Iron(3) oxide-based nanoparticles as catalysts in advanced organic aqueous oxidation

Transcription

1 Available at journal homepage: Iron(3) oxide-based nanoparticles as catalysts in advanced organic aqueous oxidation Grigory Zelmanov, Raphael Semiat Grand Water Research Institute, Rabin Desalination Laboratory Wolfson Faculty of Chemical Engineering Technion, Israel Institute of Technology Technion City, Haifa 32000, Israel article info Article history: Received 23 April 2007 Received in revised form 7 July 2007 Accepted 24 July 2007 Available online August 2007 Keywords: Fenton process Water treatment Organic pollutants TOC AOP Ethylene glycol Phenol Nanocatalyst Iron oxide Reaction rate abstract Water contaminated with dissolved organic matter is an important issue to resolve for allpurpose uses. The catalytic behavior of iron-based nanocatalysts was investigated for the treatment of contaminated water in the advanced chemical oxidation process. In this study, typical organic contaminants, such as ethylene glycol and phenol, were chosen to simulate common contaminants. It was shown that the two substances are efficiently destroyed by the Fenton-like reaction using iron(3) oxide-based nanocatalysts in the presence of hydrogen peroxide without the need for UV or visible radiation sources at room temperature. A strong effect of nanocatalyst concentration on reaction rate was shown. The kinetic reaction was found and the reaction rate coefficient k was calculated. & 2007 Elsevier Ltd. All rights reserved.. Introduction Organic pollutants are often present in drinking water, groundwater, and domestic and industrial wastewaters. A water treatment based on the chemical oxidation of organic compounds by advanced oxidation processes (AOPs) that is useful for purifying drinking water, groundwater and for cleaning industrial wastewater has been reported recently (Sigman et al., 997; Yeber et al., 2000; Perez et al., 2002). Several of these studies have focused on using these systems as a pre-treatment for biological systems when the dissolved organic matter is toxic, inhibitory or recalcitrant to microorganisms. The degradation and mineralization of organic pollutants in wastewater by AOPs is based on the generation of a very reactive free hydroxyl radical (OH*). This radical is highly reactive, non-selective and may be used to degrade a wide range of organic pollutants. It reacts with most organic compounds by forming to a double bond or by abstracting hydrogen atoms from organic molecules (Safarzadeh-Amiri et al., 997; Buxton et al., 988). The resulting organic radicals then react with oxygen, which leads to the complete mineralization of CO 2, H 2 O and mineral acids (Oliveros et al., 997; Neyens and Baeyens, 2003). Fenton and Fentonlike systems (Fe +2 /Fe +3 /H 2 O 2 ) are often used for industrial water treatment based on AOPs (Fenton, 894; Neyens and Baeyens, 2003). The degradation rate of organic pollutants with Fenton reagents strongly depends on irradiation with ultraviolet (UV) light, and increases with increased UV irradiation intensity (Safarzadeh-Amiri et al., 996). The use of UV light results in a significant increase in the cost of Corresponding author. Tel./fax: address: cesemiat@tx.technion.ac.il (R. Semiat) /$ - see front matter & 2007 Elsevier Ltd. All rights reserved. doi:.6/j.watres

2 493 Notation AOP advanced oxidation process TOC total organic carbon UV ultraviolet C concentration of organic pollutants (ethylene glycol or phenol) in solution, ppm initial concentration of organic pollutants, ppm C Fe nanocatalyst concentration, ppm C H2 O 2 initial hydrogen peroxide concentration EG ethylene glycol k m R r t t /2 a b reaction rate coefficient, /min order of reaction initial ratio of organic solute to hydrogen peroxide concentration ethylene glycol to hydrogen peroxide initial concentrations ratio, ppm/ppm time, min half-life time, the time required to decrease the initial concentration by half constant, min constant, dimensionless industrial water treatment. Post-treatment requires the elimination of Fenton reagents as colloidal precipitates. The separation of colloidal precipitates requires the use of additional processes such as coagulation, sedimentation and filtration. Nanoparticles of inorganic materials, such as metal oxides and semiconductors, have generated considerable attention due to their novel properties compared with their bulk materials (Iwasaki et al., 2000; Kamat and Meisel, 2002). A number of reports have shown that iron oxide has special photochemical and catalytic properties that strongly depend on particle size (Stramel and Thomas, 986). However, only a few studies of the catalytic activity of colloidal iron-based nanoparticles have been made. Catalyst recovery is required following water treatment using AOPs. In order to avoid this step, nanocatalysts may be immobilized on inert surfaces, foams or nanofibers without a reduction in catalytic activity (Bauer et al., 999; Yeber et al., 2000). Therefore, investigation of the effectiveness and applicability of new nanoparticle catalysts may be divided into the following main steps: Synthesis of nanoparticle catalysts and testing of their catalytic behavior. Immobilization of nanocatalysts on inert surfaces, foams or nanofibers, and investigation of their catalytic behavior. Investigation of the regeneration of the adsorbent and the nanocatalyst in successive catalytic runs. The objective of this research work is to continue the efforts of the first step: to test the catalytic behavior of iron oxidebased nanocatalysts through the degradation and mineralization of organic pollutants in wastewater as a possible method of treating contaminated groundwater and industrial wastewater. For this study, typical organic contaminants, such as ethylene glycol and phenol, were chosen as simulating pollutants. Ethylene glycol is used in large quantities as a car-cooling fluid or as an airplane and runway deicer. Large quantities of ethylene glycol have created environmental hazards, leading to the serious pollution of drinking water. Several types of industrial waste contain phenols; they are very harmful and highly toxic towards microorganisms (Cheng et al., 995). Many phenol compounds are used as solvents or reagents in industrial processes and are therefore very common contaminants in industrial wastewater and contaminated drinking water sources. 2. Experimental 2.. Materials Iron chloride hexa-hydrate, FeCl 3 6H 2 O (analytical grade; Merck KGaA, Germany), 30% hydrogen peroxide (analytical grade; PA, Panreac Quimica SA), phenol (analytical grade; Fluka) and chemically pure ethylene glycol (BIO LAB Ltd., Israel) were used as received Analysis and equipment A morphology study was performed using cryogenic transmission electron microscopy (cryo-tem), a Philips CM20 TEM optimized for cryo-tem work equipped with an Oxford Instruments CT-3500 cooling holder system. The ph was determined using a Consort P-90 electrochemical analyzer. Total organic carbon (TOC) and phenol content analyses were carried out using a TOC-5000A Shimadzu analyzer and a Hach DR/20 data-logging spectrophotometer with the phenols 4-aminoantipyrine method Preparation of iron(3) oxide-based nanocatalysts The preliminary material used for preparing the colloidal iron nanoparticles was iron chloride hexa-hydrate, FeCl 3 6H 2 O (analytical grade; Merck). Hydrolysis was used to prepare a % sol with iron nanocatalysts with initial acidity (ph ¼ 0.8). A series of iron oxide-based nanocatalysts was then prepared by diluting the initial solution. Cryo-TEM was used for characterizing the prepared nanoparticle material. Fig. shows an example of a cryo-tem image of a cluster of needle-like nanoparticles. The corresponding electron diffraction pattern (Fig. B) reveals that the needle-like nanoparticles seem to be crystals with lattice constants (characteristic inter-planar distances) of 3.3, 2.48 and.75 calculated by the Bragg equation. Analogous cryo-tem images of nanocatalysts were obtained for ph values used in the catalytic process in the ph range Currently, there is no clear identification of the crystalline structure of the

3 TOC, ppm 494 The catalytic behavior of the nanocatalyst is exhibited by a series of experiments based on the destruction of ethylene glycol in solution, as shown in Figs. 2 and 3. The results are shown on semi-logarithmic scale in order to demonstrate the low TOC level obtained in the experiments. In these experiments, the initial concentration of ethylene glycol varied from 4600 (800 ppm TOC) to 24,000 ppm (9300 ppm TOC), the concentration of nanocatalyst C Fe varied from 550 to 2500 ppm and the hydrogen peroxide concentration varied from 0.65% to 4.3%. Generally, these results show that the 0 Total organic carbon, ppm Fig. (A) Cryo-TEM image of a cluster of needle-like nanoparticles. (B) Electron diffraction of the sample in (A). Fig. 2 Degradation of ethylene glycol in the Fenton-like process at different concentrations of nanocatalyst. Concentration of nanocatalyst, ppm: m 550, & 650, 800, K, and J 500, 800, Solid line represents the initial ethylene glycol concentration of 6,000 ppm (6200 TOC) and the initial hydrogen peroxide concentration of 4.3%. Dotted line represents the initial ethylene glycol concentration of,000 ppm (4250 TOC) and the initial hydrogen peroxide concentration of 3%; dashed line represents the initial ethylene glycol concentration of 4600 ppm (800 TOC) and the initial hydrogen peroxide concentration of.2%. 0 particles; they are probably a mixture of oxides and hydroxides. The term iron oxide-based nanocatalyst is used in this article. 3. Results and discussion 3.. Degradation of ethylene glycol with nanocatalysts The catalytic behavior of the nanocatalyst was investigated by changing the concentrations of the dissolved matter, the catalyst and the peroxide. The experiments were carried out at room temperature without UV radiation, at visible room light and in some cases in a darkened vessel with no light in the reaction (the latter did not differ from regular light). The ph values in these experiments ranged from 2 to 3.5. In all of these experiments, the hydrogen peroxide concentration ranged from 0.65% to 4.3% Fig. 3 The influence of hydrogen peroxide concentration on the degradation of ethylene glycol in the Fenton-like process. Concentration of nanocatalyst: ppm. Initial concentration of H 2 O 2,%:K 4.3,. 3.2, m 2,.2, and 0.65.

4 k, /min 495 Fenton-like reaction efficiently destroys ethylene glycol with nanocatalysts in the presence of hydrogen peroxide without UV or visible radiation sources. Fig. 2 shows the significant influence of nanocatalyst concentration C Fe on the degradation rate of the organic matter. Concentration presentation as total organic content (TOC) was chosen to represent the data. This is important since the target is to remove all organic compounds and not only the EG, as by-products may be even more harmful than the original matter. The straight lines on semi-logarithmic scale confirm the assumption of first-order reaction degradation of the dissolved material. The by-products degrade also as first-order reaction, demonstrating the non-selectivity nature of the catalytic reaction. For example, an initial concentration of ethylene glycol of ¼,000 ppm (4250 ppm TOC) reacted with different concentrations of catalyst C Fe, ranging from 650 to 2500 ppm, showing an increased first-order rate of reaction with its rate coefficient k varying from to 4. 2 /min (see Fig. 4). This is also true for the other initial concentrations shown in Fig. 2. When sufficient catalysts and peroxide are available, the reaction proceeds rapidly to a level of several ppm of TOC. This ensures complete mineralization conversion to CO 2 and H 2 O with no intermediate by-products. The influence of peroxide concentration on reaction rate coefficient k is shown in Fig. 3. An increase in initial H 2 O 2 concentration from 0.65% to 4.3% for a constant nanocatalyst concentration of ppm and an initial ethylene glycol concentration of 4600 ppm (800 ppm TOC) causes the initial first-order reaction rate coefficient k to increase from to 0.6 /min. With enough catalysts at a high peroxide level, the reaction rate is very high, and TOC may be destroyed almost completely in about 40 min. Fig. 4 shows the non-linear dependence of the first-order reaction rate coefficient k on the ratio of concentrations of iron oxide-based nanocatalyst, hydrogen peroxide and ethylene glycol to the initial peroxide concentrations. Significantly increasing the catalyst and peroxide concentrations increases the value of k. Increasing the ethylene glycol to the initial hydrogen peroxide concentration ratio R from 0. to 0.56 for a nanocatalyst concentration of ppm leads to a decrease in reaction rate coefficient k from 0.6 to /min. The chemical oxidation of the proposed process can be represented by mth-order reaction kinetics (Lin and Lo, 997): dc dt ¼ kcm () with the initial condition as t ¼ 0; C ¼, (2) where C is the organic solute concentration represented as TOC (ppm), the initial TOC concentration, m the order of reaction, k the reaction rate coefficient (/min) and t the time. The choice of basing the dissolved matter on TOC measurements is related to the ability to show complete degradation without unwanted by-products. For a first-order reaction (m ¼ ), Eq. () may be rewritten as dc ¼ kdt. (3) C The solution of Eq. (3) using initial condition (2) can be written as C ¼ expð ktþ. (4) From the experimental data presented in Figs. 2 4, an expression for calculating the reaction rate coefficient k may be approximated by the dependence: EG/H2O Catalyst concentration, ppm Concentration of H 2O2, % concentration H2O2 catalyst concentration Ethylene Glycol to Hydrogen Peroxide initial concentrations ratio, ppm/ppm Fig. 4 The first-order reaction rate coefficient k as a function of the concentrations of iron-based nanocatalysts, hydrogen peroxide and ethylene glycol to the initial peroxide concentrations ratio. The symbols represent the experimental data, while the lines come from Eq. (5). k ¼ 0:65 C 2 FeOOHC H2 O 2. (5) It should be noted that the order of dependence on nanocatalyst concentration and on the ethylene glycol to hydrogen peroxide initial concentrations ratio R is 2. This fact illustrated the strong effect of nanocatalyst concentration and ethylene glycol to hydrogen peroxide initial concentrations ratio R on the reaction rate coefficient. Values of reaction rate constant k are plotted in Fig. 4, against catalyst concentration and against peroxide concentration. The dots represent experimental results and the lines are calculated using Eq. (5). The calculated rate of reaction coefficient values is in good agreement with the experimental values. Also shown in the figure is the change in the k values as the ratio of the ethylene glycol concentration to the initial concentration of peroxide. The TOC concentration obtained is a mixture of different compounds originated from ethylene glycol and includes reaction intermediates that could have been reacting at different rates. The overall catalytic reaction rate demonstrates a constant first-order rate of reaction for all byproducts during the process, including the original molecule of ethylene glycol. The range of reaction rate coefficient for ethylene glycol degradation reported in the literature is

5 Concentration of phenol, ppm 496 k EG ¼ 0.07/min (McGinnis et al., 997, 2000, 200). The values of the reaction rate coefficient for ethylene glycol degradation with iron(3) oxide-based nanocatalysts without UV light are therefore higher by a factor of 2 6 than the values reported in the literature for UV Fenton s using H 2 O Destruction of phenol with nanocatalysts For experiments with nanocatalysts carried out with initial concentrations of phenol solution of and 5000 ppm, the nanocatalyst concentrations ranged from 5 to 240 ppm. All of the experiments were carried out at room temperature. The ph ranged from 2 to 3. For the sake of convenience, the results are shown in this section as phenol ppm concentration and not as TOC. Figs. 5 and 6 show that the phenol is efficiently destroyed by the Fenton-like reaction with iron(3) oxide-based nanocatalysts in the presence of hydrogen peroxide without UV and visible radiation sources. The final concentration is a function of peroxide concentration, as will be explained later. Sufficient degradation of the phenol was shown to be possible with the appropriate concentration of catalysts and peroxide. The lines in Fig. 5 show a significant influence of nanocatalyst concentration on phenol destruction rate. As an example, increasing the iron oxide-based nanocatalyst concentration from 5 to 60 ppm resulted in a sharp decrease in phenol concentration from an initial to 400 ppm (after 0.5 min, a catalyst concentration of 5 ppm) to 2.0 ppm (after 0.5 min, a catalyst concentration of 60 ppm), and a final phenol concentration from 2 4 ppm to ppm. The curves behavior in the pictures may be divided into two regions: the first shows a sharp exponential decrease in phenol concentration (first-order reaction kinetics); the second shows a decreasing rate where asymptotic behavior is observed. From the experimental data presented in Fig. 5, the initial reaction rate coefficient for iron(3) oxide-based nanocatalysts for the region of the first-order reaction is approximately Concentration of phenol, ppm k ¼ 3 /min. The value of half-life time, the time required to decrease the initial concentration by half, is t /2 ¼ min. The rate is 35 times higher than the classic Fenton s reagent k ¼ 0.37 /min, and therefore much faster compared with t /2 ¼.8 min, as reported by Esplugas et al. (2002). The effect of hydrogen peroxide concentration on phenol oxidation rate is shown in Fig. 6 for a constant initial concentration of ppm phenol in solution, where the nanocatalyst concentration was 20 ppm and the initial H 2 O 2 concentration increased from 0.36% to 0.96%. The influence of hydrogen peroxide concentration shows an increased oxygenation rate with increased peroxide concentration, yet a slight decrease in phenol oxidation rate followed a further increase in concentration. Increasing the concentration from 0.36% to 0.5% increases the reaction rate as the phenol concentration decreases from 2 to 4 ppm. A further increase in H 2 O 2 concentration leads to a decrease in reaction Fig. 6 Influence of hydrogen peroxide concentration on phenol destruction during the catalytic process. Concentration of nanocatalyst: 5 ppm. Initial hydrogen peroxide concentration & 0.36, 0.5, m 0.72, Fig. 5 Kinetics of catalytic destruction of phenol. Concentration of nanocatalyst: 5 ppm, m 30 ppm,. 60 ppm, K 20 ppm, J 200 ppm, and 240 ppm. Solid line represents the initial phenol concentration of 5000 ppm and the initial hydrogen peroxide concentration of 2.5%; dashed line represents the initial phenol concentration of ppm and the initial hydrogen peroxide concentration of 0.5%.

6 -t/ln(c/ ) Fig. 7 Catalytic reaction of phenol t/ln(c/ ) versus reaction time t. Initial phenol concentration of 5000 ppm, initial hydrogen peroxide concentration of 2.5%, concentration of nanocatalyst, ppm: K 20 and 240. Initial phenol concentration of ppm, concentration of nanocatalyst: 5 ppm, initial hydrogen peroxide concentration & 0.36, 0.5, m 0.72,. 0.96; initial hydrogen peroxide concentration 0.5%, concentration of nanocatalyst, ppm: 5, J 30. rate. Initially, the H 2 O 2 acts as an initiator. An increase in initial hydrogen peroxide concentrations leads to an increase in the number of reactive free hydroxyl radicals (OH*). As a result, an increase in the number of reactive free hydroxyl radicals per molecule of organic matter leads to an increase in oxidation rate. The hydrogen peroxide also acts as an (OH*) scavenger (Neyens and Baeyens, 2003), hence an increase in its concentration leads to a decrease in the concentration of free hydroxyl radicals. A decrease in the concentration of free hydroxyl radicals causes a decrease in the rate of phenol degradation. As it follows from the experimental investigations, the exponential decrease in phenol concentration may simulate the process of phenol destruction with iron oxide-based nanocatalysts. It is therefore possible to describe the destruction rate in solution as C ¼ exp t, (6) a þ bt where a (min) and b (dimensionless) are two constants. In order to explain the physical meaning of the parameters, the derivation of C/ over time t can be made: dðc= Þ a C ¼ dt ða þ btþ 2. (7) When t is small (t5a/b), Eq. (7) can be rewritten as dðc= Þ ¼ C (8) dt a Eq. (8) represents the first-order reaction kinetics similar to Eq. (3), where a is the inverse proportional to the first-order reaction rate coefficient k. For a long time period (t-n), Eq. (6) can be rewritten as C t! ¼ exp (9) b or b ¼ ln C t!. () b is therefore inversely proportional to the logarithm of the maximum phenol destruction capacity. Hence, Eq. (6) can be rewritten as t ¼ a þ bt. (6-) lnðc= Þ Fig. 7 shows the function t= ln ðc= Þ versus time of reaction t. From this linear graph, the constants a and b can be calculated (Chan and Chu, 2003), and were determined for different initial concentrations of hydrogen peroxide and iron oxide-based nanocatalysts. The linearity of the curves in Fig. 7 demonstrates the validity of the model adapted for phenol destruction in solution by the chosen catalyst. The calculated values of constants a and b changed from to 0.37 min and from 0.26 to 0.5 accordingly. It can be seen that the lower the intercept on the y-axis, the higher the initial rate of the reaction. The b values are not a strong function of the catalysts or the peroxide concentration Stability of nanocatalyst aging A series of experiments were conducted to investigate the catalytic behavior of the iron(3) oxide-based nanocatalyst as a function of its aging. The experiments were carried out at room temperature without UV radiation, at visible room light and in a darkened vessel with no light in the reaction. In all these experiments the initial concentration of ethylene glycol was,000 ppm (4250 TOC), the concentration of nanocatalyst 800 ppm, and that of hydrogen peroxide 3%. The initial values of ph in these experiments ranged from 2 to 3.5. Value of the reaction rate coefficient k with fresh nanocatalyst, and for 7, 30 and 90 days after their production, was about 0.04/min. In all these experiments no aging time effect on the nanocatalyst kinetics for ethylene glycol degradation was found.

7 Conclusions This study presents information about the catalytic properties of iron-based nanoparticles for the degradation of some organic pollutants in wastewater by AOPs. The main conclusions are as follows: Ethylene glycol and phenol are efficiently destroyed by the Fenton-like reaction with iron(3) oxide-based nanocatalysts in the presence of hydrogen peroxide without ultraviolet (UV) light or any radiation sources. Complete oxygenation of ethylene glycol with iron oxidebased nanoparticles exhibits first-order reaction kinetics. A strong effect of nanocatalyst concentration on the reaction rate coefficient was shown. The ethylene glycol degradation rate with the iron-based nanoparticles without UV light or visible radiation sources is about 2 4 times higher than the values reported in the literature for Fenton s reagent/h 2 O 2 and UV. A strong effect of catalyst concentration on the phenol destruction rate was shown. The phenol destruction with the iron oxide-based nanocatalysts may be simulated by exponential decay. The phenol destruction rate with the iron oxide-based nanoparticles is about 35 times higher than the values reported in the literature for Fenton s reagent/h 2 O 2 and UV. No effect of nanocatalyst aging was observed. Acknowledgements The authors wish to acknowledge the partial support of this research project by the Ministry of Science, the Peres Center for Peace, and the Hurovitz Foundation. R E F E R E N C E S Bauer, R., Waldner, G., Fallmann, H., Hager, S., Klare, M., Krutzler, T., Malato, S., Matetzky, P., 999. The photo-fenton reaction and the TIO 2 /UV process for waste water treatment-novel developments. Catal. Today 53, Buxton, G.V., Greenstock, C.L., Helman, W.P., Ross, A.B., 988. Critical review of rate constants for reactions of hydrated electrons, hydrogen atoms and hydroxyl radicals in aqueous solution. J. Phys. Chem. Ref. Data 7, Chan, K.H., Chu, W., Modeling the reaction kinetics of Fenton s process on the removal of atrazine. Chemosphere 5, Cheng, S., Tsai, S.J., Lee, Y.F., 995. Photo catalytic decomposition of phenol over titanium dioxide of various structures. Catal. Today 26, Esplugas, S., Gimenez, J., Contrerass, S., Pascual, E., Rodriguez, M., Comparison of different advanced oxidation process for phenol degradation. Water Res. 36, Fenton, H.J.H., 894. Oxidation of tartaric acid in the presence of iron. J. Chem. Soc. 65, Iwasaki, M., Hara, M., Ito, S., Synthesis of nanometer-sized hematite single crystals through NAC-FAS method. J. Mater. Sci. 35, Kamat, P.V., Meisel, D., Nanoparticles in advanced oxidation processes. Curr. Opin. Colloid Interface Sci. 7, Lin, S.H., Lo,, Ch.C., 997. Fenton process for treatment of desizing wastewater. Water Res. 3, McGinnis, B.D., Middlebrooks, E.J., Adams, V.D., 997. Degradation of ethylene glycol in photo Fenton systems. Preprints of Extended Abstracts presented at the ACS National Meeting, American Chemical Society, vol. 37, pp McGinnis, B.D., Adams, V.D., Middlebrooks, E.J., Degradation of ethylene glycol in photo Fenton systems. Water Res. 34, McGinnis, B.D., Adams, V.D., Middlebrooks, E.J., 200. Degradation of ethylene glycol using Fenton s reagent and UV. Chemosphere 45, 8. Neyens, E., Baeyens, J., A review of classic Fenton s peroxidation as an advanced oxidation technique. J. Hazardous Mater. B 98, Oliveros, E., Legrini, O., Holb, M., Muller, T., Braun, A., 997. Industrial wastewater treatment: large scale development of a light-enhanced Fenton reaction. Chem. Eng. Process. 36, Perez, M., Torrades, F., Garcia-Hortal, J.A., Domenech, X., Peral, J., Removal of organic contaminants in paper pulp treatment effluents under Fenton and photo-fenton conditions. Appl. Catal. 36, Safarzadeh-Amiri, A., Bolton, J.R., Cater, S.R., 996. The use of iron in advanced oxidation processes. J. Adv. Oxid. Technol., Safarzadeh-Amiri, A., Bolton, J.R., Cater, S.R., 997. Ferrioxalatemediated photo degradation of organic pollutants in contaminated water. Water Res. 3, Sigman, M.E., Buchanan, A.C., Smith, S.M., 997. Application of advanced oxidation process technologies to extremely high TOC aqueous solutions. J. Adv. Oxid. Technol. 2, Stramel, R.D., Thomas, Y.K., 986. Photochemistry of iron oxide colloids. J. Colloid Interface Sci., Yeber, M.C., Rodriguez, J., Freer, J., Duran, N., Mansilla, H.D., Photo catalytic degradation of cellulose bleaching effluent by supported TiO 2 and ZnO. Chemosphere 4,