Fe(III) Oxide-modified Indonesian Bentonite for Catalytic Photodegradation of Phenol in Water

Phenol, which is a major organic pollutant, is usually detected in industrial wastewater, and thus the wastewater should be processed further before discharged into water bodies. Application of heterogeneous catalysis using natural-based materials is known to be effective and environmentally friendly in removing hazardous substances in water. In this study, local natural bentonite from the Tapanuli region in Indonesia was modified to eliminate dissolved phenol. Elimination by photodegradation reaction was conducted in a photo-Fenton system utilizing Fe(III) oxide-modified bentonite (Fe-B) as catalyst. Fe-B was prepared by a cation exchanging process using mixture solutions of NaOH and FeCl3 with OH/Fe molar ratio of 2:1 and calcined at 300 °C. Material characterization was performed by X-ray diffraction (XRD), low-angle XRD, Fourier transform infrared spectroscopy and atomic absorption spectroscopy. The reaction components consisted of ultraviolet C light, H2O2, and Fe-B, and they were processed in a batch reactor. The role of each component was analyzed by a series of reaction conditions (i.e., adsorption, photolysis, H2O2 effect, Fenton, and homogeneous photo-Fenton). The heterogeneous photo-Fenton system was found to be essential for phenol degradation, as none of the reaction conditions caused total phenol removal in the 180 min reaction time. To conclude, heterogeneous photo-Fenton gave the highest photodegradation activity, and the best experimental condition for 1.10 mM phenol removal was 5 g L catalyst, 78.35 mM H2O2, and 90 min reaction time.


Introduction
Industries are important pollution sources, and the discharged wastewaters may contain potentially harmful organic compounds. As a major organic pollutant, phenol is usually detected in various industrial wastewaters, namely, in petrochemical, tanneries, pulp and paper, chemical, and pharmaceutical industries [1]. As stated in Phenol is harmful for both water organisms and human health [3]. Biodegradability of phenol is only 90% in surface water after seven days, and its aquatic toxicity (LC50) is 12 mg L -1 (Daphnia magna, 48 h). It is also classified in chronic health hazards as a teratogenic and carcinogenic agent [4][5]. For this reason, the maximum phenol concentration for safe consumption of drinking water is limited to 1 µg L -1 in Indonesia [6]. As phenol disposal into freshwater ecosystem without further treatment is dangerous for the environment, extensive research on phenolic wastewater treatment becomes exceptionally important.
The reaction of photo-Fenton has been highlighted as one of the most promising processes for wastewater treatments; reaction is classified as part of advanced oxidation processes [7]. The Fenton reagent consists of Fe catalyst and H 2 O 2 , and additional ultraviolet (UV) light radiation is added to trigger fast radical production [8][9]. The photo-Fenton reaction has two important steps: photoreduction of Fe 3+ into Fe 2+ and reoxidation of Fe 2+ into Fe 3+ by H 2 O 2 . The high reactivity of the resultant hydroxyl radical (OH) and hydroperoxyl radical (OOH) can oxidize and degrade organic pollutants into environmentally friendly products or even mineralize pollutants completely to produce CO 2 and H 2 O [10][11]. The application of solid support to produce heterogeneous Fe catalyst in the photo-Fenton reaction can minimize the usage of Fe 2+ /Fe 3+ , thus resulting in high reproducibility and less additional Fe pollution after treatment [12].
The heterogeneous catalyst in the photo-Fenton reaction can be prepared by loading the active species to various materials, such as bentonite, sepiolite, hydrocalcite, zeolite, and mesoporous silica [13]. Bentonite has been proved to be one of the most promising support materials for catalyst because of its unique characteristics, great abundance, and cost efficiency [14]. In Indonesia, the abundance of natural bentonite as a mining resource is immense. Bentonite can be found in many places in Indonesia, especially in Sumatera, Java, Kalimantan, and Sulawesi. Unfortunately, the application of local bentonite in Indonesia is still limited compared with imported bentonite, with the national bentonite demand remaining at a ±20% deficit in 2010 [15].
Bentonite is produced by volcanic ash weathering and generally exists in form of Ca-bentonite (Ca-B), the cations of which can be exchanged to become Nabentonite (Na-B). When Na-B comes in contact with water, it may swell several times from its initial volume. The main constituent mineral of bentonite is montmorillonite, a group of smectitic clay composed of repetitions of 2:1 clay layers. Each layer consists of one Al 3+ octahedral sheet sandwiched between two Si 4+ tetrahedral sheets [16]. The substitution of Al 3+ or Fe 3+ to Si 4+ in the tetrahedral sheet and that of Mg 2+ , Zn 2+ , and Fe 2+ to Al 3+ in the octahedral sheet generate negative charges on the layer surfaces, which are neutralized by interlayer cations (e.g., K + , Na + , Ca 2+ , or Mg 2+ . The modification process, namely, ion exchange in the interlayer of montmorillonite can be established to modify bentonite as a heterogeneous catalyst [12]. The modification of bentonite as a heterogeneous catalyst for the photo-Fenton reaction can be made by the intercalation of Fe polycations using the cation exchange process and the resultant Fe-hydroxyl bentonite. By giving different treatments to Fe-bentonite, dissimilar kinds of Fe-modified bentonites can be produced. A comparative study by Chen and Zu (2009) [17] on the three types of Fe-modified bentonites, namely, Fe-hydroxyl-modified bentonite, α-Fe 2 O 3 -modified bentonite, and α-FeOOHmodified bentonite, concluded that Fe-hydroxyl-modified bentonite and α-Fe 2 O 3 -modified bentonite have a higher catalytic activity in the photo-Fenton system than the α-FeOOH-modified bentonite.
The aims of this study are (1) to prepare Fe(III) oxidemodified bentonite (Fe-B) from local natural bentonite using the cation exchange process, (2) to perform material characterization of the initial and modified bentonite, and (3) to assess the applicability of Fe-B as a heterogeneous catalyst for the photo-Fenton reaction to phenol removal in an aqueous solution. Before modifying the natural bentonite, pre-treatment, which involved the sedimentation and conversion from Ca-B to Na-B, was conducted to remove impurities and to create a desirable swelling bentonite. Material characterization for natural and modified bentonites was performed using X-ray diffraction (XRD), low-angle XRD, Fourier transform infrared (FTIR) spectroscopy, and atomic absorption spectroscopy (AAS). Phenol removal percentage at time intervals during reaction was recorded by an ultraviolet-visible (UV-vis) spectrophotometer. To ensure the effectiveness of Fe-B as a heterogeneous catalyst of a photo-Fenton reaction, seven related reaction conditions were also performed. Material characterization. The XRD pattern measurement was taken on a Shimadzu XRD 7000 using Ni-filtered Cu Kα radiation (λ = 0.154 nm) as source (voltage 40 kV; current intensity 30 mA). Scan range of 2°-50°and scan rate of 2 (2θ)/min were applied for the measurement.

Experiment
The low-angle XRD patterns were measured by Bruker D8 Advance Powder Diffractometer with a scan range of 2°-15°. The FTIR spectra were recorded on a Shimadzu IR Prestige-21 spectrometer. The specimens were prepared by mixing the powder and KBr until the ratio of 1:10 was reached. The mixture was pressed into a pellet. An average of 100 scans was collected for each of measurement at a wavelength range of 300 cm -1 -4000 cm -1 and a resolution of 4 cm -1 .
Catalytic activity. To observe the catalytic activity of modified bentonites, the photocatalytic Fenton reaction to phenol was conducted inside a solarbox equipped with an ultraviolet C (UV-C) light (Phillips Unilux 15 W, λ max = 245 nm) and a magnetic stirrer. In a typical procedure, a 50 mL beaker glass was filled with a mixture of 1.10 mM (100 ppm) phenol solution and 5 g L -1 catalyst stirred in dark condition for 1 min. The reaction began when the desired amount of H 2 O 2 was added into the beaker and the UV-C lamp was turned on. Then, the reaction mixture was stirred for 180 min and sampling was performed for every 30 min. The UVvis spectra of the sampling solutions were recorded on a Shimadzu UV-2450 spectrophotometer at a wavelength range of 200-400 nm. Through this result, phenol concentration was determined at λ max = 269 nm, which was the characteristic absorption of phenol. The concentration was shown in the following phenol removal percentage: To ensure that the reaction occurred was based on Fenton photodegradation, seven related reaction conditions were also performed to compare the effect of different reactions   Although the XRD patterns of Ca-SB and Ca-B are similar, the sedimentation process was found essential in removing visible bentonite impurities, such as amorphous carbon and calcite. Sharper diffraction peaks of montmorillonite, quartz, and dolomite in the Na-B diffraction pattern compared with Ca-B indicate that the sedimentation and cation exchange processes significantly increase the crystallinity of the bentonite structure. The higher dspacing in Ca-B and Ca-SB (15.96 Å and 15.86 Å, respectively) than in Na-B (15.31 Å) may be explained by the presence of a calcium-rich sample, which is reduced after the preparation of Na-B (Figure 1 [B]) [22].  [29]. Therefore, a uniform pillar structure is not produced and adelaminated structure is   [21,30].

Figure 1. [A] X-ray and [B] low-angle X-ray Diffraction Pattern of Ca-B, Ca-SB, Na-B, Fe-B [105], and Fe-B [300]. Different Symbols in [A] are Used to Represent the Occurrence of Montmorillonite (•), Quartz (○), Dolomite (□), and Hematite/α-Fe 2 O 3 (Δ), and the Numbers in [B] Indicate the d-spacing of Each Sample
In the Ca-SB and Na-B spectra, the weak vibration bands from quartz, silica, and dolomite suggest a decreasing amount of impurities after sedimentation.
Note that vibration bands from Fe generally overlap with the characteristic bands of montmorillonite; thus, the presence of Fe in bentonite is not resolved in the FTIR spectra, especially for the Fe-B [105] and Fe-B[300] spectra [28]. However, for both of the spectrum, low intensities are observed in the -OH bending vibration band from the structural hydroxyl groups in octahedral sheets at 3629 cm -1 and in the Si-O bending vibration band from tetrahedral sheets at 474 cm -1 . These observations may reflect the decreasing amount of free hydroxyl groups in the montmorillonite structure, and they support the XRD analysis than a high concentration of Fe was intercalated into the montmorillonite interlayer spaces.
Catalytic activity. To assess the catalytic activity of Fe-B, a photo-Fenton reaction was performed to degrade phenol. UV-C light was used as the irradiation source because, unlike UV-A, it promotes two important reactions (i.e., photoreduction of Fe(III) into Fe(II) and direct decomposition of H 2 O 2 to produce OH radicals [10]) in an organic compound degradation simultaneously. According to Iurascu et al. (2009) [31], the use of an irradiation source with a long wavelength, such as UV-A, only induces the photoreduction of Fe(III), thereby resulting in the decreased ability of system photo-degradation. For this reason, UV-C light was selected as the irradiation source for the photo-Fenton reaction in this experiment.
Based on the phenol calibration curve, the UV-vis spectra of phenol in various concentrations give two characteristic absorption peaks at 209 and 269 nm.
Although the maximum absorption of phenol at 209 nm is higher than at 269 nm, absorbance at 209 nm is likely to shift at high concentrations. Therefore, the calibration curve of phenol is determined based on the maximum absorption at 269 nm. The existence of both absorption bands is associated with the π → π * electronic transition from the aromatic ring of phenol [32]. When a photodegradation reaction occurs, the absorption bands decrease, thereby indicating the reduction of phenol concentration over the time. Partial phenol degradation produces diverse intermediates in the form of carboxylic acids, and total degradation produce CO 2 and H 2 O as its complete mineralization products [14]. (3) Note that the hydroperoxyl radical (HO 2 ·) is less reactive than the hydroxyl radical (HO·). Therefore, an increased H 2 O 2 concentration shows a diminishing return in the reaction rate [33,34].
The photo-Fenton reaction to 1.10 mM phenol with four different H 2 O 2 concentrations is shown in Figure 3. At a concentration of 39.18 mM, %phenol removal is only 60.43% after 180 min of reaction, thus indicating an inadequate H 2 O 2 concentration given to the system. At a concentration of 78.35 mM, 100% phenol removal is achieved after 90 min.
At a concentration of 117.53 mM and 156.71 mM, %phenol removal is 98.82% and 37.27%, respectively, after 180 min. This result shows that the HO• radical scavenging reaction is likely to occur at a higher concentration (Eqs. (2) and (3) As shown in the reaction condition curve in Figure 4, the observation of phenol adsorption (non-catalytic reaction) using Ca-SB and Fe-B [300] after 180 min shows that phenol can be absorbed into bentonite by as much as 44.71% and 38.71%, respectively. In the H 2 O 2 effect, which is the Fenton reaction without H 2 O 2 , only 29.98% of phenol is removed. This result shows that a strong oxidant has an important role in phenol photodegradation. Moreover, in photolysis, in which the phenol solution is irradiated only with UV-C light, the removal percentage is even lower than that in the H 2 O 2 effect at only 12.35%. UV-C irradiation without the presence of H 2 O 2 and either heterogeneous or homogeneous Fe catalyst cannot form HO•; thus, phenol degradation cannot be conducted effectively. A small percentage of phenol removal in the H 2 O 2 effect and photolysis are assumed mainly from the adsorption process, not from the oxidation-photodegradation reaction, because it cannot produce radical species. In homogeneous photo-Fenton, where Fe 3+ solution is added as the substitute for Fe-B [300] with the same Fe concentration, %phenol removal is even lower at only 2.67%. The low percentage is mainly caused by the increased turbidity of the reaction mixture after the addition of the Fe 3+ solution, which acts as a coagulant. For this reason, only a small part of the emitted UV-C light is scattered, unable to be absorbed by the reaction mixture and eventually unable to induce a photo-Fenton reaction [31].
A better result is presented in the photo-Fenton reaction using Ca-SB, in which %phenol removal is found to be relatively high at 69.93% after 180 min. A photo-Fenton reaction may occur because Ca-SB naturally has a significant iron concentration ( The existence of acidic Fenton photodegradation products, which are carboxylic acids, is verified by pH observation before and after reaction. The results are presented in Table 3. By comparing the initial and final solution pH for both of reactions, the solution pH decreased more in the photo-Fenton reaction than in the adsorption reaction, in which no photodegradation occurs. The higher ∆pH observed in the photo-Fenton reaction (2.88) than in the adsorption reaction (1.93) may have originated from the formation of carboxylic acids that have not completely  , the possible intermediates are muconic, maleic, succinic, malonic, oxalic, formic, and acetic acids in the photodegradation of phenol or phenolic compounds in general. The complete mineralization of phenol produces H 2 O and CO 2 as the end products. CO 2 is also considered to give acidic properties to the final solution because CO 2 is classified as a Lewis acid.

Conclusion
In this study, Fe-B is successfully prepared by the cation exchange process using natural Ca-B from Tapanuli, North Sumatera, Indonesia. The modifying solution with Fe 3+ concentration exceeds the CEC and high OH to Fe molar ratio (2:1) results in Fe(III) oxide pillareddelaminated bentonite. This structure is proved by the Fe-B XRD pattern that demonstrates a smectite peak broadening at 2θ =2-8°.
The application of Fe-B as a heterogeneous catalyst in the Fenton photodegradation reaction with the addition of H 2 O 2 as a strong oxidant can perform total phenol removal in the solution. The decrease in pH in the phenol solution after a photo-Fenton reaction indicates the existence of carboxylic acid intermediates that have not completely degraded. The assessment of reaction conditions indicates that each component in the heterogeneous photo-Fenton system has an important role to perform for the effective phenol removal in the solution.
We hope that our study findings will be applied as a promising alternative for industrial wastewater treatment in Indonesia.