Nowadays, numbers of dyes are being widely used in textile, paper, rubber, plastic, leather, cosmetic, pharmaceutical, and food industries, which generate large volumes of wastewater every year  . The disposal of dye wastewater without proper treatment is a big challenge as it causes drastic threat to the aquatic environment owing to their bio- resistance, visibility, toxicity and carcinogenic effects   . Some of the dyes present in wastewater even decompose into carcinogenic aromatic amines under anaerobic conditions and cause serious health problems to human beings as well as other animals  . Therefore, an effective and economic treatment of effluents containing diverse dyes has become a necessity for clean production technology for industries  . Several methods exist for the treatment of colored wastewater. These include: adsorption, coagulation and flocculation, biological treatment, photo catalytic process, advanced oxidation processes etc  . In recent years, advanced oxidation processes (AOPs) become important technologies for water treatment   and, have received considerable attention for the destructive oxidation of dyes, since many aromatic compounds have proven to be degraded effectively to CO2, H2O, and small molecules  . The Fenton oxidation process is one of the AOPs which involve the catalytic oxidation of organics in waste water using hydroxyl radicals generated from the reaction between the iron (II) cations and hydrogen peroxide (H2O2). The advantage of the Fenton oxidation process over other AOPs is that it is economical, convenient to use, produces non-toxic by-products and does not require any sophisticated or expensive instruments  . Despite its strong oxidation capacity, homogeneous Fenton or Fenton-like reaction operation has some critical limitations such as a narrow pH range, post-treatment requirements prior to discharge as a result of iron hydroxide sludge formation, and unattainable regeneration of catalyst which hampers the economic feasibility of this process  . To overcome these drawbacks, some efforts have been made to develop different heterogeneous (photo-) Fenton-like catalysts including the incorporation of iron ions or iron oxides within the structure of catalyst support    . Most of the iron sources studied includes synthesized iron oxide such as (Fe3O4), hematite (α-Fe2O3), maghemite (γ-Fe2O3) and goethite (α-FeOOH)  . However, to the best of our knowledge, little attention has been paid to the catalytic property of iron rich soils. In comparison with other solid catalysts such as natural clays, resin supports and zeolites, iron rich soil possesses higher content in Fe (II) ions. In fact, due to the lack of iron in the composition of the aforementioned catalysts, sophisticated methods such as impregnation, hydrothermal and ion exchange are employed to achieve suitable percentage of Fe (II) ions in the final catalysts    . In this study, we used iron rich soil as a catalyst for the discoloration of Tartrazine (Acid Yellow23, trisodium1-(4-sulfonato- phenyl)-4-(4-sulfonatophenylazo)-5-pyrazolone-3 carboxylate) Tartrazine is a orange- colored, water soluble powder widely used in food products, drugs, cosmetics and pharmaceuticals. The estimated amounts of tartrazine manufactured in 1996 were approximately 71.35 metric tons in Japan and 985.76 metric tons in USA  experimental results suggest that it might adversely affect human health  . The influence of several parameters such as solution pH, the H2O2 amount, catalyst amount, initial dye concentration, and inorganic salts on the heterogeneous Fenton-like oxidation of acid yellow 23 using iron rich soil were investigate. Also, the pseudo-first order rate constant (kapp) was calculated by a kinetic model (based on non-linear regression analysis) as a function of initial Acid Yellow 23 concentration.
2. Experimental Study
The acid yellow 23 dye and the hydrogen peroxide solution (50% in mass) were purchased from Merck (Germany) and used without any purification. The absorption spectra of acid yellow are characterized by two main bands, one in the visible region (λ max = 420 nm) which is responsible for the chromophoric components. In this study, the decay in the band at 420 nm was investigated as a function of time as a measure of the discoloration degree. All aqueous solutions were prepared with distilled water.
2.2. Characterization of Natural Iron Oxide
To determine the mineralogical phases of the iron rich soil sample, XRD characterization was carried out at room temperature using a Shimadzu 6000 spectrometer operating on a Cu-Kα radiation. The surface morphology and the structure of the iron rich soil were obtained using a scanning electron microscopy (SEM) (FEIQuanta 250 FEG SEM)) at an acceleration voltage of 10 kV. Nitrogen sorption analyses were obtained with a sorptometer (Micrometrics, ASAP-2010). The surface area was calculated according to the BET method.
2.3. Sorption Experiments
All equilibrium sorption experiments were conducted at pH = 2.5. The iron rich samples were mixed with variable concentrations of acid yellow 23. The equilibrium of Acid Yellow 23 adsorption was achieved within 40 minutes. After equilibrium, the samples were filtered and analyzed by UV-visible spectroscopy and adsorbed amount of the dye were calculated according to Equation (1).
Qe (mg/g) is the sorption capacity at equilibrium, Co (mg/L) is the initial and Ce (mg/L) is the acid yellow 23 concentration at equilibrium. V (L) is the volume of solution, m (g) is the mass of iron rich soil. The effect of pH on adsorption of acid yellow 23 on iron rich soil were performed with a single initial dye concentration in a 20 mL closed reaction vessel at a constant temperature. The pH was controlled by adding aliquots 0.1 M of HCl or NaOH to the stock solid suspensions. The Acid Yellow 23 residual concentration in solution was determined by measuring its absorbance using a UV-Vis spectrophotometer (PerkinElmer), with a 1 cm quartz cell.
2.4. Discoloration of Azo-Dye Acid Yellow
The experiments were carried out in a conical flask (containing 100 mL of reaction solution). The dosage of catalyst was 1.0 - 3.0 g∙L−1 while the concentrations of Acid yellow and H2O2 were 20 - 50 m∙mol∙L−1 and 4 - 24 m∙mol∙L−1, respectively. All the experiments were carried out under constant stirring. The pH of solution was adjusted by 0.1 M H2SO4 and NaOH. Before degradation reaction, the suspension containing catalyst and Acid Orange II was stirred for 40 min to achieve adsorption equilibrium. Then the degradation reaction was initiated by adding H2O2 in the Acid yellow solution. Then the degradation reaction was initiated by adding H2O2 into Acid yellow solution. Equation (2) was used to find the discoloration efficiency (DE(%))
where, Co (mg/L) is the initial concentration of Acid yellow 23 and Ct (mg/L) is the concentration of acid yellow 23 at reaction t (min).
3. Results and Discussion
3.1. Characterization of Catalysts
Figure 1 shows the XRD peaks of the iron rich soil, the peaks at d-spacing of 1.70 A°; 2.71 Å; .2.52 Å; 2.21 Å; 1.84 Å; 1.70 Å are associated with hematite phase according to the standard powder diffraction data (JCPDS). Also, the XRD peaks at d of 4.20, and 2, 51 Å are related to the goethite phase. In addition, a broad band between 05 and 20˚ observed in the XRD patterns indicates the presence of amorphous materials. So the existence of goethite (FeO(OH)) and hematite (Fe2O3), proved that the iron rich soil contains iron oxides and oxyhydroxides
Scanning electron microscopy (SEM) coupled with energy dispersive spectroscopy (EDS) that provides a qualitative elemental analysis of the material are presented on Figure 2. On the SEM micrograph, particles observed are spherical and of irregular shape. Energy dispersive X-ray elemental analysis depicts a composition characteristic of iron oxide/oxyhydroxide. The elemental composition that is presented in Figure 2 is closed to goethite composition (ideally, iron content in goethite is 63% and here the analysis indicates about 55% iron).
The surface area, pore diameter, and pore volume of soil sample were characterized by N2 adsorption-desorption. The N2 adsorption-desorption isotherms and pore size distribution curves of iron rich soil are shown in Figure 3. The shape of the isotherms seems to be nearly type IV isotherm according to the IUPAC classification. The shape of the hysteresis loops is of type H3, indicating narrow slit-shaped pores that is generally
Figure 1. XRD spectra of natural iron rich soil used in this study.
Figure 2. Scanning electron microscopy (SEM) coupled with energy dispersive spectroscopy (EDS) patterns for iron rich soil.
Figure 3. Adsorption-desorption isotherm of N2. The inset shows pore size distribution of iron rich soil sample.
associated with plate-like particles  .
The BET surface area, total pore volume and average pore size of the iron rich soil are summarized in Table 1.
Table 1. Physicochemical properties of iron rich soil.
3.2. Adsorption of Acid Yellow on Natural Iron Oxide
Adsorption Isotherms were determined to assess acid yellow distribution between solid and aqueous phases (Figure 4).
The experimental isotherm data were fitted to the Langmuir and Freundlich equations, applying linear regression analysis. The best fit was observed with the Langmuir model. The linear form of the Langmuir equation is given by Equation (3).
where, is the equilibrium concentration (mg∙L−1), is the amount of adsorbate adsorbed per unit mass of the adsorbent at equilibrium (mg/g), k is the Langmuir adsorption constant (L∙mg−1) and is the maximum amount of acid yellow 23 adsorb per unit mass of adsorbent to form a complete monolayer on the surface (mg/g). The Langmuir isotherm constants obtained by plotting Ce/Qe versus Ce are = 6.329 mg/g and KL = 0.313 L∙mg−1.
3.3. Effect of pH Dose on Discoloration Rate of Acid Yellow
The influence of initial pH of the dye solution on the discoloration of acid yellow efficiency was studied using five solutions with initial pH of 2.0, 2.5, 3.0, 3.5 and 4.0 and without any modifications or control of pH during the process. Figure 5 shows the discoloration of dye as a function of the initial pH of the solution at various reaction times.
The obtained results indicated that the discoloration of acid yellow 23 was significantly influenced by the solution pH. The increase of the initial pH led to a much lower degradation rate. It has been confirmed that low pH promotes the stability of hydrogen peroxide and slowed its decomposition   . Therefore, the decrease of pH would slow the decomposition of H2O2, which may slow the discoloration process. However, in this study the decrease of initial pH has led to a much larger degradation rate. This can be attributed to two aspects: Firstly, possible leaching of Fe from the catalyst at pH slightly acid contributing to homogeneous Fenton reaction. Secondly, the electrostatic interactions between the acid yellow 23 molecules and the iron rich soil surface at pH  . Iron concentration in reaction solutions were measured at pH 2.5 and 3.0 and the value were 4.58 mg/L and 1.15 mg/L respectively. This can justify why the discoloration efficiency at pH 2.5 is higher than the discoloration at pH 3.0.
In order to verify the second hypothesis, the sorption of acid yellow 23 on the iron rich soil was investigated by varying the pH from 2 to 10, using the same experimental conditions as for the oxidation. The results (Figure 6), showed that the adsorption quantity decrease when the pH increase. This imply that the electrostatic interactions
Figure 4. Adsorption isotherm of yellow dye onto iron rich soil.
Figure 5. Effect of pH on the discoloration efficiency of acid yellow. Experimental conditions: [acid yellow 23] = 30 mg∙L−1, [H2O2] = 16 mM, iron rich soil = 2.50 g∙L−1.
between partial negative charges on the acid yellow 23 molecule and the positive Fe-OH2+ surface groups on iron rich soil that increased as pH decreased  (PZC > 4.2, Table 1) favorite the discoloration of acid yellow 23. Consequently, both the amount of iron leaching and the surface reaction contributed to the discoloration of acid yellow 23  .
3.4. Effect of Hydrogen Peroxide Concentration
The effect of H2O2 concentration was investigated in the range of 4 to 24 m∙mol∙L−1. The results obtained are presented graphically in Figure 7. When the concentration of H2O2 was 4 mmol∙L−1, the discoloration process was slow due to the insufficient •OH in aqueous solution. As the concentration of H2O2 increased to 24 m∙mol∙L−1, the process
Figure 6. Effect of initial pH on the adsorption acid yellow. Experimental conditions: [acid yellow 23] = 30 mg∙L−1, iron rich soil = 2.50 g∙L−1.
Figure 7. Effect of hydrogen peroxide concentration on the discoloration efficiency of acid yellow. Experimental conditions: [acid yellow 23] = 30 mg∙L−1, pH = 2.5, Iron rich soil = 2.50 g∙L−1.
was significantly accelerated, because more radicals were formed.
3.5. Effect of Iron Rich Soil Dosage on the Discoloration of Acid Yellow 23
The effect of iron rich soil dosage on the discoloration of acid yellow 23 was investigated in the presence of different amounts of iron rich soil in solution (1, 1.5, 2, 2.5 and 3.0 g∙L−1), the results are depicted in Figure 8. The results indicate that when the amount of iron rich soil in solution increases from 1.0 to 2.0 g/L, the discoloration efficiency also increases from 25.12% to 97.74%. This can be ascribed to two causes; it provides additional surface area for adsorption and also additional amount of Fe for the formation of OH radicals. Further increase in the iron rich soil amount slightly decreases the discoloration efficiency, which could be ascribed to the inhibition effect caused by excess iron ions in the heterogeneous Fenton process that act as scavengers as shown by Equation (4)  .
Figure 8. Effect of Iron rich soil dosage on the discoloration efficiency of acid yellow 23. Experimental conditions: [acid yellow 23] = 30 mg∙L−1, [H2O2] = 16 mM, pH = 2.5.
3.6. Influence of Initial Concentration of Acid Yellow 23 on Discoloration
Previous studies have shown that the effectiveness of Fenton process for the degradation of organic compounds depends on the initial concentration of the pollutant For example, S. Karthikeyan et al. 2013  , have shown during the treatment of synthetic phenol solution by Fenton-like process using natural Fe-Co-Al trimetal oxide (TMO) as a heterogeneous catalyst that the process was more effective for diluted solutions more than for concentrated ones. It can be seen in Figure 9(a) that by increasing the concentration of Acid yellow from 20 to 50 mg L−1, the removal of acid yellow 23 decreased from 98.71% to 65.06% during 140 min of the reaction. There are two main reasons for explained the behavior. First, by raising the concentration of acid yellow 23, high amount of acid yellow 23 molecules would adsorb on the surface of iron rich soil. In this way, more and more hydroxyl radical (OH) for treatment of solution are needed. In other words, the amount of reactive radicals is not enough to oxidize the excessive concentration of Acid yellow due to constant rate of hydroxyl radical formation on the iron rich soil for various concentrations of acid yellow. Second, increasing the concentration of Acid yellow 23 leads to formation of various degradation intermediates or byproducts that may adsorb on the iron rich soil surface and deactivate the active sites of the iron rich soil. Consequently, the removal efficiency plunges as the concentration increases    .
In order to obtain the kinetic information, the experimental results were fit with a pseudo-first-order Equation (5).
where, Co (mg/L) is the initial concentration of Acid yellow 23 and Ct (mg/L) is the concentration of acid yellow 23 at reaction t (min) and Kapp reaction rate of the pseudo-first-order kinetic model.
As it can be seen in Figure 9(b), the reaction rate of the pseudo-first-order kinetic model (kapp) is decreased from 0.02 to 0.01 min−1 with increasing initial acid yellow concentration from 20 to 50 mg∙L−1. The heterogeneous reaction of the degradation of organics encompasses with both oxidation and adsorption that occurs simultaneously  , and can be described by the Langmuir Hinshelwood mechanism   , which can be expressed as follows:
Figure 9. (a) Effect of acid yellow concentration on the discoloration efficiency of acid yellow. Experimental conditions: [H2O2]= 16 mM, pH = 2.5 Iron rich soil = 2.50 g∙L−1. (b) Variation of apparent kinetic constant at different acid yellow concentrations.
Ks is the intrinsic surface reaction rate constant (mg∙L−1∙min−1) and Kr is the Langmuir-Hinshelwood adsorption equilibrium constant (L∙mg−1). The values of Ks and Kr can be calculated by plotting the (1/Kapp) against [Dye]. As can be seen in Figure 10, the experimental data fits reasonably well (R2 = 0.98) with the proposed Langmuir-Hinshelwood mechanism. Hence, the catalytic surface reactions between the iron rich soil active sites, acid yellow 23 and •OH radicals can be used in determining the kinetics of the heterogeneous reaction.
The Ks and Kr values were obtained as 0.579 mg∙L−1∙min−1 and 0.132 L∙mg−1, respectively. The Kr value was found to be lower than the sorption constant (KL = 0.313 L∙mg−1) in absence of an oxidant according to the Langmuir adsorption model as shown in Figure 5 indicating the existence of competitive adsorption between acid yellow 23 and H2O2 towards the active sites during the reaction  .
3.7. Spectral Changes of Acid Yellow 23 during Discoloration Process
The changes in the absorption spectra of acid yellow solution during the discoloration process at different reaction times are shown in Figure 11. It appears from the spectra that, before the treatment, the UV-Vis spectrum of acid yellow 23 is characterized by one main band at 420 nm in the visible region. This band is related to the color of acid yellow solution (n → π* transition in N=N group). Upon treatment it is observed that the adsorption band decreased gradually, indicating that the N=N bond and the conjugated π* system were completely destroyed  .
3.8. Effect of Hydroxyl Radical Scavengers on the Discoloration of Acid Yellow 23
Chloride, sulfate and nitrate are regular inorganic anions in industrial effluents and may influence the treatment efficiency, which is called the salting-out effect. Figure 12 indicates that the presence of these chemicals in the dye solutions decline the acid
Figure 10. Determination of the adsorption equilibrium constant, Ks, and the second order rate constant, Kr, for the Langmuir-Hinshelwood model. Experimental conditions: [H2O2] = 16 mM, pH = 2.5 iron rich soil = 2.5 g∙L−1.
Figure 11. UV-Vis spectral changes of acid yellow. Experimental conditions: [acid yellow 23] = 30 mg∙L−1, [H2O2] = 16 mM, pH 2.5 iron rich soil = 2 g∙L−1.
Figure 12. Effect of inorganic salts on the discoloration of acid yellow. Experimental conditions: [acid yellow 23] = 30 mg∙L−1, [H2O2] = 16 mM, pH 2.5 iron rich soil = 2 g∙L−1. [salt]=1 g/L.
yellow 23 discoloration efficiency. The degradation rates was highly affected in the presence of chlorides, this may be due to several factors: 1) the decrease in H2O2 decomposition rate and so decrease in hydroxyl radical generation rate, 2) the decrease in acid adsorption rate because of the competition with Cl− or 3) the scavenging of •OH by chloride and the formation of reactive radicals with low oxidation potential rather than −OH by the Equations (8)-(10)   : Furthermore, the generated •Cl radicals have a great tendency to react with H2O2, which declines •OH formation (Equation (11))  .
Moreover, Sulfate ions and nitrate ions can also react with •OH and lead to the formation of and NO3• radicals respectively with low oxidation potential (Equation (12) Equation (13))   . This decreases the acid yellow 23 discoloration efficiency
3.9. Application to Real Water Sample
The efficiency of iron rich soil as catalyst for the discoloration of acid yellow in real water was investigated. A sample of river water was collected and was spiked with 30 mg/L of acid yellow. Some of the characteristic of the river water are DCO = 125 mg/L, pH = 4.4. The discoloration procedure was the same as the one described in the section 2.3It appears that the discoloration percentage of acid yellow decreases from 97.75% to 75.73% respectively for distilled water and for river water for identical operating conditions. This decrease can be explained by the scavenging effect of salt present in the river water.
3.10. Stability of the Catalyst
The stability of the catalysts under the H2O2 is a crucial point for the application of Fenton’s systems in oxidation processes. The data presented in Figure 13 showed the stability and activity of iron rich soil in recycling tests. After each operation cycle, the
Figure 13. Discoloration efficiencies for three successive run. Experimental conditions: [Acid yellow 23] = 30 mg/L, [H2O2] = 16 mM, pH = 2.5 and Iron rich soil = 2 g∙L−1.
catalyst was separated, washed with H2O and used in the next run. The discoloration of acid yellow after 140 min in the 3 cycles considered were 97.75%, 96.35% and 95.10%, respectively. The slight activity decay observed may be due to the Fe leaching from iron rich soil. This behavior shows that Iron rich soil can be utilized as a promising and stable catalyst in heterogeneous Fenton process for the degradation of organic pollutants with great reusability potential.
In this study, the heterogeneous Fenton-like oxidation of acid yellow dye was investigated in the presence of an iron-rich soil catalyst. The iron-rich soil catalyst acted as an efficient and stable catalyst for this process. 97% discoloration yield was achieved after 140 minutes of reaction with 16 mM H2O2, 2.5 g/L of iron rich soil at pH = 2.5 for 30 g/L acid yellow 23 dye solution. Both homogeneous and heterogeneous reactions contribute to the degradation process. The heterogeneous Fenton-like discoloration of acid yellow followed the pseudo-first order kinetics and explained in terms of the Langmuir-Hinshelwood kinetic model. Iron rich soil exhibits stable performance after three cycles. In all of the experiments, concentration of the iron leaching was below to the standard of Environmental Quality (Sewage and Industrial Effluents) Regulation 1979 (<5 mg/dm3).
The authors gratefully acknowledge Dr. Issac NONGWE BEAS, Department of Civil and Chemical Engineering UNISA, RSA for recording the BET of the samples and also to Dr. TIYA Antoine for recording the XRD of the samples.