Green synthesis of bentonite-supported iron nanoparticles as a heterogeneous Fenton-like catalyst:Kinetics of decolorization of reactive blue 238 dye

Ahmed Khudhair Hassan *,Ghayda Yaseen Al-Kindi ,Dalal Ghanim

a Environment and Water Directorate,Ministry of Science and Technology,Baghdad 10070,Iraq

b Sanitary and Environmental Branch,Civil Engineering Department,University of Technology,Baghdad 10009,Iraq

Received 25 March 2020;accepted 18 August 2020

Available online 8 December 2020

Abstract This study aimed to synthesize green tea nano zero-valent iron(GT-NZVI)and bentonite-supported green tea nano zero-valent iron(BGT-NZVI)nanoparticles using green tea extracts in an environmentally sustainable way.Bentonite was used as a support material because it disperses and stabilizes GT-NZVI,and it helps to reduce the cost,increase the adsorption capacity of GT-NZVI,and decrease the optimum amount of GT-NZVI used in Fenton-like oxidation.A scanning electron microscope,an atomic force microscope,and a Fourier transform infrared spectroscope were used to characterize GT-NZVI and B-GT-NZVI,while the zeta potential was measured to evaluate the stability of iron nanoparticles.The decolorization kinetics of reactive blue 238(RB 238)dye in the aqueous phase in the Fenton-like oxidation process were investigated as well.The effects of various experimental conditions such as reaction time,dosages of catalysts,concentration of H2O2,temperature,addition of inorganic salts,and other parameters were investigated.The results show that the oxidative degradation efficiencies of RB 238 dye catalyzed by GT-NZVI and B-GT-NZVI were 93.5%and 96.2%,respectively,at the optimum reaction conditions as follows:c(H2O2)=5 mmol/L,ρ(GT-NZVI)=0.5 g/L orρ(B-GT-NZVI)=0.5 g/L,c(RB 238 dye)=0.05 mmol/L,and pH=2.5 at 180 min.The best catalytic performance was exhibited when B-GT-NZVI was used.Three kinetic models were employed,and the second-order model was found to be the best model representing the experimental kinetic data of RB 238 dye.The value of activation energy decreased from 38.22 kJ/mol for GT-NZVI to 14.13 kJ/mol for B-GT-NZVI.This indicates that the effect of B-GT-NZVI in decreasing the energy barrier is more pronounced than that of the GT-NZVI catalyst,leading to improved Fenton-like oxidation processes.© 2020 Hohai University.Production and hosting by Elsevier B.V.This is an open access article under the CC BY-NC-ND license(http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords:Green synthesis;Bentonite-supported iron nanoparticles;Azo dye;Fenton-like process;Kinetics

1.Introduction

The uncontrolled discharge of toxic and persistent organic chemicals into aquatic systems is the most severe form of water pollution in developing countries(Gwenzi and Chaukura,2018;Melnyk et al.,2014).The potential health hazard of textile dyes is a function of their firmness in the environment as well as their toxicity.These dyes tend to accumulate in the environment and lead to bioaccumulation(Gudelj et al.,2011).The discharged wastewater containing azo dyes has a carcinogenic potential for humans and generates a high degree of environmental toxicity.Under anaerobic conditions,azo dyes may break down into carcinogenic aromatic amines,which can cause the potential for allergic dermatitis,skin irritation,mutations,or cancer(Lellis et al.,2019).

The classical wastewater remediation methods,such as physicochemical-process-based coagulation-flocculation(Zahrim and Hilal,2013),adsorption(Elmoubarki et al.,2015),precipitation(Wang et al.,2019),and biological treatment(Paz et al.,2017),are unable to effectively degrade azo dyes.The green synthesis approach has become a simple alternative to toxic chemical procedures because it uses biological microorganisms or plant extracts,and it is environmentally friendly,economical,and easy to extend to large-scale synthesis for bioreduction of metal ions.Green synthesis typically involves the selection of green solvent media,the selection of environmental benign-reduction agents,and the stable selection of nanoparticles with non-toxic substances(Shukla and Iravani,2018).Plant extracts have been widely used for the synthesis of iron nanoparticles to minimize the negative effects of chemical synthesis,and they have a great advantage due to their natural abundance,low cost,eco-friendliness,and high amount of surface area(Demirezen et al.,2019;Bao et al.,2019;Herlekar et al.,2014).Thus,these compounds can be considered replacements for chemical synthesis of nano zero-valent iron(NZVI).Moreover,plant extracts act as both capping and dispersive agents,leading to minimization of the oxidation and agglomeration of NZVI(Wang et al.,2014a;Huang et al.,2014).Of the studied support materials,bentonite is a widely distributed clay mineral.It is regarded as one of the low-cost and efficient adsorbents.It mainly consists of montmorillonite with a high cation exchange capacity and a large specific surface area and can be used as an efficient adsorbent for removing toxic metals from wastewater(Shao et al.,2018;Wang et al.,2018).Many recent studies on green synthesis of NZVI have been conducted to remove organic pollutants.For instance,Sravanthi et al.(2019)reported the synthesis of a green catalyst of NZVI nanoparticles supported on the bentonite surface and used it to degrade 4-nitrophenol.Mahmoud et al.(2020)also studied the removal of chemical oxygen demand and biological oxygen demand from wastewater using green tea nano zerovalent iron(GT-NZVI).

Advanced oxidation technologies(AOTs)are alternative techniques used to completely remove organic compounds.AOTs have been widely used in the last decade because they can solve the problem of dye degradation in aqueous systems(Serrano-Martínez et al.,2020;Hassan et al.,2019;Trov'o et al.,2016).AOTs derive from the production of highly reactive species and non-selective ones such as hydroxyl radicals(.OH)that easily oxidize a wide range of organic pollutants(Weng et al.,2020).The green Fenton-like oxidation process is a low-cost,rapid,and environmentally friendly technique that can be easily used to degrade textile dyes in industrial water(Yuan et al.,2020;Wu et al.,2015).In the green Fenton-like oxidation process,hydroxyl radicals are produced by a green synthesis of metal-catalyzed oxidation using hydrogen peroxide and subsequently are consumed to remove azo dyes(Ali et al.,2019;Beheshtkhoo et al.,2018;Ebrahiminezhad et al.,2018).

In this study,GT-NZVI and bentonite-supported green tea nano zero-valent iron(B-GT-NZVI)nanoparticles were synthesized using the green tea aqueous leaf extract,and they were characterized with a scanning electron microscope(SEM),an atomic force microscope(AFM),and a Fourier transform infrared spectroscope(FTIR).The purpose of coating bentonite with GT-NZVI was to reduce the cost,to enhance the adsorption capacity of GT-NZVI,and to reduce the optimum amount of GT-NZVI.The goals of this study were(1)to investigate the functions of GT-NZVI and B-GT-NZVI in the Fenton-like oxidation;(2)to investigate the influences of various oxidation parameters,such as green catalyst dosage,H2O2dosage,pH,reaction temperature,and inorganic anions,on the degradation rate of the reactive blue 238(RB 238)dye aqueous solution;(3)to assess the degradation kinetic models and activation energy based on GT-NZVI and B-GT-NZVI;and(4)to develop a possible mechanism of oxidative degradation using a green Fenton-like system.

2.Materials and methods

2.1.Materials and solutions

All chemicals used in this study were of analytical reagent grade,and deionized water was used for preparing all solutions.The RB 238 dye was provided by the Al-Kut Textile Factory,in Iraq.The main characteristics and chemical structure of RB 238 dye are shown in supplementary data.The clay mineral bentonite is primarily sodium montmorillonite and was manufactured in Inner Mongolia,China.The dry green tea leaves(Ahmed brand)were purchased in the local market.Iron(III)chloride(FeCl3),hydrogen peroxide 30% weight per weight(H2O2),sodium chloride(NaCl),sodium carbonate(Na2CO3),sodium sulfate(Na2SO4),sodium sulfite(Na2SO3),and sodium nitrate(NaNO3)were obtained from Sigma-Aldrich.Sodium hydroxide(NaOH)and sulfuric acid(H2SO4)were used to adjust pH,and were purchased from AppliChem(GmbH).The UV-Vis absorption spectra and calibration graph of the RB 238 dye are provided in Fig.1.

2.2.Preparation of GT-NZVI

The GT-NZVI nanoparticles were prepared using a similar procedure described in previous studies(Wang et al.,2014b)with some modifications.The procedure was as follows:

Step 1:The green tea extract was prepared by weighing 20.0 g of green tea(Ahmed brand)in 200-mL deionized water.The solution was heated for 30 min at 85°C on a hot plate.The extract was filtered using a 0.45-μm membrane filter to remove the suspended tea particles and stored in the refrigerator for further use.

Step 2:A solution of 0.10 mol/L FeCl3was prepared by adding 3.25 g of solid FeCl3in 200-mL deionized water.After complete dissolution,this solution was filtered with a 0.45-μm membrane filter to remove the impurities.

Step 3:The tea extract in Step 1 was mixed with the solution of 0.10 mol/L FeCl3(Step 2)through slow addition for 15 min at room temperature and was constantly stirred at a magnetic stirring rate of 300 rpm.Vilardi et al.(2019b)found that the mixing intensity is very important for nanoparticle nucleation.After the addition,the solution color changed from yellow to black,indicating the reduction of iron(III)ions to iron(0)nanoparticles.Subsequently,1.0 mol/L of NaOH solution was added drop by drop until the pH was adjusted to 6.0 and was constantly stirred for 15 min at a stirring rate of 300 rpm.The black precipitate of iron particles was collected through vacuum filtration with a filter paper with a 0.45-μm pore size and quickly rinsed several times with water and ethanol.The GT-NZVI nanoparticles were dried overnight in an oven at 60°C and then ground to fine powder.

Fig.1.UV-Vis absorption spectra and calibration at various concentrations of RB 238 dye solutions.

2.3.Preparation of B-GT-NZVI

To obtain B-GT-NZVI with an iron/bentonite mass ratio of 0.3:0.7,a mass amount of bentonite was calculated to contain a 30%(weight per weight)ratio of theoretical GT-NZVI in the total bentonite mass.The B-GT-NZVI nanoparticles were prepared using the following steps:

Step 1:A quantity of 2.0 g of bentonite and 150 mL of pure water were added into a beaker and allowed to stand for 60 min at room temperature.

Step 2:A solution of 0.10 mol/L FeCl3was prepared by adding 2.5 g of solid FeCl3in 150-mL deionized water.After complete dissolution,this solution was filtered with a 0.45-μm membrane filter to remove impurities.The solution was then added to the bentonite solution(Step 1)and stirred for 60 min with an ultrasonic vibration bath.

Step 3:The green tea extract was prepared by weighing 10.0 g of green tea in 100 mL of deionized water according to the procedure of Step 1 in Section 2.2.

Step 4:The tea extract in Step 3 was added into the mixture of ferric chloride and bentonite(Step 2)slowly for 15 min at room temperature and was constantly stirred.The procedure that followed was the same as that of Step 3 in Section 2.2.

2.4.Characterization of GT-NZVI and B-GT-NZVI nanoparticles

The SEM model was used to investigate the morphology and distribution of the GT-NZVI nanoparticles.SEM is a highresolution microscope that captures digital images(electron micrographs)of approximately 0.1 nm in size.Fig.2(a)and(b)shows the SEM images of GT-NZVI and B-GT-NZVI.AFM was used to measure the surface morphology of nanoparticles(Kiruba Daniel et al.,2013).As shown in Fig.3(a)and(b),GTNZVI and B-GT-NZVI exhibited an irregular spherical shape of less than 50 nm on average as observed by AFM.

The FTIR spectra of the GT-NZVI and B-GT-NZVI nanoparticles in powdered form(Fig.4)were recorded with a potassium bromide(KBr)pellet.Thin plates of GT-NZVI and B-GT-NZVI samples were prepared by mixing the samples with spectrally pure KBr.Subsequently,infrared spectroscopic analysis was performed with the Shimadzu spectrophotometer in the spectral range of 4 000 to 400 cm-1.

The Zeta Potential Analyzer model(NanoBrook ZetaPlus)was used to evaluate the stability of nanoparticles by observing the electrophoretic behavior of fluid.In any nanoparticles,the zeta potential can be ranged from positive at low pH values to negative at high pH values.The pH of measured zeta potential was approximately 6.In terms of nanoparticles in environmental fluid stability,a zeta potential value of higher than 60 mV or lower than-60 mV denotes excellent stability,a value of 40 to 60 mV or-60 to-40 mV represents good stability,a value of 30 to 40 mV or-40 to-30 mV is considered to be stable,and a value ranging from-30 mV to 30 mV is regarded to be highly agglomerative(Setia et al.,2013;Singh et al.,2018).

2.5.Batch experiments for degradation of RB 238 dye aqueous media

Fig.2.SEM images of GT-NZVI and B-GT-NZVI.

In this work,batch experiments were carried out to assess the removal efficiency of RB 238 dye according to the following procedure.In the Fenton-like experiments,the effect of GT-NZVI dosages was studied in the range of 0.1-1.5 g/L on the degradation of 1 L of a synthetic solution with 0.05 mmol/L or 49.6 mg/L RB 238 dye at a preset H2O2concentration of 5 mmol/L and pH of 3.5.The RB 238 dye concentration during the Fenton-like oxidation was monitored by withdrawing 10-mL samples at fixed time intervals,filtering the samples with membrane filters of a 0.22-μm pore size to remove the catalyst,and transferring the samples to a glass vial containing 200μL of 1 mol/L Na2SO3to cease the reaction(Hassan et al.,2019).Subsequently,the impact of H2O2concentration(in the range of 0.64-10 mmol/L)on the removal of RB 238 dye solution(0.05 mmol/L or 49.6 mg/L)was evaluated under the optimum GT-NZVI dosage of 0.5 g/L at a pH of 3.5.Experiments were conducted at a pH value ranging from 2.0 to 9.0 to search for the best pH for the degradation of RB 238 dye.Moreover,the effects of different inorganic salts(NaCl,Na2SO4,Na2CO3,and NaNO3)on the heterogeneous Fenton-like oxidation were studied by adding 1% inorganic salt to the aqueous solution of RB 238 dye(0.05 mmol/L)at the best conditions(c(H2O2)=5 mmol/L,ρ(GT-NZVI)=0.5 g/L,and pH=2.5).Meanwhile,the initial concentration of RB 238 dye was studied in the range of 0.05-0.25 mmol/L under the best conditions.The reaction temperature was evaluated in the range of 20-50°C.Similar experiments using B-GT-NZVI instead of GT-NZVI were conducted to assess the degradation of the aqueous solution of RB 238 dye(0.05 mmol/L).

2.6.Analytical methods

The absorbance of the RB 238 dye solution was measured by a UV-Vis spectrophotometer(model SP-3000 OPTIMA,Japan)at a 605-nm wavelength.The decolorization efficiency(Ed)for RB 238 dye was expressed in terms of the decrease rate in the dye concentration:

wherec0is the initial concentration of RB 238 dye,andctis the concentration of RB 238 dye at timet.

3.Results and discussion

3.1.Characterization of GT-NZVI and B-GT-NZVI nanoparticles

The morphology and size of the synthesized GT-NZVI and B-GT-NZVI samples were verified with the SEM and AFM analyses to investigate the diameter of the nanoparticles as shown in Figs.2 and 3.Fig.2(a)and(b)presents the SEM images of the iron nanoparticles synthesized by green tea extract or green tea supported bentonite.The SEM images depicted the morphology and distributions of GT-NZVI(Fig.2(a))and B-GT-NZVI(Fig.2(b)),respectively.Irregular rectangular-shape nanoparticles were found,probably owing to the existence of polyphenols on the surface of the nanoparticles.This conforms with the results of NZVI preparation using the tree leaf extract and green tea extract(Machado et al.,2015).However,other studies found that iron nanoparticles are produced as irregular spherical particles formed in solution(Huang et al.,2014).This dependence on caffeine and polyphenol percentage presented in the extract can cause different reactions and obstruct or promote the growth of the NZVI nanoparticles,thereby producing various sizes and shapes of iron nanoparticles(Machado et al.,2015).As shown in Fig.3(a),the distribution of the particle size of GT-NZVI-synthesized iron nanoparticles had an average diameter of less than 50 nm.In contrast,B-GT-NZVI presented an average diameter of less than 150 nm with no significant agglomeration(Fig.3(b)).

Fig.3.AFM images of GT-NZVI and B-GT-NZVI.

The FTIR spectra were recorded in the range of 4 000 to 400 cm-1for GT-NZVI and B-GT-NZVI.As shown in Fig.4(a)and(b),the FTIR spectra of the synthesized GTNZVI and B-GT-NZVI displayed broadband stretching vibrations at 3 356 cm-1and 3 414 cm-1for O-H,respectively.This indicates the presence of polyphenol compounds(Abdel-Aziz et al.,2019).The absorption bands at 1 620 cm-1and 1 637.5 cm-1for GT-NZVI and B-GT-NZVI reveal the presence of a C=C aromatic ring stretching vibration,and the bands of 1 062.7 cm-1and 1 084 cm-1denote the existence of C-O-C symmetric stretching vibration of polyphenols(Wang et al.,2014a).The bands described above belonged to the characteristic peaks of green tea extracts.This indicates that the effective components in green tea extracts can be coated onto the surface of GT-NZVI and B-GT-NZVI particles.Fig.4(a)demonstrates that the presence and intensity of phenolic compound peaks can reduce Fe,and are a strong indicator of synthesized GT-NZVI(Abdel-Aziz et al.,2019).Fig.4(b)shows that new bands appeared at 1 045.4 cm-1and 918.1 cm-1.This may be attributed to the bending vibration of Si-O and Al-O vibrations,respectively(Sravanthi et al.,2019).The absorption peak near 792.7 cm-1was caused by the antisymmetric stretching vibration of Si-O-Si in the bentonite(Shao et al.,2018).The bands at 518.8 cm-1and 466.7 cm-1indicate the presence of bending vibrations of Si-O-Al and Si-O-Si in bentonite,respectively(Fig.4(b)).Additionally,the absorption band at 428.27 cm-1may be referred to as Fe-O stretches(Fida et al.,2015).

The stability of nanofluids was characterized by the zeta potential analysis technique.A negative zeta potential of-19.34 mV was found for GT-NZVI(Fig.5(a)),and-22.05 mV for B-GT-NZVI(Fig.5(b)),indicating the poor degree of coagulation to both materials.In Fig.5,the blue line denotes zero charge,while the red line represents the zeta potential values of-19.34 mV and-22.05 mV for GT-NZVI and B-GT-NZVI,respectively.Vilardi et al.(2019a)concluded that the zeta potential for classic NZVI in the range of 20-25 mV was considered the instability range of colloids.

Fig.4.FTIR spectra of green synthesized nanoparticles for GT-NZVI and B-GT-NZVI.

3.2.Influence of H2O2 concentration catalyzed byGT-NZVI and B-GT-NZVI on degradation of RB 238 dye aqueous media

The concentration of hydrogen peroxide is a very important parameter for the maximum decolorization efficiency of RB 238 dye from aqueous solutions.The blank experiment demonstrated that neither degradation nor decolorization of RB 238 dye occurred in the presence of H2O2alone.At first,the concentration of GT-NZVI and B-GT-NZVI was fixed at 0.5 g/L.Afterwards,the effect of different concentrations of H2O2on the removal of RB 238 dye aqueous solution(0.05 mmol/L or 49.6 mg/L)at an initial pH of 3.5 and at room temperature was investigated.Fig.6 illustrates the influence of H2O2concentration and oxidative reaction time on the degradation of RB 238 dye solution.The H2O2concentration varied from 0.64 mmol/L to 10 mmol/L.The decolorization efficiency increased with a rise of hydrogen peroxide concentration from 0.64 mmol/L to 5 mmol/L for both catalyzed GT-NZVI and B-GT-NZVI(Fig.6(a)and(b)).This was indeed caused by the high production of hydroxyl free radicals(.OH)when the amount of H2O2was increased.Further increase of H2O2by over 5 mmol/L led to a decreased removal efficiency.For instance,when the H2O2concentration was increased from 5 mmol/L to 10 mmol/L,the decolorization efficiencies catalyzed by GT-NZVI and B-GT-NZVI decreased from 73.5%to 66.8%and from 73.8% to 58.5%,respectively,during 60-min heterogeneous Fenton oxidation(see supplementary data).This behavior may be attributed to the recombination of hydroxyl radicals with H2O2.As a result,.OH radicals were consumed.This process can be illustrated by the following equation(Hassan et al.,2019;Hashemian,2013;Shih and Tso,2012):

Therefore,the hydrogen peroxide concentration of 5 mmol/L was chosen as the optimum concentration and was used in the following set of experiments.

Fig.5.Zeta potential graphs for GT-NZVI and B-GT-NZVI.

Three kinetic models were used to calculate the decolorization rate constants using the linear forms of the zero-order,first-order,and second-order models shown in Eqs.(3)through(5),respectively:

wherek0,k1,andk2are the rate constants for zero-,first-,and second-order reactions,respectively.As indicated in Fig.7(a)and(b),the decolorization kinetic rates of RB 238 dye increased with the rise of H2O2up to 5 mmol/L for GT-NZVI and B-GT-NZVI catalyzed by the Fenton-like oxidation reaction.All data were well fitted by the second-order model because the correlation coefficients of the second-order model were mostly higher than those of the zero-and first-order models.These results agree with the previous study by Hashemian(2013)(see supplementary data).

3.3.InfluencesofGT-NZVIandB-GT-NZVI concentrations on degradation of RB 238 dye aqueous media

The necessary amount of GT-NZVI or B-GT-NZVI for an effective treatment must be specified.Therefore,to evaluate the relationship between the decolorization efficiency and the dosage of GT-NZVI and B-GT-NZVI,a set of experiments was conducted by fixing the concentrations of H2O2and RB 238 dye and the pH value of the solution as 5 mmol/L,0.05 mmol/L,and 3.5,respectively.

Fig.6.Decolorization efficiencies of RB 238 dye with various concentrations of H2O2 catalyzed by GT-NZVI and B-GT-NZVI.

Fig.7.Linear plots of kinetic data calculated by second-order model at various concentrations of H2O2 catalyzed by GT-NZVI and B-GT-NZVI.

Fig.8.Decolorization efficiencies of RB 238 dye with various dosages of GT-NZVI and B-GT-NZVI.

Fig.8 demonstrates that the decolorization efficiencies increased with an increased amount of GT-NZVI or B-GTNZVI up to the optimum dosage(0.5 g/L).The removal efficiency of RB 238 dye solutions at the higher dosages of both GT-NZVI and B-GT-NZVI resulted in a decreased degradation.This is because the iron nanoparticles contain tea polyphenols at the surface.Increasing the amount of GT-NZVI or B-GT-NZVI may lead to more polyphenols discharge in the aqueous dye solution and consumption of.OH free radicals(Wu et al.,2015).¨Onal et al.(2017)found similar behavior in the decolorization of basic red 46 by the Fenton-like oxidation reactions using GT-NZVI as the catalyst.Fig.9(a)and(b)shows that the second-order model provided the best fit with the experimental data for both GT-NZVI-and B-GT-NZVIcatalyzed oxidation reactions.

Fig.9.Linear plots of kinetic data calculated by second-order model for GT-NZVI and B-GT-NZVI with various dosages.

3.4.Effects of pH of solutions on degradation of RB 238 dye aqueous media

The influence of initial pH values of solutions on the decolorization efficiency of RB 238 dye was investigated at different pH values ranging from 2 to 9(Fig.10(a)and(b))under the following experimental conditions:c(RB 238 dye)=0.05 mmol/L,c(H2O2)=5 mmol/L,ρ(GTNZVI)=0.5 g/L orρ(B-GT-NZVI)=0.5 g/L,and room temperature.An increase in decolorization efficiency of RB 238 with a decrease in aqueous pH for GT-NZVI and B-GTNZVI was found.This behavior may be attributed to the obstruction of ferrous hydroxide on the nano iron surface when GT-NZVI or B-GT-NZVI is oxidized in solutions with low pH values(Satapanajaru et al.,2011).After 60-min Fenton-like oxidation reactions catalyzed by GT-NZVI and B-GT-NZVI,the best pH value was found to be 2.5 with decolorization efficiencies of 93.6% and 95.1% in RB 238 dye,respectively(see supplementary data).Fig.10(a)and(b)demonstrates that the decolorization rates decreased as the pH value was increased from 2.5 to 9.This indicates that pH has a significant influence on decolorization.The kinetic parameters of the second-order model have shown a high level of adjustment in all experiments,and thek2values at pH values of 2.5,3,3.5,4,5,7,and 9 were 4 824,1 461,720.9,102.8,29.74,11.24,and 2.01 L/(mol.min),respectively,for B-GT-NZVI,and 4 042,1 460,662.1,98.84,30.09,7.636,3.772 L/(mol.min),respectively,for GT-NZVI(Fig.11(a)and(b),and supplementary data).These results show that increasing the pH value from 2.5 to 9 decreased the kinetic rates.Satapanajaru et al.(2011)also found similar decolorization kinetics of azo-dye reactive black 5 and reactive red 198 with NZVI in the pH range of 3-9.They revealed that decolorization significantly enhanced the pH values of acidic solutions.This may result from the fact that the NZVI surface has a positive charge at a low pH value,and the dye molecules with a sulfuric group(SO3)have a negative charge that may attract the NZVI particles.Consequently,the removal reaction between dye molecules and NZVI could be accomplished more easily.Therefore,this study defined a pH value of 2.5 as the optimum solution.

Fig.10.Decolorization efficiencies of RB 238 dye in solutions at various pH levels for GT-NZVI and B-GT-NZVI.

Fig.11.Linear plots of kinetic data calculated by second-order model for GT-NZVI and B-GT-NZVI at various pH levels.

3.5.Effect of RB 238 dye concentration on decolorization efficiency

Fig.12(a)and(b)presents the influence of initial dye concentration on the degradation of RB 238 dye.The experimental conditions are as follows:pH=2.5,c(H2O2)=5 mmol/L,ρ(GT-NZVI)=0.5 g/L orρ(B-GTNZVI)=0.5 g/L,and room temperature.When the initial RB 238 dye concentration increased from 0.05 mmol/L to 0.25 mmol/L for GT-NZVI and B-GT-NZVI,the decolorization efficiency decreased from 93.5% to 75.8% and from 95.1% to 80.6%,respectively(see supplementary data).Fig.13(a)and(b)indicates that the second-order kinetic model fit the experimental data well with high values ofR2.Fig.13(a)and(b)also demonstrates that the rate constant of the second-order model for decolorization efficiency decreased from 4 042 L/(mol.min)to 252.7 L/(mol.min)when the concentration of RB 238 dye was increased from 0.05 mmol/L to 0.25 mmol/L for GT-NZVI.Under the same conditions for B-GT-NZVI,the rate constant decreased from 4 824 L/(mol.min)to 337.9 L/(mol.min).

3.6.Effect of temperature on degradation of RB 238 dye aqueous media

Fig.12.Decolorization efficiencies of RB 238 dye at various concentration levels catalyzed by GT-NZVI and B-GT-NZVI.

Fig.13.Linear plots of kinetic data calculated by second-order model for GT-NZVI and B-GT-NZVI at various concentration levels.

Fig.14.Decolorization efficiencies of RB 238 dye catalyzed by GT-NZVI and B-GT-NZVI at various temperature levels.

The effect of the Fenton-like oxidation reactions at various temperature levels on the decolorization efficiency of RB 238 dye solution(0.05 mmol/L)was investigated by varying the temperature from 20°C to 50°C.The experimental conditions are as follows:pH=2.5,c(H2O2)=5 mmol/L,ρ(GT-NZVI)=0.5 g/L orρ(B-GT-NZVI)=0.5 g/L.It can be seen that the temperature significantly affected the oxidation reaction rate.Fig.14(a)and(b)demonstrates that the temperature increase had a positive impact on the degradation of RB 238 dye.After the 10-min reactions catalyzed by GT-NZVI and B-GT-NZVI,the decolorization efficiency increased from 62.8% to 94.1%and from 80% to 94.6%,respectively,when the temperature was increased from 20°C to 50°C(see supplementary data).This may be attributed to the fact that higher temperature tends to increase the oxidation reaction rate between the catalyst and H2O2,leading to an increase in the production rate of hydroxyl free radicals(.OH)or high-valence iron species(Demirezen et al.,2019).Meanwhile,high temperature can provide reactant molecules with more energy to overcome the energy for reaction activation(Hashemian,2013).However,as the temperature was increased from 30°C to 50°C,the decolorization efficiency of RB 238 dye decreased from 97.5% to 90% and from 97.6% to 93%,respectively,after the 180-min oxidation reactions catalyzed by GT-NZVI and B-GT-NZVI.This phenomenon may be attributed to the decomposition of hydrogen peroxide at high temperatures.Therefore,the best temperature for the degradation of RB 238 dye is between 30°C and 40°C.Fig.15(a)and(b)demonstrates that highR2values were obtained for the second-order kinetic model.This indicates that the second-order model was suitable for describing the experiments.Meanwhile,the apparent activation energy(Ea)of GT-NZVI and B-GT-NZVI for RB 238 dye in water was estimated according to the Arrhenius equation(Nicodemos Ramos et al.,2020).The results show thatEadecreased from 38.22 kJ/mol for GT-NZVI to 14.13 kJ/mol for B-GT-NZVI.This indicates that the effect of B-GT-NZVI on the decrease of the energy barrier is more pronounced than that of GT-NZVI,which leads to improved Fenton-like oxidation reactions(details are shown in supplementary data).This type of adsorption is considered physical adsorption that requires a low activation energy amount of 5-40 kJ/mol(Demirezen et al.,2019).

Fig.15.Linear plots of kinetic data fitted by second-order model for GT-NZVI and B-GT-NZVI at various temperature levels.

3.7.Effects of inorganic ions on degradation of RB 238 dye aqueous media

Inorganic salts are common with dyes in wastewater.Therefore,the effect of the presence of 0.1 mol/L NaCl,Na2CO3,Na2SO4,and NaNO3on the removal of RB 238 dye by Fenton-like oxidation was evaluated.As shown in Table 1,the presence of inorganic anions such as Cl-decreased the decolorization efficiency from 93.6% to 85.3% and from 95.1% to 90% in the experiments catalyzed by GT-NZVI and B-GT-NZVI,respectively(see supplementary data).The influence of the negative behavior of Cl-ions on decolorization efficiency of RB 238 dye may be attributed to the effect of chloride ions that consume the production of the hydroxyl free radicals(Hassan et al.,2019).In general,other anions did not have a significant effect on the removal of RB 238 dye solution.Table 1 also indicates that the decolorization kinetics of RB 238 dye fit the second-order model.

Table 1 Zero-order,first-order,and second-order kinetic model constants,regression coefficients,and decolorization efficiencies after addition of inorganic salt(0.1 mol/L).

3.8.Comparison of decolorization efficiency through using various reaction mechanisms

Fig.16 shows the RB 238 dye removal efficiency in an aqueous solution by various green catalyst adsorbents at the optimumpHof2.5.After reactionfor 180min,thepercentageofRB 238 dye removed by B-GT-NZVI(0.5 g/L)with hydrogen peroxide(5 mmol/L)reached 96.2%,and that for GT-NZVI(0.5 g/L)with hydrogen peroxide(5 mmol/L)was 93.5%.In contrast,those for B-GT-NZVI,GT-NZVI,and bentonite(none with H2O2)were 49.5%,28.2%,and 10.5%,respectively.The difference in removalefficiency indicates thatreactivity of B-GTNZVI and GT-NZVI with and without H2O2varied remarkably.This is probably related to their morphology and reaction mechanisms.The removal efficiency of RB 238 dye using B-GTNZVI was higher than that using GT-NZVI alone.This confirms that the presence of bentonite as a support material caused the dispersion and stabilization of GT-NZVI nanoparticles and further enhanced the reactivity of B-GT-NZVI.These results are in agreement with previous studies on bentonite-supported NZVI(Kerkez et al.,2014).When the Fenton-like oxidation process was conducted,the achieved decolorization rates were significantly high.The removal efficiencies were 96.2% and 93.5%for B-GT-NZVI/H2O2and GT-NZVI/H2O2,respectively.As shown in Fig.16,after a 30-min reaction,the maximal decolorization efficiency was almost achieved.This may be attributed to the oxidative degradation mechanism with B-GTNZVI,which accelerates initial reactions.

Fig.16.Decolorization efficiencies of RB 238 dye using various green catalyst adsorbents under optimum conditions:pH=2.5;c(H2O2)=5 mmol/L;ρ(GT-NZVI)=0.5 g/L,ρ(B-GT-NZVI)=0.5 g/L,orρ(bentonite)=0.5 g/L;and room temperature.

The oxidative degradation mechanism found in this study has been similarly reported by Wu et al.(2015).First,in an acidic medium(pH=2.5),RB 238 dye is adsorbed by the surface of GT-NZVI or B-GT-NZVI.Second,hydroxyl free radicals(.OH)are produced when Fe0on the surface of GTNZVI or B-GT-NZVI reacts with H2O2to form Fe2+.This process speeds up the decomposition of H2O2and generates highly reactive hydroxyl free radicals,when Fe2+is oxidized by H2O2and is transformed into Fe3+.Moreover,Fe3+reacts with H2O2to generate hydroperoxyl radicals(HOO.)that are less reactive than.OH(Chi et al.,2017;Shahwan et al.,2011).At the same time,Fe3+is reduced to Fe2+to continue a cycle for the production of.OH radicals in serious reactions.Finally,.OH attacks the RB 238 dye molecules on the surface of GT-NZVI or B-GT-NZVI and parts of RB 238 dye are mineralized into CO2and H2O.

4.Conclusions

In this study,GT-NZVI and B-GT-NZVI were synthesized to remove the RB 238 dye in aqueous solutions.The B-GTNZVI nanoparticles were produced to contain GT-NZVI at the ratio of 30%.They were used to investigate the removal efficiency of RB 238 dye by coating bentonite with GT-NZVI,for the purposes of reducing the cost,increasing the adsorption capacity of GT-NZVI,and decreasing the optimum amount of GT-NZVI used in the Fenton-like process.This study shows that improvement was achieved when using GTNZVI and B-GT-NZVI to remove the dye solution.Under the optimum conditions,high decolorization efficiencies of RB 238 dye(96.2%and 93.5%,respectively)were obtained by using B-GT-NZVI and GT-NZVI to catalyze the Fenton-like oxidation.

Experimental kinetic data showed that the degradation of RB 238 dye by GT-NZVI or B-GT-NZVI fit a second-order model well.The effect of temperature was investigated.The best temperature for the degradation of RB 238 dye was between 30°C and 40°C,and more than 97%of RB 238 dye was degraded in the Fenton-like oxidation process after 60 min.Meanwhile,the influence of inorganic salts(0.1 mol/L)was investigated.It was found that there was no significant decrease in decolorization efficiency.Finally,a possible mechanism of the oxidative degradation of RB 238 dye using a green Fenton-like system was determined.It demonstrates that GT-NZVI or B-GT-NZVI catalyzes the activation of H2O2to produce.OH radicals,and RB 238 dye is degraded into CO2and H2O.The findings of this research demonstrate that,as a Fenton-like catalyst,B-GT-NZVI has impressive capabilities of removing azo dyes.Future work will focus on evaluating this system by using real water samples from Iraqi textile factories(for instance,Al-Kut Textile Industry).Additionally,future work will analyze and identify the possible pathways of the intermediates in oxidative degradation.

Acknowledgements

The authors are highly indebted to Environment and Water Directorate of Ministry of Science and Technology,Iraq,and the Sanitary and Environmental Branch of the Civil Engineering Department at the University of Technology,Iraq,for providing all the facilities to carry out this work.

Appendix A.Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.wse.2020.12.001.

Declaration of competing interest

The authors declare no conflicts of interest.