a unique modal
Co and Table 4.
size distribution with characteristics RH = 1076.96±13.79 nm and
In relation to Fig. S3 and Table 4, ZO and ZO-Co showed type IV
620.71±23.48 nm suggesting good homogeneity of the solution (PD =
adsorption/desorption isotherms with H3 hysteresis curves (classified in
0.34±0.01 and 0.30±0.02, respectively). The difference in RH between
wedge-shaped pores, cones or parallel plates) [70]. ZO and ZO-Co
the ZO may be due to the cobalt ions preventing the coalescence of
demonstrated mesoporous characteristics with SBET = 60±0.06 m2/g
atoms in the calcination process demonstrating a smaller RH, where the
and 53±0.04 m2/g, Dp = 11.2±0.2 nm and 11.1±0.1 nm, and Vp =
particle size distribution was consistent with results found in the liter­
0.123±0.002 cm3 g− 1 and 0.121±0.001 cm3 g− 1 similar to zeolite
ature [82]. It is noteworthy that the size of the nanoadsorbents corre­
structures [71,72]. The reduction of SBET may be due to the insertion of
sponds to the hydrodynamic size, which is over estimated due to the
Co2+ ions with the surface energy obstructing crystalline structure [73].
strong electrostatic interaction between the solvent and the ZO and ZO-
The BET constant (c) is related to the adsorption energy in the first

Fig. 3. PhZCP of the ZO and ZO-Co.

6
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

Fig. 4. (a) SEM micrography of the ZO with 4500x magnification, (b) EDX results of the ZO, (c) SEM micrography of the ZO-Co with 7700x magnification, (d) EDX
results of the ZO-Co, (e) FEG-SEM micrography of the ZO with 12.5kx magnification, (f) FEG-SEM micrography of the ZO-Co with 12.5kx magnification, (g) FEG-SEM
micrography of the ZO with 31.3kx magnification, (h) FEG-SEM micrography of the ZO-Co with 31.3kx magnification, (i) FEG-SEM micrography of the ZO with
62.5kx magnification, (j) FEG-SEM micrography of the ZO-Co with 62.5kx magnification and (k) average particle size of ZO and ZO-Co.

Co constituents generally oxides [83]. The stability of ZO oxides can be adsorption process [88]. Fig. 4(b and d) show the elemental composition
explained by the tendency of atoms to relax reducing surface energy and of ZO and of ZO-Co which demonstrate a Si/Al ratio of 7.22 and 1.17,
generating small agglomerates in solution [84]. Fig. 4 shows the SEM- respectively, classified as high and low zeolite [89]. The ZO and ZO-Co
EDX and FEG-SEM micrographs of the ZO and ZO-Co. denoted elements (Si, Al, C, O, Ca, Na), where differences found in the
According to Fig. 4(a, c, e, f, g, h, i, j and k), ZO and ZO-Co entry of cobalt ions into the structure demonstrating success in the hy­
demonstrated a heterogeneous structure with large aggregated parti­ drothermal method.
cles (semi-cubic shape) and particle diameter around 45 nm and 42 nm TGA thermograms for ZO and ZO-Co shown in Fig. S5 were used to
similar to those found in the literature [85,86]. ZO and ZO-Co contain determine the thermal stability of the adsorbent. ZO and ZO-Co
many holes of <200 nm, where they are considered beneficial to in­ demonstrated mass loss of 9.78 % and 8.63 %, respectively in the tem­
crease the water flow leading to a better result in the adsorption process perature range of ~294.76 K to ~500.35 K attributed to water loss from
[87]. It is noteworthy that the morphology of the semi-cubic particle hydration complexes [90]. The % mass loss beyond ~500.35 K was
facilitates the ionic exchange process increasing the kinetic speed of the attributed to ZO dihydroxylation [91]. Furthermore, the difference in

7
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

Fig. 4. (continued).

8
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

Table 5
Coded matrix of the CCRD 23 for RhB adsorption.
[RhB] (mg L− 1) [ZO-Co] (g L− 1) pH qmax (mg g− 1)

150 (1) 0.3 (− 1) 10 (1) 74.46

16 (− 1.68) 0.55 (0) 7 (0) 19.07
150 (1) 0.8 (1) 10 (1) 119
100 (0) 1 (1.68) 7 (0) 46.80
150 (1) 0.8 (1) 4 (− 1) 8.88
100 (0) 0.55 (0) 7 (0) 45.51
50 (− 1) 0.8 (1) 4 (− 1) 8.12
100 (0) 0.55 (0) 2 (− 1.68) 5.05
50 (− 1) 0.8 (1) 10 (1) 42.31
184 (+1.68) 0.55 (0) 7 (0) 47.89
50 (− 1) 0.3 (− 1) 4 (− 1) 12.19
100 (0) 0.55 (0) 7 (0) 27.5
50 (− 1) 0.3 (− 1) 10 (1) 132.56
100 (0) 0.13 (− 1.68) 7 (0) 58.57
100 (0) 0.55 (0) 12 (+1.68) 48.30
150 (1) 0.3 (− 1) 4 (− 1) 48.87
100 (0) 0.55 (0) 7 (0) 31.65

Table 6
Fig. 5. ANN performance: correlation between number of nodes of the hidden
ML model performance.
and RSME.
Model R2train RMSEtrain R2test RMSEtest Best model configuration

XGB 0.8515 2.6242 0.6371 3.0139 Number of decision trees: RhB adsorption model using ZO-Co was not predicted/significant due to
50 the Fcalculated (1.6) being smaller than the Ftabeled (4.02) adjusting 41 %
max. tree depth: 15 m
of the experimental values to the model. Thus, to increase the coefficient
of determination and find an improved ideal condition different models
ANN 0.9139 7.6113 0.9025 8.050 Number of epochs: 80,000 of machine learning models were used.
Activation function: ReLu
Neural network: 3–15-1
weight updating function:
lbfgs 3.3. Machine learning

RF 0.8514 2.6242 0.6371 3.0139 Booster method: dart A machine learning study was performed for comparative purposes
max. tree depth: 15 m with CCRD. Thus, more than one parameter in each model (number of
decision trees, max tree depth, number of hidden layers and activation
function) was varied. Table 6 shows the performance of each model
mass loss may be due to the ratios of the ions (Si4+, Al3+ and Co2+) applied to training data.
absorbing humidity from the air [92]. It is noteworthy that ZO denoted a According to Table 6, the best algorithm selected as the generalized
different analysis due to the entry of Co2+ causing a stronger crystal model was the ANN model due to the R2 values, which were greater than
lattice providing an increase in thermal stability [93]. The destruction of 0.90. The difference in the RMSE for the training and testing dataset was
the crystal structure of ZO occurs around of 1173.15 K, however the small of |RSMEtrain – RMSEtest| = 0.44. The best configuration for the
maximum limit for which the adsorption properties of ZO are not proposed model was a framework of 3 neurons as input data, 15 in the
compromised is 783.15–883.15 K, thus, nanoadsorbents can be used in hidden layer and 1 response, the RhB removal (%R). The similar R2
high-temperature processes for adsorption. values for training and testing evaluated suggest a low probability of
overfitting [96]. Fig. 5 shows the performance of the neural network
3.2. Experimental design of ZO-Co tested in this work, in terms of neurons (number of nodes).
According to Fig. 5, the performance of the proposed ANN model
Table 5 shows the experimental design encoded of the adsorption of increased as the number of neurons increased, which is verified through
RhB using ZO-Co, where the pH and the concentration of ZO-Co and RhB the RSME value. In this work, the number of nodes tested for the hidden
were used as independent variables, while removal was a dependent layer was 1–15, where from 9 to 15 a steady value was achieved.
variable. However, regarding the difference between RSMEtrain and RMSEtest
According to Table 5, the ideal condition for adsorption was [RhB] generated, the structure with the highest node was selected for further
of = 50 mg L− 1, [ZO-Co] of = 0.3 g/L and pH ≈ 10 with qmax of 132.56 study, in which RSME were 7.61 and 8.05 mg g− 1 for training and testing
mg g− 1. The independent variables of the adsorption process (Pareto data. Therefore, the neural network structure with n = 15 was selected
chart) and 2D response surface are shown in Fig. S6(a and b). Eq. (30) as the best ANN algorithm for the machine learning-based model
presents the model adjusted by CCRD 23 to the adsorption of RhB. development. Table S7 informs a study performed in batch adsorption
based on 10 data generated from the ANN model, where the actual and
q = 43.29 + 7.62*pH (30)
predicted values of dye removal (%) were compared.
In relation to Fig. S6(a and b), it was possible to verify that the pH of According to Table S7, three data points showed relative error higher
the solution had a positive influence (linear) for the adsorption of RhB than 5 %, indicating that the proposed ANN can be characterized as a
dye, where basic pH >10 showed qmax >100 mg g− 1 due to the changes good regression model, being suitable to predict the removal of RhB
in the interaction’s electrostatics of the molecules dye and the nano­ onto the adsorbent prepared in this work. Also, to further investigate the
adsorbent concerning to the pH of the solution [94]. However, the influence of each parameter on the response (%R), a feature importance
concentration of RhB and ZO-Co showed no significant difference study was performed, whose result is shown in Fig. S8.
possibly due to the large number of variables in the adsorption process As can be seen in Fig. S8, all the parameters (pH, concentration of
and the CCRD function used [95]. Moreover, it was indicated that the RhB, and ZO-Co) showed a strong influence on the response variable (%

9
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

3.4. Adsorption study

Fig. 6 and Table S9 show the kinetic curves data adjusted according
to the results obtained in the condition determined by the ML for the
kinetic models PFO, PSO, IPD and Avrami.
According to Table S9, comparing PFO and PSO data the best
adjustment was in PFO at [RhB] = 130 mg/L demonstrating the q1 =
1458 mg g− 1, k1 = 0.0006 min− 1, R2adj = 0.9904, ARE = 14.91 % and
SSE = 19.91 (mg g− 1)2 indicating physical adsorption due to ZO-Co
components. Fig. 6 shows the kinetic models, however, the Avrami
model was the best adjustment with smaller errors qe = 259.17 mg g− 1,
KAV = 0.0069 min− 1, n = 1.5214, R2adj = 0.99721, ARE = 10.40 % and
SSE = 7.21 (mg g− 1)2 probably due to the ability to explain complex
reaction pathways [97]. The constants n and KAV represent the dimen­
sionality of the growth of the adsorption sites, where it demonstrated
unidimensional growth for ZO-Co adsorbing RhB [98]. It is noteworthy
that in cases of homogeneous adsorption, the probability of adsorption
occurring is the same for any region in each time interval (0–180 min),
additionally, the Avrami kinetic constant is independent of the initial
concentration of the adsorbate indicating an optimized model for
Fig. 6. Kinetic adsorption of RhB dye using the models kinetics under ideal several adsorbents used in the adsorption process [99]. Fig. 7 and
condition by ML. [RhB] of = 130 mg L− 1, [ZO-Co] of = 0.6 g L− 1, pH ≈ 7, and T
Table S10 shows the RhB adsorption isotherms using ideal condition by
= 298.15 ± 2 K
ML.
According to Fig. 7 and Table S10, the experimental adsorption data
R) of the ANN model. The value of 0.25 for the Permutation Score was were well adjusted for the Khan isotherms (Qk = 85.23 mg g− 1, Bk =
selected as the criterion for defining whether a variable contributes or 0.308 L mg− 1 and Ak = 0.847), where the experimental points are close
not to change in the removal of the RhB adsorption. Thus, a Permutation to the lines representing these models and the highest values of R2 =
Score greater or equal to 0.25 for a feature (input variable) indicated 0.6554 and the smallest errors ARE = 17.32 % and SSE = 648.29 (mg
that it affects the %R. According to Fig. S8, the concentration of RhB g− 1)2. Khan model is an extension derived from the Langmuir and
showed a higher contribution to the removal of the RhB from the Freundlich isotherms to reduce errors in both generally used to fit
aqueous solution, followed by pH and adsorbent concentration of ZO- experimental data of adsorption of pure adsorbates [100], according to
Co. The ideal condition determined by machine learning was [RhB] this study. It is noteworthy that the variable Ak (<1) is close to the
of = 130 mg/L, [ZO-Co] of = 0.6 g/L, pH ≈ 7, T = 298.15±2 K with %R Langmuir model indicating a homogeneous surface of ZO-Co and one-to-
= 84.74–86.92. one adsorption of the active site and RhB [101]. R2 (<0.7) is observed in
all isotherms used suggesting that the adsorption of RhB for ZO-Co did
not reach equilibrium. Table 7 shows the thermodynamic parameters

Fig. 7. Isotherms adsorption of RhB dye onto the ZO-Co under ideal condition by ML. [RhB] of = 130 mg/L, [ZO-Co] of = 0.6 g/L, pH ≈ 7, and T = 298.15±2 K.

10
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

Table 7 Briefly, pH <7 was the most used in the tests with RhB removal re­
Thermodynamic parameters for the adsorption of RhB with ZO-Co. sults in around 72–98.8 % possibly due to it is providing a pH at which
Temperature (K) △G◦ (kJ mol− 1) △H◦ (kJ mol− 1) △S (kJ mol− 1
K− 1) the RhB molecule and the adsorbate present a stronger electrostatic
interaction increasing the adsorption. Most of the ZO were synthesized
298.15 − 5.47 55.65 20.39
308.15 − 7.86 by the hydrothermal method having the advantage of uniformity/ho­
318.15 − 8.12 mogeneity of the pores increasing the control of the adsorption process.
It is noteworthy that the adsorption of RhB for ZO is affected by the
[RhB] of = 130 mg/L, [ZO-Co] of = 0.6 g/L and pH ≈ 7.
characteristics of the adsorbent such as surface area, surface charge and
porosity and characteristics of the adsorption process such as concen­
tration and RhB, pH of the solution and test time. Therefore, the
respective work produced a cobalt-doped ZO from the sludge and
pumice stone through the hydrothermal method obtaining a removal of
83.66 % of RhB (qmax = 259.17 mg g− 1).

3.6. Phytotoxicity of ZO and ZO-Co

The CCRD 22 results for the ecotoxicity tests with Lactuca sativa L.
seeds are shown in Table 9, where the independent variables were [ZO-
Co] and solution pH and the response variable was seed root radicular
growth (Rg).
In relation to Table 9, the ideal ecotoxicity condition was [ZO-Co] =
0.55 g/L and pH = 11.2 with response variable seed root radicular
growth of 29.72±1.88 mm. Fig. 9 shows; (a) the effects of independent
variables on the ecotoxicity of ZO-Co by Pareto chart; (b) the 2D
response surface; (c) an approximate linear equation of root radicular
growth; and (d) a comparison in terms of root radicular growth of ZO-Co
Fig. 8. Effect of reuse of ZO-Co for RhB dye removal. [RhB] of = 130 mg/L, with ZO.
[ZO-Co] of = 0.6 g/L, pH ≈ 7, and T = 298.15±2 K. According to Fig. 9(a – c), the linear pH interaction promoted direct
and positive effects on the root radicular growth in Lactuca sativa L.
evaluated from this study. seeds (around 59 %) denoting hypotheses such as the pH change can
According to Table 7, the values evaluated for enthalpy and entropy interact with the seeds; (i) alternating the surface charge of ZO-Co
variation were 55.65 kJ mol− 1 and 20.39 kJ mol− 1 K− 1, respectively, [114]; (b) modifying the agglomerated particle size and surface area
which suggests that adsorption is an endothermic process (chemo­ of ZO-Co [115]; and (c) spherical impediment processes occurring in
sorption) and with the increase of randomness in the interface of RhB seeds due to ZO components (Si, Al, Na and Co) modifying the absorp­
with ZO-Co [102,103]. The △G◦ modulus was higher at higher tem­ tion of micronutrients [116]. The 2D response surface of Fig. 9(b)
peratures, demonstrating and confirming that the dye adsorption was demonstrated that at acidic pH (pH <5) the root radicular growth of
more spontaneous at 328 K [104]. seeds was reduced possibly due to the decrease in the bioavailability of

Table 9
3.5. Reuse of ZO-Co CCDR 22 for ecotoxicity of ZO-Co.
[ZO-Co] (g L− 1) pH Rg (mm)
Fig. 8 shows the reuse effect of ZO-Co after IV cycles, where the reuse
of ZO-Co showed the removal of RhB from 83.66 % (181.25 mg g− 1) to 0.55 2 0.88±0.22
0.904 7 16.54±1.64
75.38 % (163.33 mg g− 1) in the IV cycle possibly due to the agglom­
0.196 7 14.54±1.11
eration of RhB in the active sites causing a reduction in active sites due 0.3 10 12.51±0.85
to the fouling of the available pores. It is noteworthy that the chemical 0.55 7 11.44±1.22
adsorption process of RhB decreases the adsorption capacity of repeated 0.8 4 4.44±0.46
cycles leading to a decrease in the adsorption capacity [105]. It is worth 0.3 4 2.85±0.31
0.55 7 11.11±0.97
mentioning that the lost mass of biosorbent was around 0.002 g after the 0.8 10 9.87±0.63
IV cycles, which was considered tiny and could be due to the experi­ 0.55 11.2 29.74±1.88
mental procedure. Table 8 shows the comparison of ZO to RhB 0.5 7 10.05±0.77
adsorption carried out on the ScienceDirect platform.

Table 8
Studies about the [RhB] dye by adsorption carried out with ZO-Co.
Adsorbent Methodological aspects [Adsorbent] (g L− 1) [RhB] (mg/L) pH qmax (mg g− 1) Removal (%) Reference

ZIF-8 Hydrothermal/adsorption (20 min) 0.5 5 7 10 90 [106]

Ɓ-ZO Seeding/adsorption (60 min) 1 50 3 27.97 98.8 [107]
ZO-m Hydrothermal/adsorption (20 min) 2 7 — — 72 [108]
NaP-ZO Hydrothermal/adsorption (20 min) 0.05 20 6 24 98.26 [109]
ZO Fusion/adsorption (280 min) 0.3 300 6 66.86 — [110]
NZ Mineral benefit/adsorption (1440 min) 0.25 250 2 — 92 [111]
Raw zeolites Hydrothermal/adsorption (20 min) 0.02 50 7 4.41 94.11 [112]
Ɓ-ZO Hydrothermal/adsorption (20 min) 1 80 6 578 95.7 [113]
ZO-Co Hydrothermal/adsorption (180 min) 0.6 130 7 259.17 83.66 This study

ZIF-8: zeolite imidazolate framework-8, ZO: zeolite, ZO-m: zeolitic mixture and NZ: cuban natural zeolite.

11
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

Fig. 9. Results of the experimental design, (a) Pareto chart, (b) 2D graphic chart pH and [ZO-Co], (c) linear relation of pH and Rg parameters and (d) ZO-Co and ZO
comparison using the ideal condition ([ZO-Co] = 0.55 g/L and pH = 11.2) with statistical difference of p < 0.05*, p < 0.01**, p < 0.001*** and p < 0.0001****.

12
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

micronutrients [117]. Equation (31) presents the adjusted model of the Appendix A. Supplementary data
phytotoxicity study.
Supplementary data to this article can be found online at https://doi.
Rg(mm) = 2.36*pH + 0.05*pH 2 (31)
org/10.1016/j.jphotochem.2023.115404.
Regarding Fig. 9(d), it was observed root radicular growth in Lactuca
sativa L. seeds containing ZO-Co (29.74±1.88 mm) compared to ZO References
(19.8±1.31 mm) possibly due to a spherical impediment caused by co­
[1] V.B. Yennawar, A.B. Bhosle, P.R. Shaikh, Asssesment of Industrial Wastewater
balt ions modifying the absorption of micronutrients, also, as the
Quality and Management, Science Park. 1 (2014) 27–38, https://doi.org/
decrease in the average hydrodynamic size of 620.71±23.48 nm. It is 10.9780/23218045/1202013/49.
noteworthy that in the analysis of variance (ANOVA) of the CCRD 22 it [2] S. Samsami, M. Mohamadizaniani, M.H. Sarrafzadeh, E.R. Rene, M. Firoozbahr,
was possible to demonstrate the main parameters and regression co­ Recent advances in the treatment of dye-containing wastewater from textile
industries: Overview and perspectives, Process Saf. Environ. Prot. 143 (2020)
efficients were not predicted since the Fcalculated (1.8) was lower than the 138–163, https://doi.org/10.1016/j.psep.2020.05.034.
Ftabeled (5.05) indicating that 59 % of the experimental data adjusted to [3] M.D.R. Vučić, J.Z. Mitrović, M.M. Kostić, N.D. Velinov, S.M. Najdanović, D.
the model. V. Bojić, A.L. Bojić, Heterogeneous photocatalytic degradation of anthraquinone
dye Reactive Blue 19: optimization, comparison between processes and
identification of intermediate products, Water SA. 46 (2020) 291–299, https://
4. Conclusion doi.org/10.17159/wsa/2020.v46.i2.8245.
[4] V.W. Atul, G.S. Gaikwad, M.G. Dhonde, N.T. Khaty, S.R. Thakare, Removal of
organic pollutant from water by heterogenous photocatalysis: A review, J. Chem.
ZO-Co from sludge and pumice residues was prepared for the RhB Environ. 17 (2013) 84–94.
dye removal by adsorption with crystalline phase, SBET = 53 m2/g and a [5] N.Y. Donkadokula, A.K. Kola, I. Naz, D. Saroj, A review on advanced physico
negative charge (ZP = -8.69 mV) indicating an electrostatic interaction chemical and biological textile dye wastewater treatment techniques, Rev.
Environ. Sci. Biotechnol. 19 (2020) 543–560, https://doi.org/10.1007/s11157-
with RhB. The ideal condition was [RhB] = 130 mg L− 1, [ZO-Co] = 0.6 020-09543-z.
g/L, pH ≈ 7 and T = 298.15 with qmax = 259.17 mg g− 1. Ecotoxicity tests [6] S. Ledakowicz, K. Paździor, Recent Achievements in Dyes Removal Focused on
demonstrated the root radicular growth using Lactuca sativa L. seeds, Advanced Oxidation Processes Integrated with Biological Methods, Molecules. 26
(2021) 870–887, https://doi.org/10.3390/molecules26040870.
indicating the absence of toxicity. Therefore, ZO-Co showed good
[7] G.B. Gerber, A. Léonard, P.H. Hantson, Carcinogenicity, mutagenicity, and
adsorption capacity for RhB removal being a potential source for the teratogenicity of manganese compounds, Crit. Rev. Oncol. Hematol. 42 (2002)
preparation of alternative nanoadsorbents relating to nanotechnology 25–34, https://doi.org/10.1016/S1040-8428(01)00178-0.
and sustainability issues. [8] A.A. Pivtsaev, V.I. Azova, Detection of the carcinogenic properties of synthetic
and natural dyes using positron annihilation lifetime spectrometry, Toxicol. Res.
Consent to Publish. 5 (2016) 1306–1308, https://doi.org/10.1039/c6tx00147e.
All authors read and approved the final manuscript. [9] J. Sharma, S. Bhatt, V. Soni, Toxic effects of Rhodamine B on antioxidant system
and photosynthesis of Hydrilla verticillata, J. Hazard. Mater. 3 (2022)
100069–100078, https://doi.org/10.1016/j.hazl.2022.100069.
CRediT authorship contribution statement [10] Q. Guo, Z. Ma, C. Zhou, Z. Ren, X. Yang, Single Molecule Photocatalysis on TiO2
Surfaces, Chem. Rev. 119 (2019) 11020–11041, https://doi.org/10.1021/acs.
M.E.S. Pedebos: Writing – review & editing, Writing – original draft, chemrev.9b00226.
[11] S. Imam, H. Babamale, A short review on the removal of rhodamine B dye using
Validation, Investigation, Formal analysis, Data curation, Conceptuali­ agricultural waste-based adsorbents, Asian, J. Chem. Sci. 7 (2020) 25–37,
zation. D.M. Druzian: Writing – review & editing, Writing – original https://doi.org/10.9734/ajocs/2020/v7i119013.
draft, Validation, Investigation, Formal analysis, Data curation, [12] J.F.D. Neto, D.S. Pereira, V.C. Da-Silva, H.C. Ferreira, G.A. Neves, R.R. Menezes,
Study of equilibrium and kinetic adsorption of rhodamine B onto purified
Conceptualization. L.R. Oviedo: Writing – review & editing, Writing – bentonite clays, Cerâmica. 64 (2018) 598–607, https://doi.org/10.1590/0366-
original draft, Validation, Investigation, Formal analysis, Data curation, 69132018643722429.
Conceptualization. Y.P.M. Ruiz: Writing – review & editing, Formal [13] C.R.B. Rhoden, F.S. Bruckmann, T.R. Salles, C.G.K. Junior, J.S.R. Mortari, Study
from the influence of magnetite onto removal of hydrochlorothiazide from
analysis, Conceptualization. A. Galembeck: Writing – review & editing,
aqueous solutions applying magnetic graphene oxide, J. Water Process. Eng. 43
Conceptualization. G. Pavoski: Writing – review & editing, Formal (2021) 102262–102276, https://doi.org/10.1016/j.jwpe.2021.102262.
analysis, Conceptualization. D.C.R. Espinosa: Writing – review & [14] J. Rajeev, M. Mathur, S. Shalini, A. Mittal, Removal of the hazardous dye
editing, Formal analysis, Conceptualization. W.L. da Silva: Conceptu­ rhodamine B through photocatalytic and adsorption treatments, J. Environ.
Manage. 85 (2007) 956–964, https://doi.org/10.1016/j.jenvman.2006.11.002.
alization, Data curation, Formal analysis, Investigation, Validation, [15] Y.P. Moreno, W.L. Da Silva, F.C. Stedile, C. Radtke, J.H.Z. Dos Santos, Micro and
Writing – original draft, Writing – review & editing. nanodomains on structured silica/titania photocatalysts surface evaluated in RhB
degradation: Effect of structural properties on catalytic efficiency, Appl. Surf. Sci.
3 (2021) 100055–100069, https://doi.org/10.1016/j.apsadv.2021.100055.
Declaration of competing interest [16] J. Zhang, H. Tong, W. Pei, W. Liu, F. Shi, Y. Li, Y. Huo, Integrated photocatalysis-
adsorption-membrane separation in rotating reactor for synergistic removal of
The authors declare that they have no known competing financial RhB, Chemosphere. 270 (2021) 129424–129436, https://doi.org/10.1016/j.
chemosphere.2020.129424.
interests or personal relationships that could have appeared to influence [17] Q. Hu, M. Zhang, L. Xu, S. Wang, S.T. Yang, M. Wu, W. Lu, Y. Li, D. Yu,
the work reported in this paper. Unraveling timescale-dependent Fe-MOFs crystal evolution for catalytic
ozonation reactivity modulation, J. Hazard. Mater. 431 (2022) 128575–128587,
https://doi.org/10.1016/j.jhazmat.2022.128575.
Data availability [18] R.T. Mogharbel, M.M. Aljohani, A.Q. Alorabi, A.M. Al-bonayan, H.M. Abumelha,
T.M. Habeebullah, I. Althagafi, N.M. El-Metwaly, Post-synthetic Ln@modified
Data will be made available on request. MIL (Ti) framework in sono-photocatalysis degradation of carbaryl, Mater. Chem.
Phys. 305 (2023) 127990–127999, https://doi.org/10.1016/j.
matchemphys.2023.127990.
Acknowledgments [19] S.H. Alrefaee, A.M. Al-bonayan, H.H. Alsharief, M. Aljohani, K.F. Alshammari, F.
A. Saad, H.M. Abumelha, N.M. El-Metwaly, Efficient removal of carbofuran by
Department of Fundamental Chemistry at Federal University of sono-photo active CdS@MIL based Ti Framework, Surfaces and Interfaces 40
(2023) 103133–103140, https://doi.org/10.1016/j.surfin.2023.103133.
Pernambuco (UFPE), University of Sao Paulo (LAREX USP, Brazil – SP) [20] L.D. Pompeu, P.C.L. Muraro, G. Chuy, B.S. Vizzotto, G. Pavoski, D.C.R. Espinosa,
and Franciscan University (UFN, Brazil – RS) for the support and assis­ L.S. Fernandes, W.L. Silva, Adsorption for rhodamine b dye and biological activity
tance to carry out the present work. of nano-porous chitosan from shrimp shells, Environ. Sci. Pollut. Res. 29 (2022)
49858–49869, https://doi.org/10.1007/s11356-022-19259-y.
[21] S. Sallam, A.Q. Alorabi, A.R.Z. Almotairy, S.F. Ibarhiam, E. Aljuhani, S.D. Al-
Qahtani, N.M. El-Metwaly, Superior and effective adsorption of amoxicillin by
using novel metal organic framework and its composite: Thermodynamic, kinetic,

13
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

and optimization by Box-Behnken design, Appl. Organomet. Chem. 37 (2023) [44] C.C. Silva, M.P.F. Graça, M.A. Valente, A.S.B. Sombra, Crystallite size study of
e7184. nanocrystalline hydroxyapatite and ceramic system with titanium oxide obtained
[22] T.O. Ajiboye, O.A. Oyewo, D.C. Onwudiwe, Adsorption, and photocatalytic by dry Ball milling, J. Mater. Sci. 42 (2007) 3851–3855, https://doi.org/
removal of Rhodamine B from wastewater using carbon-based materials, 10.1007/s10853-006-0474-0.
FlatChem. 29 (2021) 100277–100294, https://doi.org/10.1016/j. [45] C. Rosa, F. Auriemma, Crystals and Crystallinity in Polymers: Diffraction Analysis
flatc.2021.100277. of Ordered and Disordered Crystals, 1. Ed., Elsevier, Inglaterra, 2014.
[23] M. Oumam, A. Abourriche, S. Mansouri, M. Mouiya, A. Benhammou, [46] T. Virtanen, G. Rudolph, A. Lopatina, B. Al-Rudainy, H. Schagerlöf, L. Puro,
Y. Abouliatim, Y. El-Hafiane, H. Hannache, M. Birot, R. Pailler, R. Naslain, M. Kallioinen, F. Lipnizki, Analysis of membrane fouling by Brunauer-Emmett-
Comparison of chemical and physical activation processes at obtaining Teller nitrogen adsorption/desorption technique, Sci. Rep. 10 (2020) 3427–3444,
adsorbents from Moroccan oil shale, Oil Shale. 37 (2020) 139–157, https://doi. https://doi.org/10.1038/s41598-020-59994-1.
org/10.3176/oil.2020.2.04. [47] F.I.A. Fadl, A.M. Elbarbary, Radiation synthesis and characterization of
[24] G. Zhong, D. Liu, J. Zhang, Incorporation of Functional Groups Expands the heterogeneous magnetic nanocomposites of 2-hydroxyethyl methacrylate for
Applications of UiO-67 for Adsorption, Catalysis and Thiols Detection, catalytic degradation of sandocryl blue dye, Sep. Purif. Technol. 272 (2021)
ChemistrySelect. 3 (2018) 7066–7080, https://doi.org/10.1002/slct.201800840. 118972–118994, https://doi.org/10.1016/j.seppur.2021.118972.
[25] I. Boughaita, C. Bouchelta, M.S. Medjram, P. Magri, Activated Carbon Supported [48] A.A. Oyekanmi, A. Ahmad, K. Hossain, M. Rafatullah, Statistical optimization for
Cobalt as Efficiency Adsorbent: Application Chemical Agricultural Polluant 2,4-D adsorption of Rhodamine B dye from aqueous, J. Mol. Liq. 281 (2019) 48–58,
Herbicide Removal from Aqueous Solution, Orient. J. Chem. 33 (2017) https://doi.org/10.1016/j.molliq.2019.02.057.
2226–2236. https://doi.org/10.13005/ojc/330510. [49] P. Mañas, J. Heras, Phytotoxicity test applied to sewage sludge using Lactuca
[26] H. Yu, S. Wang, Y. Li, Q. Qiao, K. Wang, X. Li, Recovery of cobalt from spent sativa L. and Lepidium sativum L. seeds, Int. J. Environ. Sci. Technol. (2018)
lithium-ion battery cathode materials by using choline chloride-based deep 273–280, https://doi.org/10.1007/s13762-017-1386-z.
eutectic solvent, Green Process. Synth. 11 (2022) 868–874, https://doi.org/ [50] G. Cirini, Kinetic and equilibrium studies on the removal of cationic dyes from
10.1515/gps-2022-0073. aqueous solution by adsorption onto a cyclodextrin polymer, Dyes and Pigm. 77
[27] Y.F. Zhou, R.J. Haynes, Water treatment sludge can be used as an adsorbent for (2008) 415–426, https://doi.org/10.1016/j.dyepig.2007.07.001.
heavy metals in wastewater streams, Int. J. Environ. Waste Man. 379 (2010) [51] V.S. Munagapati, J.C. Wen, C.L. Pan, Y. Gutha, J.H. Wen, Enhanced adsorption
341–352, https://doi.org/10.2495/WM100341. performance of Reactive Red 120 azo dye from aqueous solution using quaternary
[28] M. Ahmaruzzaman, V.K. Gupta, Rice Husk, and Its Ash as Low-Cost Adsorbents in amine modified orange peel powder, J. Mol. Liq. 285 (2019) 375–385, https://
Water and Wastewater Treatment, Ind. Eng. Chem. Res. 50 (2011) 13589–13613, doi.org/10.1016/j.molliq.2019.04.081.
https://doi.org/10.1021/ie201477c. [52] S. Singh, A.G. Anil, S. Khasnabis, V. Kumar, B. Nath, V. Adiga, P.C. Ramamurthy,
[29] S. Candamano, A. Policicchio, G. Conte, R. Abarca, C. Algieri, S. Chakraborty, Sustainable removal of Cr(VI) using graphene oxide-zinc oxide nanohybrid:
S. Curcio, V. Calabrò, F. Crea, R.G. Agostino, Preparation of foamed and Adsorption kinetics, isotherms and thermodynamics, Environ. Res. 203 (2022),
unformed geopolymer/NaX zeolite/activated carbon composites for CO2 111891, https://doi.org/10.1016/j.envres.2021.111891.
adsorption, J. Clean. Prod. 330 (2022) 129843–129862, https://doi.org/ [53] W. Samek, K.R. Towards explainable artificial intelligence. In: Samek, W.,
10.1016/j.jclepro.2021.129843. Montavon, G., Vedaldi, A., Hansen, L.K., Müller, K.-R. (Eds.), Explainable AI:
[30] G. Osman, Characteristics of fired clay bricks with pumice aditive, Energy Build. Interpreting, Explaining and Visualizing Deep Learning, (eds.).,. Springer Cham,
102 (2015) 217–224, https://doi.org/10.1016/J.ENBUILD.2015.05.031. Switzerland, p. 5-22. https://doi.org/10.1007/978-3-030-28954-6, 2019.
[31] B. Rahimi, N. Fari, A. Abdolahnejad, H. Farrokhzadeh, A. Ebrahimi, Application [54] T. Yin, X. Meng, S. Wang, X. Yao, N. Liu, L. Shi, Study on the adsorption of low
of efficient photocatalytic process using a novel BiVO/TiO2– NaY zeolite concentration VOCs on zeolite composites based on chemisorption of metal-
composite for removal of Acid Orange 10 dye in aqueous solutions: modeling by oxides under dry and wet conditions, Sep. Purif. Technol. 280 (2022) 119634-
response surface methodology (RSM), J. Environ. Chem. Eng. 7 (2019) 119649. https://d oi.org/10.1016/j.seppur.2021.119634.
103253–103264, https://doi.org/10.1016/j.dib.2021.106869. [55] J. Coronas, Present and future synthesis challenges for zeolites, Chem. Eng. J. 156
[32] M. Yoldi, E. Fuentes-Ordoñez, S. Korili, A. Gil, Zeolite synthesis from industrial (2010) 236–242, https://doi.org/10.1016/j.cej.2009.11.006.
wastes, Micropor. Mesopor. Mat. 287 (2019) 183–191, https://doi.org/10.1016/ [56] S. Sivalingam, S. Sen, Rapid ultrasound assisted hydrothermal synthesis of highly
j.micromeso.2019.06.009. pure nanozeolite X from fly ash for efficient treatment of industrial effluent,
[33] S.H.S. Filho, L. Bieseki, A.R. Da-Silva, A.A.B. Maia, R.A.S. San-Gil, S.B.C. Pergher, Chemosphere 210 (2018) 816–823, https://doi.org/10.1016/j.
Synthesis of Zeolite A employing Amazon kaolin waste, Cerâmica. 61 (2015) chemosphere.2018.07.091.
409–421, https://doi.org/10.1590/0366-69132015613601898. [57] L.R. Oviedo, V.R. Oviedo, L.D.D. Nora, W.L. Da Silva, Adsorption of organic dyes
[34] W. Zhang, W. Huang, J. Tan, D. Huang, J. Ma, B. Wu, Modeling, optimization and onto nanozeolites: A machine learning study, Sep. Purif. Technol. 315 (2023),
understanding of adsorption process for pollutant removal via machine learning: 123712, https://doi.org/10.1016/j.seppur.2023.123712.
Recent progress and future perspectives, Chemosphere 311 (2023) [58] S. Shahmohammadi, H. Babazadeh, Isotherms for the adsorption of zinc and
137044–137064, https://doi.org/10.1016/j.chemosphere.2022.137044. copper onto kaolinite: comparison of various error functions, Int. J. Environ. Sci.
[35] A. Goel, A.K. Goel, A. Kumar, The role of artificial neural network and machine Technol. 11 (2014) 111–118, https://doi.org/10.1007/s13762-013-0260-x.
learning in utilizing spatial information, Spat. Inf. Res. 31 (2023) 275–285, [59] C.M. Sayago, R. Bingre, L. Huang, G. Lutzweiler, Q. Wang, B. Louis, CO2
https://doi.org/10.1007/s41324-022-00494-x. Adsorption Capacities in Zeolites and Layered Double Hydroxide Materials, Front.
[36] J. Montomoli, L. Romeo, S. Moccia, M. Bernardini, L. Migliorelli, D. Berardini, Chem. 7 (2019) 551–561, https://doi.org/10.3389/fchem.2019.00551.
A. Donati, A. Carsetti, M.G. Bocci, P.D.W. Garcia, T. Fumeaux, P. Guerci, R. [60] J. Wang, Y. Zhou, X. Hu, Adsorption of CO2 by a novel zeolite doped amine
A. Schüpbach, E. Frontoni, M.P. Hilty, Machine learning using the extreme modified ternary aerogels, Environ. Res. 214 (2022) 113855–113867, https://
gradient boosting (XGBoost) algorithm predicts 5-day delta of SOFA score at ICU doi.org/10.1016/j.envres.2022.113855.
admission in COVID-19 patients, J. Intensive Care Med. 1 (2021) 110–116, [61] P. Djomgoue, D. Njopwouo, FT-IR Spectroscopy Applied for Surface Clays
https://doi.org/10.1016/j.jointm.2021.09.002. Characterization, J. Surf. Eng. Mater. Adv. Technol. 456 (2013) 275–282, https://
[37] J. Li, L. Pan, M. Suvarna, Y.W. Tong, X. Wang, Fuel properties of hydrochar and doi.org/10.4236/jsemat.2013.34037.
pyrochar: Prediction and exploration with machine learning, Appl. Energ. 269 [62] C. Li, J. Zhang, Y. Lin, Y. Chen, X. Xie, H. Wang, L. Wang, In situ growth of
(2020) 115166–115180, https://doi.org/10.1016/j.apenergy.2020.115166. layered double hydroxide on disordered platelets of montmorillonite, Appl. Clay
[38] A. Antoniadis, S. Lambert-Lacroix, J.M. Poggi, Random forests for global Sci. 119 (2016) 103–108, https://doi.org/10.1016/j.clay.2015.06.032.
sensitivity analysis: a selective review, Realib. Eng. Syst. 206 (2021), 107312, [63] G. Sheng, J. Hu, X. Wang, Sorption properties of Th(IV) on the raw
https://doi.org/10.1016/j.ress.2020.107312. diatomite–effects of contact time, pH, ionic strength and temperature, Appl.
[39] N. Taoufik, W. Boumya, R. Elmoubarki, A. Elhalil, M. Abdennouri, N. Barka, Radiat. Isot. 66 (2008) 1313–1320, https://doi.org/10.1016/j.
Experimental design, machine learning approaches for the optimization and apradiso.2008.03.005.
modeling of caffeine adsorption, Mater. Today Chem. 23 (2022), 100732, https:// [64] J.S. Rodrigues, K. Palmezani, R.R.M. Freitas, J.O. Silva, V.H. Lima, V.R. Botaro,
doi.org/10.1016/j.mtchem.2021.100732. Isolation and Characterization of Acetosolv Lignin Present in Sugarcane Bagasse,
[40] L. Fernandes, M.G. Sabino, H.L. Rosetto, Method of silica extraction from rice Rev. Virtual Quim. 12 (2020) 930–937. https://doi.org/10.21577/1984-6835
husk, Cerâmica. 60 (2014) 160–163, https://doi.org/10.1590/S0366- .20200069.
69132014000100022. [65] W. Rongchapo, C. Keawkumay, N. Osakoo, K. Deekamwong, N. Chanlek,
[41] Y. Zhang, Z. Leng, F. Zou, L. Wang, S.S. Chen, D.C.W. Tsang, Synthesis of zeolite A S. Prayoonpokarach, J. Wittayakun, Comprehension of paraquat adsorption on
using sewage sludge ash for application in warm mix asphalt, J. Clean. Prod. 172 faujasite zeolite X and Y in sodium form, Adsorp, Sci. Technol. 36 (2018)
(2018) 686–695, https://doi.org/10.1016/j.jclepro.2017.10.005. 684–693, https://doi.org/10.1177/0263617417715394.
[42] W, Ahmad, U, Mehmoodb, A, Al-Ahmed, F.A. Al-Sulaiman, M.Z. Aslam, M.S. [66] M. Sprynskyy, T. Kowalkowski, H. Tutu, E.M. Cukrowska, B. Buszewski, Ionic
Kamal, R.A. Shawabkeh, Synthesis of zinc oxide/titanium dioxide (ZnO/TiO2) liquid modified diatomite as a new effective adsorbent for uranium ions removal
nanocomposites by wet incipient wetness impregnation method and preparation from aqueous solution, Colloids Surf. A Physicochem. Eng. Asp. 465 (2015)
of ZnO/TiO2 paste using poly(vinylpyrrolidone) for efficient dye-sensitized solar 159–167, https://doi.org/10.1016/j.colsurfa.2014.10.042.
cells, Electrochim. Acta. 222 (2016) 473-480. https://doi.org/10.1016/j. [67] P. Laoot, K. Rangsriwatananon, A. Chaisena, Synthesis of sodium-type
electacta.2016.10.200. fluorophlogopite mica from perlite and diatomite, J. Ceram. Process. Res. 460
[43] A. Shakya, M. Vithanage, T. Agarwal, Influence of pyrolysis temperature on (2011) 273–278.
biochar properties and Cr (VI) adsorption from water with groundnut shell [68] Z. Jiang, J. Yang, H. Ma, X. Ma, J. Yuan, Synthesis of pure NaA zeolites from coal
biochars: Mechanistic approach, Environ. Res. 215 (2022) 114243–114262, fly ashes for ammonium removal from aqueous solutions, Clean Technol. Environ,
https://doi.org/10.1016/j.envres.2022.114243. Policy. 18 (2016) 629–637, https://doi.org/10.1007/s10098-015-1072-0.

14
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

[69] A. Al-Futaisi, A. Jamrah, R. Ai-Hanai, Aspects of cationic dye molecule adsorption [92] M.K. Król, P. Jeleń, The Effect of Heat Treatment on the Structure of Zeolite A,
to palygorskite, Desalination. 214 (2007) 327–342, https://doi.org/10.1016/j. Materials (basel). 14 (2021) 4642–4651, https://doi.org/10.3390/ma14164642.
desal.2006.10.024. [93] H. Liu, Z. Guo, Q. Zhang, B. Jin, R. Peng, Zeolite Imidazolate Frameworks-67
[70] C. Schlumberger, M. Thommes, Characterization of Hierarchically Ordered Precursor to Fabricate a Highly Active Cobalt-Embedded N-Doped Porous
Porous Materials by Physisorption and Mercury Porosimetry—A Tutorial Review, Graphitized Carbon Catalyst for the Thermal Decomposition of Ammonium
Adv. Mater. Interfaces. 8 (2021) 2002181, https://doi.org/10.1002/ Perchlorate, ACS Omega. 6 (2021) 25440–25446, https://doi.org/10.1021/
admi.202002181. acsomega.1c03427.
[71] B. Muir, T. Bajda, A. Jeleń, Removal of chromium (VI) from aqueous solutions [94] Y. Yu, N.M. Bandaru, J.G. Shapter, K. Constantopoulos, N.H. Voelcker, A.V. Ellis,
using zeolites modified with HDTMA and ODTMA surfactants, Clay Miner. 50 Benzene carboxylic acid derivatized graphene oxide nanosheets on natural
(2015) 103–115, https://doi.org/10.1180/claymin.2015.050.1.10. zeolites as effective adsorbents for cationic dye removal, J. Hazard. Mater. 260
[72] F. Sumari, A. Fajaroh, R.K. Santoso, Wardani, Performance of Activated Natural (2013) 330–338, https://doi.org/10.1016/j.jhazmat.2013.05.041.
Zeolite/Cu as a catalyst on Degradation of Glycerol into Ethanol Assisted by [95] L.A. Kafshgari, M. Ghorbani, A. Azizi, S. Agarwal, V.K. Gupta, Modeling and
Ultrasonic, J. Phys. Conf. Ser. 1093 (2018) 012036–012043, https://doi.org/ optimization of Direct Red 16 adsorption from aqueous solutions using
10.1088/1742-6596/1093/1/012036. nanocomposite of MnFe2O4/MWCNTs: RSM-CCRD model, J. Mol. Liq. 233 (2017)
[73] K. Komura, E. Imai, T. Ikeda, GAM-5: A microporous cobalt-containing alumino 370–377, https://doi.org/10.1016/j.molliq.2017.03.047.
phosphate with ATV topology formed by the interzeolite conversion of CoAPO-5 [96] A.N. Ebelegi, N. Ayawei, D. Wankasi, Interpretation of Adsorption
via AWO-type zeolite, Microporous Mesoporous Mater. 344 (2022) Thermodynamics and Kinetics, J. Phys. Chem. 10 (2020) 166–182, https://doi.
112245–112253, https://doi.org/10.1016/j.micromeso.2022.112245. org/10.4236/ojpc.2020.103010.
[74] A.K. Ladavos, A.P. Katsoulidis, A. Iosifidis, K.S. Triantafyllidis, T.J. Pinnavaia, P. [97] B. Royer, N.F. Cardoso, E.C. Lima, J.C.P. Vaghetti, N.M. Simon, T. Calvete, R.
J. Pomonis, The BET equation, the inflection points of N2 adsorption isotherms C. Veses, Applications of Brazilian pine-fruit shell in natural and carbonized forms
and the estimation of specific surface area of porous solids, Microporous as adsorbents to removal of methylene blue from aqueous solutions-Kinetic and
Mesoporous Mater. 151 (2012) 126–133, https://doi.org/10.1016/j. equilibrium study, J. Hazard. Mater. 164 (2009) 1213–1222, https://doi.org/
micromeso.2011.11.005. 10.1016/j.jhazmat.2008.09.028.
[75] A. Füger, F. Konrad, A. Leis, M. Dietzel, V. Mavromatis, Effect of growth rate and [98] M. Avrami, Kinetics of phase change. I: General theory, J. Chem. Phys. 7 (1939)
pH on lithium incorporation in calcite, Geochim. Cosmochim. Acta. 248 (2019) 1103–1112, https://doi.org/10.1063/1.1750380.
14–24, https://doi.org/10.1016/j.gca.2018.12.040. [99] E.W. Menezes, E.C. Lima, B. Royer, F.E. de Souza, B.D. dos Santos, J.R. Gregório,
[76] X. Liu, P.M. Arvela, A. Aho, Z. Vajglova, V.M. Gun’ko, I. Heinmaa, N. Kumar, K. T.M.H. Costa, Y. Gushikem, E.V. Benvenutti, Ionic silica-based hybrid material
Eränen, T. Salmi, D.Y. Murzin, Zeta Potential of Beta Zeolites: Influence of containing the pyridinium group used as an adsorbent for textile dye, J. Colloid
Structure, Acidity, pH, Temperature and Concentration, Molecules. 23 (2018) Interface Sci. 378 (2012) 10–20, https://doi.org/10.1016/j.jcis.2012.04.021.
946-960. https://doi.org/10.3390/molecules23040946. [100] H. Sereshti, H. Gaikani, H.R. Nodeh, The effective removal of mercury ions (Hg2
[77] K. Shen, M.A. Gondal, Removal of Hazardous Rhodamine Dye from Water by +) from water using cadmium sulfide nanoparticles doped in polycaprolactam
Adsorption onto Exhausted Coffee Ground, J. Saudi Chem. Soc. 21 (2017) nanofibers: kinetic and equilibrium studies, J. Iran. Chem. Soc. 15 (2017)
120–127, https://doi.org/10.1016/j.jscs.2013.11.005. 743–751, https://doi.org/10.1007/s13738-017-1274-y.
[78] L.D. Pompeu, P.C.L. Muraro, G. Chuy, B.S. Vizzotto, G. Pavoski, D.C.R. Espinosa, [101] A.R. Khan, R. Ataullah, A. Al-Haddad, Equilibrium adsorption studies of some
L.S. Fernandes, W.L. da Silva, Adsorption for rhodamine b dye and biological aromatic pollutants from dilute aqueous solutions on activated carbon at different
activity of nano-porous chitosan from shrimp shells, Environ. Sci. Pollut. Res. Int. temperatures, J. Colloid Interface Sci. 194 (1997) 154–165, https://doi.org/
29 (2022) 49858–49869, https://doi.org/10.1007/s11356-022-19259-y. 10.1006/jcis.1997.5041.
[79] A.K. Al-Buriahi, A.A. Al-Gheethi, P.K. Senthil, R.M.S. Radin, H. Yusof, A. [102] S. Chowdhury, R. Mishra, P. Saha, P. Kushwaha, Adsorption thermodynamics,
F. Alshalif, N.A. Khalifa, Elimination of rhodamine B from textile wastewater kinetics and isosteric heat of adsorption of malachite green onto chemically
using nanoparticle photocatalysts: A review for sustainable approaches, modified rice husk, Desalination 265 (2011) 159–168, https://doi.org/10.1016/j.
Chemosphere. 287 (2022) 132162–132176, https://doi.org/10.1016/j. desal.2010.07.047.
chemosphere.2021.132162. [103] G.W. Kajjumba, S. Emik, A. Öngen, S. Aydın, G.W. Kajjumba, S. Emik, A. Öngen,
[80] F. Davar, Z. Fereshteh, H.S. Razavi, R.S. Razavi, M.R. Loghman-Estarki, Synthesis S. Aydın, Modelling of Adsorption Kinetic Processes—Errors, Theory and
and characterization of cobalt oxide nanocomposite based on the Co3O4–zeolite Application, Adv. Sorption Process Appl. (2018), https://doi.org/10.5772/
Y, Superlattices Microstruct. 66 (2014) 85–95, https://doi.org/10.1016/j. INTECHOPEN.80495.
spmi.2013.11.024. [104] J.R. Regalbuto, K. Agashe, A. Navada, M.L. Bricker, Q. Chen, A Scientific
[81] E.C. Ngeno, V. Shikuku, F. Orata, D.L. Baraza, S.J. Kimosop, Caffeine and Description of Pt Adsorption onto Alumina, Stud. Surf. Sci. Catal. 118 (1198) 147-
Ciprofloxacin Adsorption from Water onto Clinoptilolite: Linear Isotherms, 156. https://doi.org/10.1016/S0167-2991(98)80177-8.
Kinetics, Thermodynamic and Mechanistic Studies, S. Afr. J. Chem. 72 (2019) [105] J. Wang, S. Wang, Preparation, modification, and environmental application of
136–142. https://doi.org/10.17159/0379-4350/2019/v72a17. biochar: A review, J. Clean. Prod. 227 (2019) 1002–1022, https://doi.org/
[82] M.A. Tony, Zeolite-based adsorbent from alum sludge residue for textile 10.1016/j.jclepro.2019.04.282.
wastewater treatment, Int. J. Sci. Environ. Technol. 17 (2020) 2485–2498, [106] V.A. Trana, B. Khanh, T. Voe, L. Van, D. Ha, Experimental and computational
https://doi.org/10.1007/s13762-020-02646-8. investigation on interaction mechanism of Rhodamine B adsorption and
[83] M. Castro, P. Losch, W. Park, M. Haouas, F. Taulelle, C. Loerbroks, E. Breynaert, photodegradation by zeolite imidazole frameworks-8, Appl. Surf. Sci. 538 (2021)
G. Brabants, C.E.A. Kirschhock, R. Ryoo, W. Schmidt, Unraveling Direct 148065–148077, https://doi.org/10.1016/j.apsusc.2020.148065.
Formation of Hierarchical Zeolite Beta by Dynamic Light Scattering, Small Angle [107] C. Zhi-Lin, L. Xiang, L. Zan, Study on adsorption of rhodamine B onto Beta zeolites
X-ray Scattering, and Liquid and Solid-State NMR: Insights at the Supramolecular by tuning SiO2/Al2O3 ratio, Ecotoxicol. Environ. Saf. 148 (2018) 585–592,
Level, Chem. Mater. 30 (2018) 2676–2686, https://doi.org/10.1021/acs. https://doi.org/10.1016/j.ecoenv.2017.11.005.
chemmater.8b00109ff. [108] S. Filho, P. Vinanches, S. Pergher, Zeolite synthesis in basic media using expanded
[84] G. Yuming, L. Ziteng, C. Yu, X. Gu, L. Xu, Y. Gao, J. Ma, Zeolite Adsorption perlite and its application in Rhodamine B adsorption, Mater. Lett. 227 (2018)
Isotherms Predicted by Pore Channel and Local Environment Descriptors: Feature 258–260, https://doi.org/10.1016/j.matlet.2018.05.095.
Learning on DFT Binding Strength, J. Phys. Chem. c. 124 (2020) 9314–9328, [109] Y. Zhang, H. Han, X. Wang, X. Wang, M. Zhang, Y. Chen, C. Zhai, H. Song,
https://doi.org/10.1021/acs.jpcc.0c00130. Utilization of NaP zeolite synthesized with different silicon species and NaAlO2
[85] Z. Yang, R. Zhang, R. Liu, S. Zhang, Elucidating the Zeolite Particle Size Effect on from coal fly ash for the adsorption of Rhodamine B, J. Hazard. Mater. 415 (2021)
Butene/Isobutane Alkylation, Ind. Eng. Chem. Res. 61 (2022) 1032–1043, 125627–125639, https://doi.org/10.1016/j.jhazmat.2021.125627.
https://doi.org/10.1021/acs.iecr.1c02038. [110] C. Li, Y. Yu, Q. Zhang, Adsorption of Cu2+ and rhodamine B by zeolite obtained
[86] G. Ye, Y. Sun, Z. Guo, K. Zhu, H. Liu, X. Zhou, M.O. Coppens, Effects of zeolite from electrolytic manganese residue, OP Conf. Ser.: Earth, Environ. Sci. 612
particle size and internal grain boundaries on Pt/Beta catalyzed isomerization of (2020) 12038–12050, https://doi.org/10.1088/1755-1315/612/1/012038.
n-pentane, J. Catal. 360 (2018) 152–159, https://doi.org/10.1016/j. [111] T. Pineira, O. Mozo, B. Rosabal, L. Ménorval, Cuban clinoptilolite for rhodamine B
jcat.2018.01.033. adsorption. Characterization of the obtained zeolite-dye materials, Rev. Cub.
[87] W. Franus, Characterization of X-type zeolite prepared from coal fly ash, Pol. J. Quim. 30 (2018) 175–190.
Environ. Stud. 21 (2012) 337–343. [112] W. Cheng, N. Li, Y. Pan, L. Jin, The Adsorption of Rhodamine B in Water by
[88] E.P. Botella, S. Valencia, F. Rey, Zeolites in Adsorption Processes: State of the Art Modified Zeolites, Mod. Appl. Sci 10 (2016) 67–76, https://doi.org/10.5539/
and Future Prospects, Chem. Rev. 122 (2022) 17647–17695, https://doi.org/ mas.v10n5p67.
10.1021/acs.chemrev.2c00140. [113] M. Yin, Y. Pan, C. Pan, Adsorption properties of graphite oxide for Rhodamine B,
[89] B. Reiprich, K.A. Tarach, K. Pyra, G. Grzybek, K.G. Marek, High-Silica Layer-like Mater. Sci. Eng. 14 (2019) 1121–1212, https://doi.org/10.1049/mnl.2018.5820.
Zeolites Y from Seeding-Free Synthesis and Their Catalytic Performance in Low- [114] N.R. Fajdetić, T.B. Lačić, K. Mirosavljević, S. Antunović, R. Benković, M. Rakić,
Density Polyethylene Cracking, ACS Appl. Mater. Interfaces. 14 (2022) S. Milošević, B.J. Palenkić, Influence of Seed Treated by Plasma Activated Water
6667–6679, https://doi.org/10.1021/acsami.1c21471. on the Growth of Lactuca sativa L, Sustainability. 14 (2022) 16237–16244,
[90] A. Chaibi, Y. Boucheffa, N.B. Lalaoui, TGA investigation of water and ethanol https://doi.org/10.3390/su142316237.
adsorption over LTA zeolites, Microporous Mesoporous Mater. 324 (2021) [115] C. Wang, K. Jiang, B. Wu, J. Zhou, Y. Lv, Silver nanoparticles with different
111285–111297, https://doi.org/10.1016/j.micromeso.2021.111285. particle sizes enhance the allelopathic effects of Canada goldenrod on the seed
[91] A.K. Mosai, L. Chimuka, E.M. Cukrowska, I.A. Kotzé, H. Tutu, The Recovery of
Rare Earth Elements (REEs) from Aqueous Solutions Using Natural Zeolite and
Bentonite, Water Air Soil Pollut. 230 (2019) 188–205, https://doi.org/10.1007/
s11270-019-4236-4.

15
M.E.S. Pedebos et al. Journal of Photochemistry & Photobiology, A: Chemistry 449 (2024) 115404

germination and seedling development of lettuce, Ecotoxicology. 27 (2018) of Salicylic Acid, Yeast, and Zeolite, Life. 12 (2022) 1538–1553, https://doi.org/
1116–1125, https://doi.org/10.1007/s10646-018-1966-9. 10.3390/life12101538.
[116] M. Babaousmail, M.S. Nili, R. Brik, M. Saadouni, S.K.M. Yousif, R.M. Omer, N. [117] G.I. Gadotti, G.E. Meneghello, M.A.A. Tillmann, Requirement range and pH
A. Osman, A.A. Alsahli, H. Ashour, A.M. El-Taher, Improving the Tolerance to influence on the germination test, Revista Fac, Agron. Univ. Nac. La Plata. 112
Salinity Stress in Lettuce Plants (Lactuca sativa L.) Using Exogenous Application (2013) 27–34, https://doi.org/10.1590/0102-33062017abb0001.

16