ISSN 1061933X, Colloid Journal, 2012, Vol. 74, No. 1, pp. 85–90. © Pleiades Publishing, Ltd., 2012. Original Russian Text © V.I. Lesin, L.M. Pisarenko, O.T. Kasaikina, 2012, published in Kolloidnyi Zhurnal, 2012, Vol. 74, No. 1, pp. 90–95.
Colloidal Catalysts Based on Iron (III) Oxides. 1. Decomposition of Hydrogen Peroxide V. I. Lesina, L. M. Pisarenkob, and O. T. Kasaikinab a
b
Institute of Oil and Gas Problems, Russian Academy of Sciences, ul. Gubkina 3, Moscow, 119333 Russia Semenov Institute of Chemical Physics, Russian Academy of Sciences, ul. Kosygina 4, Moscow, 119991 Russia email:
[email protected],
[email protected] Received November 1, 2010
Abstract—The catalytic activity of a colloidal catalyst based on iron(III) oxides in decomposition of H2O2 is studied. The catalyst is obtained by hydrolysis followed by peptization of FeCl3 ⋅ 6H2O salt in water in the presence of 1% ethanol. The structure, composition, and size of colloidal particles of the catalyst are studied by the methods of Mössbauer spectroscopy, Xray fluorescence, Xray diffraction analysis, and transmission electron microscopy. The obtained catalyst is based on αFe2O3 crystals with an admixture of other crystalline structures of iron oxides and carboncontaining compounds. The activity of the catalyst with respect to H2O2 decomposition undergoes nonlinear and nonmonotonic variations and its particle size enlarges beginning from 1 to 3 nm with increasing initial concentration of FeCl3 ⋅ 6H2O. The catalyst obtained under optimal conditions exhibits high activity corresponding to the most efficient agents of H2O2 decomposition. DOI: 10.1134/S1061933X12010103
INTRODUCTION
colloidal particles [14–16]. It has been assumed that it is due to the presence of inclusions of iron oxide– based colloidal particles that the treatment of oil and oil products with magnetic field decreases their viscos ity, decelerates paraffin deposition, accelerates water separation, etc. [14–18]. The analysis of these facts was the starting point for developing iron oxidebased polyfunctional and envi ronmentally safe catalysts for oxidative decomposition of organic residues of plant with the use of Н2О2 and/or atmospheric air as an oxidizing agent. Under the action of transition metals, Н2О2 is decomposed to yield active hydroxyl radicals НО•, which easily detach hydrogen atoms from the majority of organic molecules and are added to unsaturated bonds to yield freeradical products. In the presence of oxygen, the radicals being formed can continue the chain process of oxidative decomposition of organic components [7, 19–21]. Colloidal catalysts based on iron oxides exhibit high activity in oxidative decomposition of lignocellu lose biomasses of sawdust, peat, technical lignin, and straw and demonstrate the capability of “selftuning” with respect to a specific substrate [22]. The activity of metallocomplex catalysts is known to strongly depend on ligand structure, which may be changed in the course of a reaction as a result of interaction with a substrate and products of its transformations. To some extent, the kinetics of Н2О2 decomposition reflects the changes to which a catalyst undergoes during the reac tion. The rate of Н2О2 decomposition was shown to be
Catalysts based on iron oxides are widely used in various chemical processes due to their low toxicity and relatively low cost. In recent years, in connection with the development of nanotechnologies, highly dis persed heterogeneous catalysts based on iron oxides are tested in the processes of oxidation and postoxida tion of organic pollutants in water with both atmo spheric oxygen [1–3] and an environmentally safe oxidizing agent, hydrogen peroxide [4–7]. It is known that, in nature, organic pollutants (such as wood waste and even lignin, which is very difficult to chemically decompose) are processed by microorgan isms. A necessary condition for microorganism devel opment, reproduction, and functioning is the pres ence of water containing metal ions [8–11]. Moreover, hydrogen peroxide and atmospheric air are involved in the mechanism of enzymatic oxidative decomposition [12]. As a rule, natural water contains dissolved organic substances, such as products of biomass oxida tion, hydrogen peroxide, and iron compounds, which, to some extent, facilitate its selfpurification [13]. It was demonstrated [7] that catalysts based on nanopar ticles of iron(III) oxide in combination with Н2О2 make it possible to remove phenol and ethylene glycol from water. It has been known that particles of iron oxides are present in many minerals, soil, petroleum, etc. It was determined by direct methods that mag netic nanoparticles of iron oxides are present in petro leum as components of fractal aggregates of organic 85
86
LESIN et al. ×105, 2O2
[H2O2]/[H2O2]0 1.1 1.0 0.9 0.8 0.7 0.6 0.5 8 0.4 0.3 3 0.2 0.1 0
1
2
WH 1
М/s
25 20 15
2
10 5 4, 5, 6
7
0 3
4
5
6
7 t, h
Fig. 1. Kinetic curves for Н2О2 consumption in the pres ence of the catalyst prepared using different amounts of FeCl3 ⋅ 6H2O: (1) 0.014, (2) 0.062, (3) 0.145, (4) 0.214, (5) 0.312, (6) 0.412, and (7) 0.667 g; temperature, 60°С and initial Н2О2 concentration, 1.2 M. Curve 8 represents the consumption of 1.2 M Н2О2 in the presence of 0.014 g of untreated FeCl3 ⋅ 6H2O.
directly related to the rate of decomposition of an organic component (pollutant) in water [7, 19]. The structure and properties of a catalyst surface can play a significant role in the case of heterogeneous catalysts. The results of studying the structure and catalytic activity of iron(III) oxidebased colloidal catalyst in Н2О2 decomposition in the absence of organic sub strates are presented in this paper. The main attention is focused on the effect of the concentration of the iron salt used to synthesize the catalyst on its structure and activity. EXPERIMENTAL FeCl3 ⋅ 6H2O (Merck KGaA, Germany) and 30% hydrogen peroxide (analytical grade, Reakhim, Usol’ekhimprom) were used in the experiments. The catalyst was produced by hydrolyzing iron chloride in water containing a surfactant (1 wt %) [22]. A colloidal precipitate comprising the surfactant was formed. The amount of the salt was varied in a range of 0.014–0.67 g per 800 ml of H2O. Catalyst samples were studied by transmission electron microscopy (TEM) on an LEO 912 AB Omega microscope (Carl Zeiss); the structure of the catalyst was investigated by Xray diffraction analysis and Mössbauer spectroscopy at the Faculty of Physics, Moscow State University. In these experiments, a sus pension of particles was applied onto special substrates selected in accordance with the requirements for sam ple preparation for the corresponding study and kept at room temperature until complete drying up.
0.1 0.2 0.3 0.4 0.5 0.6 0.7 FeCl3 ×6H2O, g/100 ml
Fig. 2. Dependence of the initial rate of Н2О2 consump tion on the amount of FeCl3 ⋅ 6H2O used to synthesize the colloidal catalyst.
Since the catalyst + Н2О2 system is intended for oxidative decomposition of organic woodandplant mass, the kinetics of Н2О2 decomposition was studied in an appreciably concentrated solution of this perox ide. The resulting colloidal catalyst and a 1.2 M H2O2 solution (100 ml) were placed into a glass vessel equipped with a magnetic stirrer and a reflux con denser and thermostated at 60°С. Samples were taken in the course of the reaction, and Н2О2 concentration was determined by iodometry. RESULTS AND DISCUSSION Figure 1 shows the kinetic curves of Н2О2 con sumption in the presence of colloidal catalyst samples obtained using different amounts of FeCl3 ⋅ 6H2O. It is clear that the catalyst sample prepared using 0.014 g of FeCl3 ⋅ 6H2O has almost no effect on Н2О2 consump tion (curve 1). Meanwhile, in the presence of the same amount of FeCl3 ⋅ 6H2O (0.014 g or 0.52 mM of Fe(3+) ions per 100 ml of the Н2О2 solution), 60% of hydrogen peroxide rapidly decomposes and the process ceases (curve 8). The catalysis factor formally calculated for Fe(3+) ions is, in this case, (1.2 × 0.6)/(5.2 × 10–4) = 1.4 × 103. Thus, the hydrolysis of iron chloride and the formation of colloidal particles result in qualitative changes in properties of the substance in terms of cat alyzing hydrogen peroxide decomposition. The colloidal catalyst synthesized at higher FeCl3 ⋅ 6H2O concentrations decomposes Н2О2 (Fig. 1, curves 2–7). Here, hydrogen peroxide is completely consumed, in contrast to the situation with the initial FeCl3 ⋅ 6H2O salt (curve 8). The initial rate of Н2О2 consumption to a certain extent characterizes the activity of the catalyst with respect to hydrogen perox ide decomposition. As can be seen from Fig. 2, the ini tial rate nonlinearly and nonmonotonically varies with COLLOID JOURNAL
Vol. 74
No. 1
2012
COLLOIDAL CATALYSTS BASED ON IRON (III) OXIDES
0
1
2
FeKa
3
4
5
FeKb
FEKesc
SKa SKb ClKa ClKb
CKa FeLaOKa FeLI
Pulse number 1000 002 900 800 700 600 500 400 300 200 100
6
7
8 9 10 Energy, keV
Fig. 3. Xray fluorescence spectrum of a sample that repre sents a dense cluster of particles formed on a silicon plate after water evaporating from colloidal catalyst synthesized from 0.1 g of FeCl3 ⋅ 6H2O. The spectrum has been obtained under the action of an electron beam in a scan ning electron microscope at accelerating voltage of 15 kV, current of 2.6 nA, and irradiation time of 100 s.
an increase in the amount of FeCl3 ⋅ 6H2O used to pre pare the colloidal catalysts. The samples synthesized from the 0.05–0.2g weighed portions of FeCl3 ⋅ 6H2O exhibit the highest activity. A set of analytical methods, including atomic force microscopy (AFM), scanning electron microscopy (SEM), TEM, Mössbauer spectroscopy, Xray diffrac tion analysis, and Xray fluorescence, were used to investigate the physicochemical characteristics of cat alyst colloidal particles. Figure 3 presents the characteristic Xray fluores cence spectrum of a catalyst sample produced from 0.05 g of FeCl3 ⋅ 6H2O. As follows from analysis of the spectrum, iron, oxygen, and carbon are the major components of catalyst particles. The molar fractions
20 μm
(а)
of carbon, iron, and oxygen are 30–35, 30–40, and 20–25%, respectively. As impurities, catalyst particles contain chlorine, sulfur, sodium, and other elements at concentrations of approximately 0.01–3%. It should be noted that the scanned sample was produced from a colloidal dispersion by evaporating the solvent. Therefore, the impurities of salts contained in water could distort the data on the composition of dispersed particles. This is evident from the nonuniform distri bution of impurities over the surface of the sample. Nevertheless, it is evident that the presence of carbon (the surfactant included in the particles) is essential for the synthesis of the colloidal catalyst. The investigation of catalyst colloidal particles by the TEM and AFM methods [23, 24] demonstrated that the particles have a high tendency toward aggre gation in the entire examined range of FeCl3 ⋅ 6H2O mass concentrations. The AFM examination using special probes showed that the particles posses mag netic moments, electrostatic charges, and sizes of sev eral nanometers. Significant magnetic moments of the particles in an aqueous solution are evident from par ticles concentrating in the regions of the highest mag netic field gradient during sedimentation onto the bot tom of a cylindrical vessel subjected to a magnetic field. The powder Xray diffraction method was applied to estimate the size of catalyst particles prepared from a 0.02g weighed portion of FeCl3 ⋅ 6H2O; the value of 3 nm was obtained from the measured width of the dif fraction line using the Debye–Scherrer formula. It should be noted that particles of this size are charac terized by diffraction bands (Fig. 4) that are so wide that some authors [23] consider them to have an amor phous structure. The bands in the diffraction spectrum became narrower with increasing initial concentration of FeCl3 ⋅ 6H2O. The crystalline structure of the parti cles was estimated to be close to that of αFe2O3 with
(b)
20 μm
Fig. 4. (a) Transmission electron microscopy and (b) powder Xray diffraction data for catalyst sample prepared from 0.02 g of FeCl3 ⋅ 6H2O. COLLOID JOURNAL
Vol. 74
No. 1
2012
87
88
LESIN et al. channels
N, pulses
1.00
710000
705000 0.99 700000 0.98 695000
–6
–4
–2 0 2 4 6 Скорость движения источника, mm/s
Fig. 5. Mössbauer spectrum recorded at room temperature for the catalyst sample synthesized from 0.15 g of FeCl3 ⋅ 6H2O.
an admixture of foreign crystals. The characteristic size of the crystals observed in the TEM images increased in accordance with decreasing width of the diffraction lines. Note that, during water evaporation from a droplet of the dispersion of the colloidal cata lyst, aggregates consisting of several hundreds and thousands of particles were formed on the surface of all used substrates (silicon, carbon, etc.). This phenome non hampered the estimation of particle sizes by both TEM and AFM methods. Analysis of the particle sizes obtained by different methods demonstrated that the average particle size increases from 1.5–3 to 5–9 nm with enlarging weighed portions of FeCl3 ⋅ 6H2O in a range of 0.018–0.18 g. Iron occurs in particles in the trivalent state. The Mössbauer spectrum of the synthesized colloi dal catalyst (Fig. 5) almost coincides with that pre sented in [25] for Н2О2 decomposition catalyst pro duced by the thermal treatment of iron oxalate dihy drate FeC2O4 ⋅ 2H2O in air. It was found [25] that the thermally induced solidphase oxalate decomposition results in the formation of amorphous Fe2O3 nanopar ticles followed by their gradual crystallization into hematite αFe2O3. Depending on the duration of heating, the particles thus obtained had different degrees of crystallinity and surface areas. The catalytic activity of synthesized iron(III) oxide nanoparticles in Н2О2 decomposition was studied. It was established that the activity of the catalyst nonmonotonically depends on its surface area. The maximum value of the effective rate constant of Н2О2 consumption turned out to be 26.4 × 10–3 min–1 (g/l)–1 at a relatively small
surface area of the sample (337 m2/g), but at a high degree of its crystallinity, which resulted from longer heating at 175°С. The authors of [25] claimed that this value of the effective rate constant is the highest among those published for Н2О2 decomposition under the action of ironcontaining catalysts. The Möss bauer spectrum similar to that presented in Fig. 5 cor responds to a lowcrystallinity sample of Fe2O3 nano particles [25]. Iron oxide Fe2O3 with carbon inclusions is likely to be the major component of the colloidal catalyst synthesized in this work. Assuming that all iron ions contained in initial salt FeCl3 ⋅ 6H2O pass into Fe2O3, the initial rates of Н2О2 consumption (W0 can be used to calculate the effective rate constant of H2O2 decomposition as kef = W0/[Н2О2]0mFe 2O3 , where mFe 2O3 is the maximum amount of Fe2O3 (g) per 1 l of the system. The plot kef versus the weighed portion of FeCl3 ⋅ 6H2O appeared to be a nonmonotonic curve with a maximum value of 116 × 10–3 min–1 (g/l)–1 (Fig. 6). This value of the constant is approximately four times as high as that obtained in [25], most likely at room temperature of 22–25°C (the experimental temperature was not mentioned in the cited article). In our experiments, the reaction mixture was thermostated at 60°С. The data in Fig. 6 demonstrate that the efficiency of the colloidal catalyst is a nonlinear function of its amount. The maximum activity is observed for the catalyst pre pared from 0.1–0.2 g of FeCl3 ⋅ 6H2O in 800 ml of water, and the estimated kef characterizes the resulting colloidal catalyst as a very efficient agent for hydrogen peroxide decomposition. COLLOID JOURNAL
Vol. 74
No. 1
2012
COLLOIDAL CATALYSTS BASED ON IRON (III) OXIDES
The Mössbauer spectra and magnetic properties were studied in [26, 27] for Fe2O3 particles obtained using the same method as in [25], i.e., the thermal decomposition of iron oxalate at different tempera tures, which was accompanied by changes in both the sizes and magnetic properties of the particles. Based on the performed measurements, the authors of [27] concluded that the firstorder magnetic phase transi tion occurs upon increasing particle size; i.e., super paramagnetic particles are transformed into magneti cally ordered ones. The Mössbauer spectrum corre sponding to the superparamagnetic particles, in which the magnetic moment vector undergoes rapid thermal fluctuations, has the form of a doublet similar to that shown in Fig. 5 for colloidal catalyst particles with sizes of 5–9 nm. The fluctuation rate dramatically decreases with increasing particle size, and the mag netic moment becomes fixed relative to a particle. By comparing the results of this study with the data reported in [26, 27], one may assume that the varia tions in the catalyst activity are associated with changes in both sizes and, probably, magnetic proper ties of the particles. It should be noted that almost invariable effective rate constant kef of decomposition, which is observed when weighed portions of FeCl3 ⋅ 6H2O equal to 0.27– 0.7 g per 800 ml of water are applied to synthesize the catalyst (Fig. 6), can be explained by the constant ratio between the surface area and volume of the particles. This situation takes place when the particle size remains unchanged, or when the particles are so flat that the side surface area can be ignored. It is notewor thy that small colloidal particles of iron oxides pre pared from 0.014 g of FeCl3 ⋅ 6H2O do not induce the decomposition of hydrogen peroxide, whereas the same amount of the iron salt causes the intensive reac tion (Fig. 1). Thus, the study of the structure and catalytic activ ity of an iron(III) oxide–based colloidal catalyst with respect to Н2О2 decomposition, with the catalysis being synthesized by hydrolyzing FeCl3 ⋅ 6H2O in water containing 1% of a surfactant, demonstrated the following. 1. The obtained catalyst is mainly composed of α Fe2O3 crystals with an admixture of other crystalline structures of iron oxides, as well as carboncontaining compounds. 2. The activity of the catalyst with respect to Н2О2 decomposition varies nonlinearly and nonmonotoni cally and its particle size grows starting from 1–3 nm with an increase in the initial concentration of FeCl3 ⋅ 6H2O used to synthesize the catalyst. 3. The catalyst obtained under the optimal condi tions demonstrates a high activity, which is at the level of the most efficient agents for Н2О2 decomposition. COLLOID JOURNAL
Vol. 74
No. 1
2012
89
kef , min–1 (g/l)–1 0.12 0.10 0.08 0.06 0.04 0.02 0
0.1
0.2
0.3
0.4
0.5 0.6 0.7 FeCl3 · 6H2O, g
Fig. 6. Dependence of the effective rate constant for Н2О2 decomposition on the amount of iron chloride used to pre pare catalysts.
ACKNOWLEDGMENTS We are grateful to Prof. A.S. Ilyushin and Dr. Sci. (Phys.–Math.) G.B. Khomutov (Moscow State Uni 4 versity) for assistance in studying the catalyst struc ture. REFERENCES 1. Dondur, V., Radic, N., Grbic, B., and Drofenic, M., Mater. Sci. Forum, 2006, vol. 518, p. 85. 2. Shin, E.J., Miser, D.E., Chan, W.G., and Hajaligol, M.R., Appl. Catal. B, 2005, vol. 61, p. 79. 3. Arena, F., Gatti, G., Stievano, L., Martra, G., Coluc cia, S., Frusteri, F., Spadaro, I., and Parmaliana, A., Catal. Today, 2006, vol. 117, p. 75. 4. Lin, S.S. and Gurol, M.D., Environ. Sci. Technol., 1998, vol. 32, p. 1417. 5. Cuzzola, A., Bertini, M., and Salvadori, P., Appl. Catal., vol. 36, p. 231. 6. Kwan, W.P. and Voelker, B.M., Environ. Sci. Technol., 2002, vol. 36, p. 1467. 7. Zelmanov, G. and Semiat, R., Water Res., 2008, vol. 42, p. 492. 8. Kirk, T.K., in Microbial Degradation of Organic Com pounds, Gibson, D.T., Ed., New York: Marcel Dekker, 1 1984, vol. 13, p. 399. 9. Boeran, W., Ralph, J., and Baucher, M., Annu. Rev. Plant Biol., 2003, vol. 54, p. 519. 10. Flaig, W., Geohim. Cosmochim. Acta, 1964, vol. 28, 2 p. 1523. 11. Kirk, T.K. and Farrell, R.L., Annu. Rev. Microbiol., 1987, vol. 41, p. 465. 12. Kersten, P.J. and Kirk, T.K., J. Bacteriol., 1987, 3 vol. 169, p. 2195. 13. Latifoglu, A. and Kilic, A., Fresenius Environ. Bull., 2002, vol. 11, p. 894.
90
LESIN et al.
14. Lesin, V.I., Koksharov, Yu.A., and Khomutov, G.B., Neftekhimiya, 2010, vol. 50, p. 114. 4 15. Lesin, V.I., Koksharov, Yu.A., and Khomutov, G.B., Neft. Khoz., 2009, no. 3, p. 95. 4 16. Lesin, V.I., Koksharov, Yu.A., and Khomutov, G.B., 5 Georesursy Geoenerg. Geopolit., 2010, no. 1. www.oilgas journal.ru 17. Lesin, V.I., Neft. Khoz., 2004, no. 1, p. 68. 18. Lesin, V.I., Neftepromysl. Delo, 2008, no. 1, p. 43. 19. Denisov, E.T. and Afanas’ev, I.B., Oxidation and Antiox idants in Organic Chemistry and Biology, Boca Raton: CRC, 2005. 20. Lipid Oxidation Pathways, vol. 2, KamalEldin, A. and Min, D.B., Eds., Champaign: AOCS, 2008. 21. Chou, S.S. and Huang, C.P., Appl. Catal. A, 1999, vol. 185, p. 237. 22. Lesin, V.I., Pisarenko, L.M., and Kasaikina, O.T., Pat. Appl. no. 2010105825/04/008255, Russia (19 February 2010). 4
23. Kanevskii, V.M., Vlasov, V.P., Lesin, V.I., and Musli mov, A.E., Abstracts of Papers, XIII nats. konf. po rostu kristallov (NKRK2008) (XIII Natl. Conf. on Crystal Growth), Moscow, 2008, p. 442. 24. Kanevskii, V.M., Vlasov, V.P., Lesin, V.I., and Musli mov, A.E., Abstracts of Papers, XIII nats. konf. po rostu kristallov (NKRK2008) (XIII Natl. Conf. on Crystal Growth), Moscow, 2008, p. 444. 25. Hermanek, M., Zboril, R., Medrik, I., Pechousek, J., and Gregor, C., J. Am. Chem. Soc., 2007, vol. 129, p. 10929. 26. Suzdalev, I.P., Buravtsev, V.N., Maksimov, Yu.M., Imshennik, V.N., Novichikhin, S.V., Matveev, V.V., and Plachinda, A.S., Ross. Khim. Zh., 2001, vol. 45, no. 3, p. 66. 27. Suzdalev, I.P., Maksimov, Yu.M., Imshennik, V.N., Novichikhin, S.V., Matveev, V.V., Tret’yakov, Yu.D., Lukashin, A.V, Eliseev, A.A., Malygin, A.A., and Sos nov, E.A., 2006, vol. 1, nos. 1–2, p. 134.
SPELL: 1. Dekker, 2. Flaig, 3. Bacteriol, 4. Khomutov, 5. Georesursy, Скорость движения источника COLLOID JOURNAL
Vol. 74
No. 1
2012
