Polyol-made Mn3O4 nanocrystals as efficient Fenton-like catalysts

Applied Catalysis A: General 386 (2010) 132–139

Contents lists available at ScienceDirect

Applied Catalysis A: General journal homepage: www.elsevier.com/locate/apcata

Polyol-made Mn3 O4 nanocrystals as efficient Fenton-like catalysts Tarik Rhadfi a,b , Jean-Yves Piquemal a,∗ , Lorette Sicard a , Frédéric Herbst a , Emmanuel Briot c , Marc Benedetti d , Ahmed Atlamsani b a

ITODYS, Université Paris Diderot (Paris 7), CNRS – UMR 7086, Bâtiment Lavoisier, 15 rue Jean-Antoine de Baïf, 75205 Paris Cedex 13, France Laboratoire de Physico-Chimie des Interfaces et Environnement, Faculté des Sciences, Université Abdelmalek Essaâdi, BP2121, 93000 Tétouan, Morocco c PECSA, Université P. et M. Curie (Paris 6), CNRS–UMR 7195, Bâtiment F, 4 place Jussieu, 75005 Paris, France d Laboratoire de Géochimie des Eaux, Université Paris Diderot (Paris 7) IPGP, CNRS – UMR 7154, 75205 PARIS Cedex 13, France b

a r t i c l e

i n f o

Article history: Received 12 May 2010 Received in revised form 21 July 2010 Accepted 23 July 2010 Available online 30 July 2010 Keywords: Mn3 O4 Nanomaterials Polyol Hydrogen peroxide Fenton chemistry

a b s t r a c t Nanosized Mn3 O4 single crystals are prepared using the polyol method and characterized with powder X-ray diffraction, transmission electron microscopy and physisorption experiments. Depending on the reaction time, either oval-like nanocrystals with a mean TEM size of 7.9 nm or rhomboedron-like particles with a mean TEM size of 12.6 nm are isolated. These particles are active towards the decomposition of aqueous hydrogen peroxide at room temperature and the data show that the reaction kinetics depend on the particle size and shape. The degradation of a model substrate, methylene blue is also investigated.

1. Introduction In recent years manganese oxides have attracted considerable attention and are used in many domains such as energy storage [1], ion exchange [2] or heterogeneous catalysis. Indeed, bulk or supported manganese oxides have been used for the oxidation of alcohols [3–7] or CO [8–10] the epoxydation of olefins [11], the complete oxidation of hydrocarbons [9,12–15], the soot combustion [16] or the selective catalytic reduction of NOx [17]. They are also involved in the treatment of wastewater using hydrogen peroxide as the primary oxidant via Fenton-type chemistry [6,7,18–20]. In such a process, hydrogen peroxide is decomposed and highly reactive hydroxyl radicals (• OH) are generated. These non-selective species are among the most oxidizing known compounds and are able to degrade a wide range of organic pollutants. Considering the classical Fenton reagent, i.e. Fe2+ /Fe3+ –H2 O2 , given the acidic pH associated with this process (in the range 2–4), the effluent must be subsequently neutralized with a base, giving rise to the formation of iron sludge [21]. In order to overcome this problem, heterogeneous catalysts have been proposed. Iron-based systems have been extensively studied, such as ␣-Fe2 O3 [22] and Fe3 O4 [23] nanoparticles, iron oxide nanoparticles supported on alumina coated mesoporous silica [24] or iron-containing ZSM-5

∗ Corresponding author. Tel.: +33 1 57 27 87 66; fax: +33 1 57 27 72 63. E-mail address: [email protected] (J.-Y. Piquemal). 0926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.apcata.2010.07.044

© 2010 Elsevier B.V. All rights reserved.

zeolites [25–27]. Despite their high potential in oxidation catalysis, manganese oxides have not been investigated so much as Fentontype catalysts [6,7,18,20,28–30]. Particularly, the Mn3 O4 spinel, with Mn(II) and Mn(III) mixed valencies is a good candidate. Mn3 O4 particles can be synthesized by heating at high temperature manganese oxides, hydroxides, oxyhydroxides or manganese salts in air. However, these methods lead to particle sizes higher than a hundred nanometers. Moreover, the grains are often coarsened and non-uniform in size. Spray hydrolysis was also reported but the high temperatures (more than 450 ◦ C) do not permit to obtain small particles [31]. Finally, sol–gel methods also require a post-calcination treatment. In order to obtain monodisperse and very small particles (less than 50 nm in size), low temperature syntheses in basic media are more appropriated. Some microwave procedures have been carried out either in water [19], water and ethanol [32], or benzylalcool [33]. Hydrothermal [12,34] and solvothermal syntheses [35–37] were also developed but they require long reaction times. Moreover, the precipitation of Mn3 O4 nanoparticles was achieved in solution under atmospheric pressure in water in the absence [38] or in the presence [39] of surfactants; in long chain amines [40–42]; in ethanol [43] and in polyol [44]. In this last case, particles of 30–40 nm were obtained. Nevertheless, a unique synthesis procedure permitting to obtain very well crystallized and monodisperse nanoparticles in a short time and above all allowing to vary the size and shape was seldom reported. We have developed a very simple synthesis method which consists in heating a manganese salt in diethylene glycol at relatively

T. Rhadfi et al. / Applied Catalysis A: General 386 (2010) 132–139

low temperature [45]. Depending on the reaction time, the particles vary in form and size: they are about 7 nm ovals after 5 min and 12 nm rhomboedrals after 2 or more hours. At any moment of the reaction, even after a few minutes, the particles are very well crystallized and uniform in size without the addition of surfactants. This work reports on the preparation of Mn3 O4 nanocrystals by the polyol process, a “chimie douce route” developed in our laboratory, and their use in advanced oxidation techniques. The catalytic activity of these nanomaterials for the decomposition of H2 O2 and the degradation of methylene blue, a basic cationic dye used extensively for dying cotton, wool and silk, are examined. 2. Experimental 2.1. Chemicals Diethyleneglycol (99%, Acros), ethanol (Normapur, VWR), dimethyl sulfoxide (VWR), MnII (CH3 COO)2 ·4H2 O (≥99%, Aldrich), methylene blue (>98.5%, RAL), d-mannitol (A.C.S. reagent, Aldrich) and 30 wt.% aqueous hydrogen peroxide (AnalaR Normapur, VWR; stabilizer: stannate) were used as received. The H2 O2 concentration, determined by idometric titration, was 9.9 ± 0.2 mol L−1 . 2.2. Preparation of nanosized Mn3 O4 3.06 g of manganese(II) acetate tetrahydrate, 10 mL of distilled water and 125 mL of diethyleneglycol (DEG) were introduced in a three-neck flask. The solution was heated up to 100 ◦ C (at a heating rate of 6 ◦ C min−1 ) and left under vigorous stirring for 5 min (sample Mn-A) or 16 h (sample Mn-B). The solution was then cooled down. The brown powders were recovered by centrifugation, washed 3 times with ethanol and dried at 50 ◦ C overnight. 2.3. Characterization techniques X-ray diffraction patterns were obtained with a Panalytical X’Pert Pro diffractometer equipped with a PIXcel detector using Ni-filtered Cu-K␣ radiation. The data were collected at room temperature with a 0.026◦ step size in 2, from 2 = 20◦ to 70◦ . The crystalline phase was identified by comparison with ICSD reference files. The size of coherent diffraction domains were determined using MAUD software [46] which is based on the Rietveld method combined with Fourier analysis, well adapted for broadened diffraction peaks. The transmission electron microscopy (TEM) and highresolution TEM (HRTEM) studies were performed using a Jeol JEM 100CXII transmission electron microscope operating at 100 kV. One drop of an ethanolic suspension of the as-produced particles was deposited on the carbon membrane of the microscope grid for the observations. The particle size distribution was obtained from TEM images using a digital camera and the SAISAM software (Microvision Instruments), diameter dp from Eq. (1): dp =

 2 nd i i i nd i i i

(1)

where ni is the number of particles with di diameter. The statistical result of the particle size was obtained by counting about 200 particles considering a spherical particle shape. dp is given in nanometer. Adsorption and desorption nitrogen isotherms were obtained at 77 K using a Micromeritics ASAP 2020 apparatus. The samples were outgassed at 423 K and 0.1 Pa for 8 h before measurements. Specific surface area (SBET ) values were obtained using the Brunauer–Emmett–Teller equation [47] with relative pressures in the range 0.05–0.20.

133

UV–vis transmission spectra and measurements were recorded at room temperature on a Varian Cary 5E spectrometer equipped with a double monochromator. Total carbon (TC), inorganic carbon (IC) and dissolved organic carbon (DOC) analyses were carried out using a Shimadzu TOC-VCSH total organic carbon analyzer. 2.4. Reactivity measurements 2.4.1. H2 O2 decomposition volumetry The H2 O2 decomposition was measured at 293 K by determining the volume of oxygen evolved as a function of time, using a known mass of catalyst (1.0–10.0 mg), 2.0 mL of 30 wt.% aqueous H2 O2 and 20.0 mL of distilled water. A Mettler Toledo MX5 micro balance (m ± 1 ␮g) was used for weighing the powders. Measurements were performed in an all-glass reactor vessel (50 mL) and O2 evolution was determined using a home-made gas burette system. Indeed, volumetry has been reported to be more precise and convenient to follow the reaction kinetics than titrymetry [18]. The initial pH was 5.1 and the pH of the reaction mixture was not controlled via the addition of a buffer in order to minimize the depletion of • OH radicals with the buffer. A blank experiment was realized without adding manganese oxide nanoparticles in order to evaluate the degradation of hydrogen peroxide without catalyst in the same operating conditions. Using these conditions, the conversion of hydrogen peroxide was evaluated to be about 1.5% after 24 h at 293 K. Catalytic decomposition of hydrogen peroxide was also realized using HO• radical scavengers in order to assess the formation of these radicals. d-mannitol [48], methylene blue [49,50] and dimethylsulfoxide [48,51] have been reported as efficient HO• scavengers. Either a 0.1 g L−1 aqueous solution of methylene blue, a 13.8 g L−1 aqueous solution of d-mannitol or a 35.4 g L−1 aqueous solution of dimethyl sulfoxyde were used in this study. To 20 mL of one of these solutions was first added the catalyst then 2.0 mL of 30 wt.% aqueous H2 O2 . The volumes of O2 generated were measured as above and compared to that obtained with pure water. 2.4.2. Methylene blue degradation Kinetic measurements were realized in a standard Schlenk tube at room temperature (293 K) under moderate magnetic stirring. In a typical catalysis test, in 20.0 mL of a 100 mg L−1 methylene blue (MB) solution, were added the nanocrystals (1.0, 2.0, 5.0 or 10.0 mg) and the desired amount of 30 wt.% aqueous hydrogen peroxide (in the range 0.5–2.0 mL). The conversion of methylene blue was determined from the UV–vis absorbance using the Beer–Lambert relation. The UV–vis spectrum of MB shows two main peaks located at  = 607 nm and  = 660 nm. The best linear fit is obtained for the peak at 607 nm which was thus chosen to obtain a calibration curve using a series of standard solutions. No conversion of MB was observed after 5 h when the following blank experiments were performed: (i) 20.0 mL of a 100 mg L−1 MB aqueous solution and 2.0 mL of 30 wt.% aqueous hydrogen peroxide without nanocatalysts and (ii) 20.0 mL of the MB solution and 5.0 mg of Mn3 O4 nanoparticles without H2 O2 . Even if a small amount of MB can adsorb onto the particle surface [52], this quantity can be considered as negligible towards the initial dye stock. 3. Results and discussion 3.1. Synthesis and characterization of Mn3 O4 nanoparticles The X-ray diffraction patterns of the Mn3 O4 nanocrystals isolated after 5 min (Mn-A) and 16 h (Mn-B) are presented in Fig. 1. The

134

T. Rhadfi et al. / Applied Catalysis A: General 386 (2010) 132–139

cal, the average diameter (d) or edge length (a) can be calculated using the following equations: d = 6/Mn3 O4 · SBET for a sphere or a = 6/Mn3 O4 · SBET for a cube. With a density of 4.83 g cm−3 for hausmannite, the calculated mean diameter for Mn-A is 9.7 nm and the mean edge length for Mn-B is 13.8 nm. These values agree fairly well with those determined using TEM images. 3.2. Decomposition of hydrogen peroxide aqueous solutions Mn3 O4 nanocatalysts have been tested for the decomposition of hydrogen peroxide, in order to evaluate their activity towards the generation of hydroxyl radicals OH• from H2 O2 . First, to assess the possible formation of HO• radicals, the decomposition of hydrogen peroxide was performed in water or in the presence of different HO• scavengers (see Section 2.4.1). The results presented in Fig. 4 show unambiguously that the emission of gaseous O2 is higher in water than in the presence of the HO• scavengers used in this study. Moreover, d-mannitol and methylene blue appear to be more efficient than DMSO for the quenching of the radicals, resulting in a lower evolution of gaseous oxygen for the formers. This can be explained considering the reaction mechanism of OH• with DMSO: H3 C• are generated first leading subsequently to the formation of methane and ethane [54]. Based on previous studies on spinel metal oxides [6,7,49,55,56], our results can be explained considering the following mechanism for the production of HO• radicals: (i) Generation of HO• radicals via the Haber Weiss reaction on the Mn2+ surface atoms: Mn2+ surf + H2 O2 → Mn3+ surf + HO• + OH–

(2)

(ii) The Mn3+ surface atoms are reduced to Mn2+ via: Mn3+ surf + H2 O2 → Mn2+ surf + HOO• + H+ Fig. 1. X-ray diffraction patterns for Mn3 O4 nanoparticles: (a) Mn-A and (b) Mn-B. The observed intensity data are shown by dotted lines. The solid lines correspond to the calculated pattern of Mn3 O4 . The difference between the observed and calculated intensities is shown by the bottom dotted line.

particles crystallize with the hausmannite structure (JCPDS no 00024-0734). Rietveld analyses have been performed and the mean crystallite size is found to be 7.7 and 11.5 nm for Mn-A and Mn-B, respectively. These values are very close to the mean sizes measured from the TEM images that show the monocrystalline nature of the particle (see Fig. 2). This was also confirmed with HRTEM images (vide infra). Both XRD and HRTEM studies show well crystallized particles. Different size and morphologies can be obtained in DEG, depending on the reaction time. In a very short time, i.e. 5 min, oval shaped particles are obtained. Their mean diameter dp is 7.9 nm and the standard deviation is small with a ratio, /dp = 14%. These particles are well crystallized as revealed by their XRD spectrum (Fig. 1) and by the HRTEM images (see Fig. 2b). For longer reaction times, a different morphology is observed and more facetted, rhomboedron-like particles are generated after 16 h. Their mean size is 12.6 nm with /dp = 13% (Fig. 2c and d). Mn3 O4 nanocrystals were also analyzed using physisorption techniques. The materials exhibit type IV isotherms according to the IUPAC classification with a H2 hysteresis type (Fig. 3). Given that the materials are non-porous, this can be explained by interparticle porosity or percolation effects [53]. The moderate C BET [47] constants found for Mn-A and Mn-B are in the same range, 54 and 44, respectively, and are indicative of the absence of microporosity within the materials. The specific surface areas, determined using the BET method, are 128 and 90 m2 g−1 for Mn-A and Mn-B, respectively. Assuming an isotropic shape, either cubic or spheri-

(3)

(iii) To account for the formation of O2 , several reactions can be proposed: H2 O2 + HO• → H2 O + HOO•

(4)

Then: Mn3+ surf + HOO• → Mn2+ surf + H+ + O2 Or, a recombination of the 2 HO• → H2 O + ½ O2

HO•

(5)

radicals [57]: (6)

(iv) HO• radicals can attack an organic substrate RH such as methylene blue: RH + HO• → R • + H2 O

(7)

A part of the HO• radicals are quenched by the scavengers and thus cannot react via Eqs. (4)–(6). The decomposition of hydrogen peroxide is then significantly reduced as observed experimentally. For the two sets of particles, the data show (see Fig. 5) that the reaction follows a pseudo first-order kinetic rate law: −d[H2 O2 ]/dt = kobs [H2 O2 ], which can also be written ln([H2 O2 ]0 /[H2 O2 ]) = kobs t, where kobs is the observed first-order rate constant and [H2 O2 ] and [H2 O2 ]0 are the hydrogen peroxide concentration at any time t and at time zero, respectively. Similar results were reported for different heterogeneous catalysts, including iron-based solids such as goethite (␣-FeOOH) [58], hematite (␣-Fe2 O3 ) [22], 10 nm sized MnFe2 O4 nanoparticles [59] or manganese oxide-based solids [18,29]. Fig. 6 presents a plot of the observed first-order rate constants as a function of the Mn3 O4 concentration, for the two series of nanoparticles. The calculated kobs values are reported in Table 1.

T. Rhadfi et al. / Applied Catalysis A: General 386 (2010) 132–139

135

Fig. 2. TEM and HRTEM images of Mn3 O4 nanoparticles obtained in DEG after 5 min: Mn-A (a) and (b), and 16 h: Mn-B (c) and (d). Inset: numerical diffraction pattern; zone axis: [1 1 1].

Fig. 3. N2 physisorption isotherms obtained at 77 K for samples Mn-A and Mn-B: 䊉, : adsorption branches; , : desorption branches.

Two different behaviours are clearly observed depending on the sample used. For Mn-B particles, a linear relationship is found when plotting kobs as a function of the nanoparticles concentration (see Fig. 6), indicating a first-order with respect to the catalyst concentration, i.e. v = k [Mn3 O4 ][H2 O2 ], yielding a second global order where k = 0.227 min−1 L g−1 and kobs = k [Mn3 O4 ]. Whereas, for Mn-A sample, the kobs constant does not vary linearly with the Mn3 O4 concentration, indicating a more complex kinetic rate law. Intrinsic constants kint values (min−1 m−2 ) can also be calculated by dividing kobs by SBET (m2 g−1 ) and the catalyst weight (g). It allows the direct comparison of the performances of the two Mn3 O4 samples towards H2 O2 decomposition at constant specific surface area. With Mn-B particles, this intrinsic constant is found to be nearly

Fig. 4. Gaseous oxygen produced in the decomposition of hydrogen peroxide over Mn-A nanocatalysts in: H2 O (), 0.1 g L−1 methylene blue aqueous solution (), 13.8 g L−1 d-mannitol () and 35.4 g L−1 dimethylsulfoxide (). See Section 2.4.1 for experimental details. The dashed line corresponds to the volume of dioxygen produced for the total decomposition of H2 O2 .

constant (see Table 1) which is not the case with Mn-A catalysts (see Table 1 and Fig. 7) for which there is a monotonous increase of the kint value with the Mn3 O4 concentration. These results indicate different kinetic behaviours for the same material but with different size and shape. At low Mn3 O4 concentration, sample Mn-B appears to be more active than sample Mn-A. On the one hand, on the basis of a thermodynamic analysis, it has been recently reported that reaction orders can vary depending on

136

T. Rhadfi et al. / Applied Catalysis A: General 386 (2010) 132–139 Table 1 Pseudo first-order (kobs ) and intrinsic (kint ) rate constants determined for the decomposition of hydrogen peroxide by Mn3 O4 nanoparticles. Catalyst

mcata a (mg)

kobs b (min−1 )

kint c (m−2 min−1 )

Mn-A

2.0 5.0 8.0 10.0

1.19 × 10 5.31 × 10−2 12.2 × 10−2 17.3 × 10−2

4.65 × 10−2 8.30 × 10−2 11.9 × 10−2 13.5 × 10−2

Mn-B

1.9 5.0 6.4 8.0 10.6

1.53 × 10−2 4.85 × 10−2 6.31 × 10−2 7.97 × 10−2 11.50 × 10−2

8.95 × 10−2 10.8 × 10−2 11.0 × 10−2 11.1 × 10−2 12.1 × 10−2

a b c

−2

mcata : catalyst weight. Reaction conditions: V(H2 O2 ) = 2 mL; V(H2 O) = 20 mL; T = 293 K. kint = kobs /(SBET × mcata ); SBET (Mn-A) = 128 m2 g−1 ; SBET (Mn-B) = 90 m2 g−1 .

the size of the nanoclusters for structure sensitive reactions [60]. On the other hand, for hematite nanoparticles prepared by controlled decomposition of iron oxalate, Hermanek et al. have shown recently that, if it is well accepted that the surface area has a marked influence on the catalytic properties, the cristallinity or the “surface quality” of the particles is of major importance: very small (∼3 nm) amorphous particles with high surface areas (∼400 m2 g−1 ) show lower activity than bigger (∼6 nm) well-crystallized particles with lower surface area (∼340 m2 g−1 ) [22]. However, in the present work, Mn-A particles as well as Mn-B particles are very well crystallized (vide supra), and a “crystalline quality” effect can thus be discarded. These results could rather be explained by differences in the nature and/or quantity of surface active sites due to the different morphologies, i.e. oval-like or rhomboedron-like shapes for Mn-A and Mn-B samples. 3.3. Degradation of methylene blue using aqueous hydrogen peroxide

Fig. 5. Pseudo first-order plots obtained for the H2 O2 decomposition at 293 K with (a) Mn-A and (b) Mn-B nanoparticles.

Fig. 6. Pseudo first-order constants kobs vs Mn3 O4 nanoparticles concentration for () Mn-A and () Mn-B.

The catalytic properties of Mn3 O4 nanocrystals were also evaluated towards the decomposition of a basic dye: methylene blue. Note that the adsorption of methylene blue on the particle surface seems unlikely and cannot explain the high conversion observed. Indeed, the zero point charge (ZPC) of Mn3 O4 is about 5.5 [61], which is approximately the same value as the pH of the catalytic tests. To observe a significant adsorption onto the particle surface, the pH should be higher than the ZPC. Moreover a blank experiment realized without hydrogen peroxide but in the presence of the nanoparticles showed no conversion of MB (see Section 2.4.2).

Fig. 7. Variation of the intrinsic rate constant kint as a function of the Mn3 O4 concentration. : Mn-A; : Mn-B.

T. Rhadfi et al. / Applied Catalysis A: General 386 (2010) 132–139

137

Table 2 Total carbon (TC), inorganic carbon (IC) analyses obtained with Mn-B nanocatalysts. Dissolved organic carbon (DOC) was calculated as follow: DOC = TC − IC.

Fig. 8. Relative absorbance A/A0 of the solutions illustrating the catalytic decomposition of methylene blue using Mn-A () and Mn-B () catalysts and a mixture of MnII and MnIII acetate salts (). Experimental conditions: T = 293 K, 20.0 mL of a 100 mg L−1 methylene blue solution; m(Mn3 O4 ) = 5.0 mg; 2 mL H2 O2 .

Using an initial H2 O2 concentration of 0.90 M (2 mL) and 5.0 mg of catalyst ([Mn3 O4 ] = 0.227 g L−1 ), comparable activities were displayed by Mn-A and Mn-B particles (see Fig. 8), but initial rates are higher with Mn-A nanoparticles. After 400 min, the maximum conversion is obtained for the two sets of catalysts, i.e. there is no further significant evolution of the relative absorbance A/A0 . In order to evaluate the influence of the hydrogen peroxide concentration, several experiments were conducted with different initial H2 O2 concentrations (see Fig. 9). When the H2 O2 concentration is increased from 0.24 to 1.29 M, the initial rate for the conversion of MB is monotonously enhanced. The A/A0 value after 400 min, which corresponds to the maximum conversion (vide supra) was also plotted as a function of the H2 O2 concentration. An optimum is clearly visible corresponding to 0.90 M. For higher initial concentration, i.e. 1.29 M, the final A/A0 value was found to be higher,

Time (h)

TC (mg C L−1 )

0 2 4 21

48.5 48.0 49.7 48.3

± ± ± ±

0.7 0.7 0.8 0.7

IC (mg C L−1 ) 0.1 2.8 3.4 3.8

± ± ± ±

0.1 0.1 0.1 0.1

DOC (mg C L−1 ) 48.4 45.2 46.3 44.5

± ± ± ±

0.8 0.8 0.9 0.8

indicating a lower conversion. This can be easily explained taking into account that at low H2 O2 concentrations, the HO• radicals preferentially react with the MB cation whereas for higher concentrations, some of the HO• are scavenged by excess H2 O2 , generating HO2 • radicals which exhibit much lower oxidation abilities than HO• [62]. This leads to a decrease in the degradation efficiency of the catalytic system. These results indicate that the dye was decolorized but do not imply its mineralization. We thus have performed total carbon (TC) and total inorganic carbon (TIC) measurements in order to assess a possible mineralization of MB. With some catalysts, e.g. TiO2 , a complete mineralization of MB has been reported under UV photocatalytic degradation [63] while with Fe3−x Tix O4 catalysts, only small amounts of MB were mineralized [64]. Our results (see Table 2) indicated only a very low diminution of the dissolved organic carbon content of the solution: even if the MB cation has been partially oxidized by HO• radicals, total mineralization to CO2 , SO4 2− , NH4 + and NO3 − is not effective with the Mn3 O4 nanocrystals. The fact that there is only a very small diminution of the total organic carbon can be explained considering that the hydroxyl radicals can indeed react by two different and competitive pathways [49]: (i) the oxidation of an organic substrate or (ii) the reaction of OH• with H2 O2 leading to the formation of O2 according to Eqs. (4) and (5). If hydroxyl radicals react preferentially following the second pathway, it can explain the low mineralization observed for the methylene blue molecule. To the best of our knowledge, there is no paper dealing with the degradation products of methylene blue using Mn3 O4 catalysts under Fenton conditions. Because of the homolytic nature of the reaction, the reaction is not selective and several products are expected. However, based on recent studies using electrospray mass spectrometry [28,65], some oxidation products have been identified. The MB degradation is thought to be initiated via OH• incorporation in the aromatic ring followed by cleavage of the dye into oxidative intermediate species. Finally, in order to ensure that the observed reactivity was not due to a partial dissolution of the nanoparticles, leading to an homogeneous system, a test was performed with a mixture of MnII and MnIII acetate salts in proportions corresponding to the spinel stoichiometry under homogeneous conditions. After 5 min, the MB conversion was about 14% but remained constant over the remaining time (see Fig. 8). This result shows that the activity observed for the heterogeneous system cannot be explained, or to a lesser extent, by manganese ions released from the particle surface (see also next section).

3.4. Re-use of the catalyst

Fig. 9. Influence of the H2 O2 concentration on the initial rate () of decomposition of MB and on the measured relative absorbance after 400 min reaction time ().

The re-use of the Mn-B catalyst was also investigated. Fig. 10 clearly shows that a quite similar activity is observed for the recycled catalyst as compared to the fresh catalyst. Furthermore, a TEM study performed on the particles recovered after two successive catalytic tests showed that the mean size of the particles was nearly the same than that of the starting powder (see Fig. 10, inset), discarding a possible partial dissolution of the nanocrystals.

138

T. Rhadfi et al. / Applied Catalysis A: General 386 (2010) 132–139

Fig. 10. Re-use of the Mn-B catalyst for the degradation of methylene blue using hydrogen peroxide: () first test and () second test. Experimental conditions: T = 293 K; 20.0 mL of a 100 mg L−1 methylene blue solution; m(Mn3 O4 ) = 5.0 mg; 2 mL H2 O2 . Inset: HRTEM image of the recovered nanocrystals after the catalysis tests.

4. Conclusion Mn3 O4 (hausmannite) nanocrystals were prepared using the polyol process. Depending on the reaction time, two sets of particles were obtained. With very short reaction times, oval-like 8 nm sized crystals were observed whereas with 16 h reaction times, rhomboedron-like particles with a mean size of 13 nm were generated. The particles were active towards the degradation of hydrogen peroxide via well-known Fenton reactions. For the two sets of particles an observed first-order reaction rate towards the H2 O2 concentration is found, but different kinetics were observed as a result of differences in the number and/or nature of the surface active sites. Methylene blue, a classical dye was used as model substrate to evaluate the activity of the catalysts. A nearly total discoloration was observed after 3 h. The particles are true heterogeneous catalysts and can be recycled without loss of activity and without noticeable morphologic modification. Work is in progress in order to compare the reactivity of these particles with other spinel type oxides such as Fe3 O4 or Co3 O4 which are studied in our group. Acknowledgements T. Rhadfi gratefully acknowledges the Agence Universitaire de la Francophonie for financial support. The authors would like to thank Pr. R. Losno for constructive discussions. References [1] A.R. Armstrong, P.G. Bruce, Nature 381 (1996) 499–500. [2] Q. Feng, H. Kanoh, K. Ooi, J. Mater. Chem. 9 (1999) 319–333. [3] S.S.T. Bastos, J.J.M. Orfao, M.M.A. Freitas, M.F.R. Pereira, J.L. Figueiredo, Appl. Catal. B 93 (2009) 30–37. [4] Y.-C. Son, V.D. Makwana, A.R. Howell, S.L. Suib, Angew. Chem. Int. Ed. 40 (2001) 4280–4283.

[5] V.D. Makwana, L.J. Garces, J. Liu, J. Cai, Y.-C. Son, S.L. Suib, Catal. Today 85 (2003) 225–233. [6] Y.-F. Han, F. Chen, K. Ramesh, Z. Zhong, E. Widjaja, L. Chen, Appl. Catal. B 76 (2007) 227–234. [7] Y.-F. Han, F. Chen, Z. Zhong, K. Ramesh, L. Chen, D. Jian, W.W. Ling, Chem. Eng. J. 134 (2007) 276–281. [8] Y.-F. Han, F. Chen, Z. Zhong, K. Ramesh, L. Chen, E. Widjaja, J. Phys. Chem. B 110 (2006) 24450–24456. [9] J. Luo, Q. Zhang, J. Garcia-Martinez, S.L. Suib, J. Am. Chem. Soc. 130 (2008) 3198–3207. [10] Y. Ren, Z. Ma, L. Qian, S. Dai, H. He, P.G. Bruce, Catal. Lett. 131 (2009) 146–154. [11] R. Ghosh, Y.-C. Son, V.D. Makwana, S.L. Suib, J. Catal. 224 (2004) 288–296. [12] X. Li, L. Zhou, J. Gao, H. Miao, H. Zhang, J. Xu, Powder Technol. 190 (2009) 324–326. [13] V.P. Santos, M.F.R. Pereira, J.J.M. Orfao, J.L. Figueiredo, Top. Catal. 52 (2009) 470–481. [14] G. Laugel, J. Arichi, H. Guerba, M. Moliere, A. Kiennemann, F. Garin, B. Louis, Catal. Lett. 125 (2008) 14–21. [15] M. Baldi, E. Finocchio, F. Milella, G. Busca, Appl. Catal. B 16 (1998) 43–51. [16] I. Atribak, A. Bueno-Lopez, A. Garcia-Garcia, P. Navarro, D. Frias, M. Montes, Appl. Catal. B 93 (2010) 267–273. [17] M. Kang, E.D. Park, J.M. Kim, J.E. Yie, Appl. Catal. A 327 (2007) 261–269. [18] M.A. Hasan, M.I. Zaki, L. Pasupulety, K. Kumari, Appl. Catal. A 181 (1999) 171–179. [19] L.-X. Yang, Y.-J. Zhu, H. Tong, W.-W. Wang, G.-F. Cheng, J. Solid State Chem. 179 (2006) 1225–1229. [20] N. Sui, Y. Duan, X. Jiao, D. Chen, J. Phys. Chem. C 113 (2009) 8560–8565. [21] E. Neyens, J. Baeyens, J. Hazard. Mater. 98 (2003) 33–50. [22] M. Hermanek, R. Zboril, I. Medrik, J. Pechousek, C. Gregor, J. Am. Chem. Soc. 129 (2007) 10929–10936. [23] L. Gao, J. Zhuang, L. Nie, J. Zhang, Y. Zhang, N. Gu, T. Wang, J. Feng, D. Yang, S. Perrett, X. Yan, Nat. Nanotechnol. 2 (2007) 577–583. [24] H. Lim, J. Lee, S. Jin, J. Kim, J. Yoon, T. Hyeon, Chem. Commun. (2006) 463–465. [25] M.B. Kasiri, H. Aleboyeh, A. Aleboyeh, Appl. Catal. B 84 (2008) 9–15. [26] K. Fajerwerg, H. Debellefontaine, Appl. Catal. B 10 (1996) L229–L235. [27] G. Centi, S. Perathoner, T. Torre, M.G. Verduna, Catal. Today 55 (2000) 61–69. [28] T. Sriskandakumar, N. Opembe, C.-H. Chen, A. Morey, C. King’ondu, S.L. Suib, J. Phys. Chem. A 113 (2009) 1523–1530. [29] S.-H. Do, B. Batchelor, H.-K. Lee, S.-H. Kong, Chemosphere 75 (2009) 8–12. [30] Z. Yang, Y. Zhang, W. Zhang, X. Wang, Y. Qian, X. Wen, S. Yang, J. Solid State Chem. 179 (2006) 679–684. [31] K. Konstantinov, Y. Tournayre, H.K. Liu, Mater. Lett. 61 (2007) 3189–3192. [32] A. Askarinejad, A. Morsali, Ultrason. Sonochem. 16 (2008) 124–131. [33] I. Bilecka, I. Djerdj, M. Niederberger, Chem. Commun. (2008) 886–888. [34] Z. Fang, K. Tang, L. Gao, D. Wang, S. Zeng, Q. Liu, Mater. Res. Bull. 42 (2007) 1761–1768. [35] Z. Weixin, W. Cheng, Z. Xiaoming, X. Yi, Q. Yitai, Solid State Ionics 117 (1999) 331–335. [36] L.-X. Yang, Y. Liang, H. Chen, Y.-F. Meng, W. Jiang, Mater. Res. Bull. 44 (2009) 1753–1759. [37] I. Djerdj, D. Arcon, Z. Jaglicic, M. Niederberger, J. Phys. Chem. C 111 (2007) 3614–3623. [38] T. Ozkaya, A. Baykal, H. Kavas, Y. Koeseoglu, M.S. Toprak, Phys. B: Condens. Matter 403 (2008) 3760–3764. [39] P. Gibot, L. Laffont, J. Solid State Chem. 180 (2007) 695–701. [40] W.S. Seo, H.H. Jo, K. Lee, B. Kim, S.J. Oh, J.T. Park, Angew. Chem. Int. Ed. 43 (2004) 1115–1117. [41] D.-S. Wang, T. Xie, Q. Peng, S.-Y. Zhang, J. Chen, Y.-D. Li, Chem. Eur. J. 14 (2008) 2507–2513. [42] N. Zhao, W. Nie, X. Liu, S. Tian, Y. Zhang, X. Ji, Small 4 (2008) 77–81. [43] N. Wang, L. Guo, L. He, X. Cao, C. Chen, R. Wang, S. Yang, Small 3 (2007) 606–610. [44] C. Feldmann, Adv. Funct. Mater. 13 (2003) 101–107. [45] L. Sicard, J.-M. LeMeins, C. Méthivier, F. Herbst, S. Ammar, J. Magn. Magn. Mater. 322 (2010) 2634–2640. [46] L. Lutterotti, S. Matthies, H.R. Wenk, MAUD: a friendly Java program for material analysis using diffraction, IUCr: Newsletter of the Commission on Powder Diffraction, 1999, p. 14. [47] S. Brunauer, P.H. Emmett, E. Teller, J. Am. Chem. Soc. 60 (1938) 309–319. [48] N. Shimizu, C. Ogino, M.F. Dadjour, K. Ninomiya, A. Fujihira, K. Sakiyama, Ultrason. Sonochem. 15 (2008) 988–994. [49] R.C.C. Costa, M.F.F. Lelis, L.C.A. Oliveira, J.D. Fabris, J.D. Ardisson, R.R.V.A. Rios, C.N. Silva, R.M. Lago, J. Hazard. Mater. B 129 (2006) 171–178. [50] A.Y. Satoh, J.E. Trosko, S.J. Masten, Environ. Sci. Technol. 41 (2007) 2881–2887. [51] C. Ogino, M.F. Dadjour, Y. Iida, N. Shimizu, J. Hazard. Mater. 153 (2008) 551–556. [52] A.-N. Chowdhury, M.S. Azam, M. Aktaruzzaman, A. Rahim, J. Hazard. Mater. 172 (2009) 1229–1235. [53] K.S.W. Sing, D.H. Everett, R.A.W. Haul, L. Moscou, R.A. Pierotti, J. Rouquerol, T. Siemieniewska, Pure Appl. Chem. 57 (1985) 603–619. [54] M.K. Eberhardt, R. Colina, J. Org. Chem. 53 (1988) 1071–1074. [55] F. Magalhaes, M.C. Pereira, S.E.C. Botrel, J.D. Fabris, W.A. Macedo, R. Mendonca, R.M. Lago, L.C.A. Oliveira, Appl. Catal. A 332 (2007) 115–123. [56] R.C.C. Costa, M.d.F.F. Lelis, L.C.A. Oliveira, J.D. Fabris, J.D. Ardisson, R.R.V.A. Rios, C.N. Silva, R.M. Lago, Catal. Commun. 4 (2003) 525–529. [57] M. Falcon, B. Peyrille, P. Reilhac, J.N. Foussard, H. Debellefontaine, Rev. Sci. Eau 6 (1993) 411–426. [58] S.-S. Lin, M.D. Gurol, Environ. Sci. Technol. 32 (1998) 1417–1423.

T. Rhadfi et al. / Applied Catalysis A: General 386 (2010) 132–139 [59] T. Valdes-Solis, P. Valle-Vigon, S. Alvarez, G. Marban, A.B. Fuertes, Catal. Commun. 8 (2007) 2037–2042. [60] D.Y. Murzin, Chem. Eng. Sci. 64 (2009) 1046–1052. [61] M. Ocana, Colloid Polym. Sci. 278 (2000) 443–449. [62] K. Dutta, S. Mukhopadhyay, S. Bhattacharjee, B. Chaudhuri, J. Hazard. Mater. 84 (2001) 57–71.

139

[63] A. Houas, H. Lachheb, M. Ksibi, E. Elaloui, C. Guillard, J.-M. Herrmann, Appl. Catal. B 31 (2001) 145–157. [64] S. Yang, H. He, D. Wu, D. Chen, X. Liang, Z. Qin, M. Fan, J. Zhu, P. Yuan, Appl. Catal. B 89 (2009) 527–535. [65] L.C.A. Oliveira, M. Goncalves, M.C. Guerreiro, T.C. Ramalho, J.D. Fabris, M.C. Pereira, K. Sapag, Appl. Catal. A 316 (2006) 117–124.