ISSN 0036-0244, Russian Journal of Physical Chemistry A, 2007, Vol. 81, No. 9, pp. 1511–1514. © Pleiades Publishing, Ltd., 2007. Original Russian Text © O.V. Udalova, E.V. Khaula, Yu.N. Rufov, 2007, published in Zhurnal Fizicheskoi Khimii, 2007, Vol. 81, No. 9, pp. 1697–1701.

CHEMICAL KINETICS AND CATALYSIS

The Heterogeneous Catalytic Oxidation of Propylene in the Presence of Singlet Oxygen in the Reaction Mixture O. V. Udalova, E. V. Khaula, and Yu. N. Rufov Semenov Institute of Chemical Physics, Russian Academy of Sciences, ul. Kosygina 4, Moscow, 117977 Russia e-mail: [email protected] Received July 24, 2006

Abstract—Three methods for the introduction of singlet oxygen into the reaction mixture were tested, including thermal generation of singlet oxygen on the catalyst itself, the introduction of singlet oxygen from an external source, and photogeneration of singlet oxygen on the catalyst. Zeolites with admixtures of Mo, Bi, V, and Ni and SiO2 with deposited Mo, V, and Bi were used. Common to all reactions was an increase in the yield of deep oxidation products in the presence of singlet oxygen. A sharp increase in the yield of mild oxidation products was observed in the oxidation of propylene on a Bi/SiO2 catalyst. The generation of singlet oxygen under irradiation at 240–260 nm was found to cause deep oxidation only. Mild oxidation products could only form under the action of total mercury lamp light. DOI: 10.1134/S0036024407090324

INTRODUCTION The reactivity of electronically excited molecules in heterogeneous catalysis has been studied poorly. The main reasons for this is comparatively short lifetimes and low concentrations of excited particles under experimental conditions. In this respect, singlet oxygen é2(1∆g) has certain advantages over other excited molecules, because its radiative lifetime is 64.6 min at a 1 + 0.977 eV energy. The O2( Σg ) excited state has the radiative lifetime ~7 s at a 1.626 eV internal energy [1] and is therefore virtually ignored in studies of heterogeneous reactions. The interest of researchers in é2(1∆g) is also caused by its electronic structure responsible for the selective oxidation of organic molecules at double bonds. These properties inspire hopes that heterogeneous catalytic reactions with the participation of singlet oxygen should have special features of their own. There are several molecules with comparable radiative lifetimes (Ne, 430 s; Xe, 149 s; and C, 3230 s), but they are of no interest for heterogeneous catalysis. Studies of the role played by é2(1∆g) in heterogeneous oxidation were performed either at low pressures [2] or at low concentrations in reaction mixtures [3–5]. The main difficulty of such studies is that of increasing the concentration of é2(1∆g). We used several techniques for increasing the concentration of 1é2 in reaction mixtures. These were: (1) The selection of catalysts with maximum thermal generation of é2(1∆g) at the catalysis temperature. (2) The introduction of singlet oxygen into the reaction mixture from an external source.

(3) Irradiation of catalysts with maximum photogeneration of é2(1∆g) during the reaction. We tested Y-type zeolites with admixtures of Zn, Ni, and Ca and silica gel and Al2O3 with V, Bi, and Mo ions deposited on them. EXPERIMENTAL The starting material for the preparation of substituted zeolites was the Na form of zeolite Y with the ratio Si/Al = 2.6 synthesized from aqueous silica alumina gels at 80–110°ë. The Zn form of zeolite Y was obtained by singly treating NaY with a solution of Zn(NO3)2 · 6H2O at room temperature. The amount of Zn in the solution was 10 times higher than the equivalent amount of Na in the zeolite. The duration of treatment was 24 h. In all experiments, the zeolite sample weight was 3 g, and the volume of the solution was 200 ml. The NiY and CaY zeolite forms were prepared by treating NaY (3 g) with three portions of a solution of Ni(NO3)2 · 6H2O or CaCl2 (100 ml) with intermediate decantation. Exchange was performed at 60°ë for 4 h. Each portion contained the amount of Ni or Ca equivalent to the amount of Na in the zeolite. We also studied a series of catalysts which were transition metal oxides deposited on silica gel, namely, V2O5/SiO2, MoO3/SiO2, and Bi2O3/SiO2. The substrate was silica gel Silochrom S-120. Deposition was performed by precipitation from aqueous solutions. The amount of a salt necessary for the preparation of a solution was calculated to obtain a 1.5 mol % concentration of deposited metal oxides with respect to silica gel (NM/NSi + NM). Silica gel (fraction 0.15–0.25 mm) was placed into a solution of the corresponding salt. The

1511

1512

UDALOVA et al. C3H6/Air Water O2/He 3 2 1 5 4

Discharge

140°ë He For analysis

6

Fig. 1. Quartz reactor for studying the heterogeneous catalytic oxidation of propylene with é2(1∆g) thermally generated on the catalyst itself or introduced into the reaction mixture from an external source: 1, catalyst bed; 2, pocket for a thermocouple; 3, halogen lamps; 4, six-pass cock for taking samples for chromatography; and 5, reactor furnace.

Exit

1

2 Fig. 2. Quartz reactor for studying the heterogeneous catalytic oxidation of propylene with é2(1∆g) photogenerated on the catalyst itself: 1, catalyst bed and 2, thermocouple.

lnV [arb. units] 4

3

V2O5 /SiO3 MoO3 /SiO2 SiO2 BiO3 /SiO2

2

1

0

1.53

1.74

2.75 103/T, K–1

Fig. 3. Temperature dependence of the photogeneration of é2(1∆g) for transition metal oxides deposited on silica gel.

solution was periodically stirred with a magnetic stirrer while water was evaporated during 12 h at 110°ë. The samples were then calcined for 5 h at 400 and 500°ë. Bismuth was deposited from a solution of bismuth nitrate and vanadium and molybdenum, from solutions of ammonium vanadate and molybdate [6]. The heterogeneous catalytic oxidation of propylene while é2(1∆g) was thermally generated on the catalyst was studied on a flow unit, which was a quartz reactor of volume 20 cm3 with a pocket for a thermocouple (Fig. 1). The diameter of the reactor was 12 mm, and the mean catalyst bed length was ~1 cm. The thermocouple was in the center of the catalyst bed. The oxidation products were transferred from the reactor directly into the thermostat of a chromatograph heated to 140°ë. The catalytic activity of samples in the presence of singlet oxygen from an external source was studied on a reactor with an additional entrance sealed into it. It was found experimentally that, at the catalysis temperature (~400°ë), singlet oxygen did not decay before it entered the reactor because of singlet oxygen generation on heated reactor quartz walls. The photochemical singlet oxygen generator was a glass tube 180 mm long and 15 mm in diameter. Its inside surface was frosted and treated with an alcoholic solution of methylene blue. The tube had a water jacket over the whole its perimeter. Irradiation was performed by six halogen lamps of power 50 W. é2(1∆g) was introduced into the reaction mixture with a flow of a He–O2 mixture. The optimum concentration of O2 (taking into account the decay of é2(1∆g) in the gas phase) was ~ 2.5%. At a flow rate of 210 cm3/min, the concentration of photogenerated é2(1∆g) was 1.2 × 1013 molecules/cm3. Specially selected catalysts were irradiated during the heterogeneous catalytic oxidation of propylene using a quartz reactor (Fig. 2) of volume 115 cm3 and diameter 3.5 cm. The reactor had a partition with a plate of silica, on which catalysts were placed. The thermocouple was inside the silica layer. The source of ultraviolet radiation was a PRK-8 mercury lamp. Singlet oxygen é2(1∆g) was continuously determined by the chemiluminescent method. The reaction of é2(1∆g) with 10-methoxymethyl-101-methyl-9,91biacridylidene causes the emission of a quantum recorded by a photoelectron multiplier. The photoelectron multiplier signal is amplified and transmitted to a computer through an analog-to-digital converter. A detailed description of the unit is given in [7]. At an air pressure in the reactor of 80–90 torr, the flow rate was 3–3.3 cm3/s. This was an optimum pressure. At higher pressures, the decay of singlet oxygen in the gas phase was observed. RESULTS AND DISCUSSION A comparison of the experimental data on the influence of singlet oxygen on the catalytic oxidation of pro-

RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A

Vol. 81

No. 9

2007

THE HETEROGENEOUS CATALYTIC OXIDATION OF PROPYLENE

pylene was performed after selecting conditions that most clearly revealed 1é2 effects. The photocatalytic oxidation of propylene had to be conducted at lowered temperatures, because the photogeneration of é2(1∆g) decreased above 200–250°ë on all the catalysts studied. The temperature dependence of the photogeneration of é2(1∆g) can be caused by the chemisorption mechanism of singlet oxygen deactivation. According to [8], the desorption of water vapor empties singlet oxygen deactivation centers. This process likely occurs at comparatively low temperatures (up to ~100°ë). At higher temperatures, a decrease in the generation of é2(1∆g) is observed for all samples with a small spread in activation energies (4–7 kcal/mol). This decrease may be related to the desorption of adsorbed gases from the surface. Nor can we rule out the possibility that, under irradiation, excitation energy is dissipated into the solid without the formation of singlet oxygen (Figs. 3 and 4). We were unable to determine the concentration of singlet oxygen participating in the catalytic reaction for

lnV [arb. units] 2.5 2.0 1.5

1513

CaY ZnY NiY

1.0 0.5 0

1.52

1.78

2.16

–0.5

2.72 103/T, K–1

Fig. 4. Temperature dependence of the photogeneration of é2(1∆g) for substituted zeolite Y.

the following reasons. First, the reaction occurred at atmospheric pressure, whereas the concentration of singlet oxygen was determined at lowered pressures (~ 80 torr) to exclude the decay of 1é2 in the gas phase. This difficulty can formally be circumvented by mere

Heterogeneous catalytic propylene oxidation products at various techniques for supplying singlet oxygen Sample

SiO2 ZnY NiY MoO3 /SiO2 (1.5%)

V2O5 /SiO2 (1.5%)

Bi2O3 /SiO2 (1.5%)

CaY ZnY NiY

IllumiN × 10–11 x, % nation

[CO], % E

S

[CO2], % E

S

[C3H4O], % E

S

[C2H4O2], % E

S

Singlet oxygen photogeneration at 200°C 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.9 0.7 0.0 0.0 0.0 0.0 0.3 40.1 0.2 32.3 45.1 7.9 17.6 34.2 75.9 1.4 3.0 0.3 0.6 5.1 46.5 11.3 24.3 32.1 69.0 1.7 3.7 0.4 0.8 65.7 9.1 13.8 55.1 83.9 0.2 0.4 0.3 0.4 0.7 65.5 17.3 26.4 46.1 70.5 0.4 0.6 0.6 0.9 0.8 0.0 0.0 0.0 0.0 0.1 9.9 0.7 82.6 0.03 6.1 2.6 42.9 2.7 44.2 0.1 1.7 0.5 9.0 6.0 1.4 23.9 3.6 60.3 0.2 2.7 0.5 8.1 5.6 1.3 23.6 3.7 65.8 0.0 0.4 0.4 6.6 12.5 5.1 40.9 5.1 40.9 1.0 8.1 0.4 3.1 0.4 15.6 6.4 40.8 7.4 47.2 1.2 7.5 0.3 1.8 12.3 4.3 34.7 6.6 53.6 1.0 8.5 0.3 2.6 7.3 1.5 20.4 5.6 75.9 0.2 2.2 0.0 0.5 6.0 1.8 30.6 3.8 38.9 0.1 2.5 0.2 3.4 0.6 7.8 2.7 34.6 3.8 48.7 0.9 11.3 0.3 3.5 5.7 2.5 44.1 2.9 53.9 0.1 2.1 0.1 1.7 3.3 0.9 27.9 2.2 35.4 0.0 0.0 0.1 3.5 7.6 2.1 28.1 4.5 24.7 0.8 10.0 0.1 1.5 O2(1∆g) thermal generation and photogeneration from an external source at 420°C <1 78.8 30.9 39.6 45.5 58.3 0.0 0.0 0.8 1.1 VI 123 79.7 42.4 53.3 55.8 70.0 0.0 0.0 1.0 1.2 223 32.1 26.0 81.3 32.2 98.1 0.0 0.0 1.3 4.1 VI 346 35.1 31.1 88.5 27.5 78.1 0.0 0.0 1.2 3.5 74 70.2 17.7 25.2 51.7 73.7 12.1 17.2 0.6 0.9 VI 197 70.5 24.5 34.8 52.1 73.9 6.5 9.2 0.5 0.7 I II I II I II I II III IV I II III IV I II III IV V

[C3H4O2], % E

S

0.0 0.2 1.3 1.0 0.9 1.0 0.1 0.1 0.3 0.2 0.9 0.4 0.1 0.1 0.0 0.1 0.1 0.0 0.1

0.0 27.7 2.9 2.1 1.4 1.6 7.6 2.3 5.0 3.7 6.8 2.8 0.6 1.1 0.4 1.2 1.1 1.4 1.1

0.0 0.0 0.0 0.0 0.0 0.0

0.0 0.0 0.0 0.0 0.0 0.0

Note: x is the conversion, E is the yield, S is the selectivity, and N is the concentration of singlet oxygen (molecules/cm3); I, without irradiation by a mercury lamp; II, with irradiation by a mercury lamp; III, mercury lamp with a UFS-1 filter; IV, mercury lamp with a BS-8 filter; V, irradiation by a mercury lamp and a halogen lamp; and VI, thermal generation plus external singlet oxygen source. RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A

Vol. 81

No. 9

2007

1514

UDALOVA et al.

extrapolation of the concentration of 1é2 to atmospheric pressure. Secondly, the question of the real concentration of singlet oxygen participating in the catalytic reaction remains open, because 1é2 can decay on the path to the catalytic center. The results presented in the table show that the formation of é2(1∆g) occurs most effectively under UV irradiation at wavelengths below 240 nm virtually on all the systems studied, both deposited and massive. This leads us to suggest that 1é2 is formed in charge transfer complexes by the mechanism described in [1]. Charge transfer was suggested in [7, 9] for the formation of ëé2. Additional catalyst illumination in the visible range insignificantly increased the yield of ëé2. The formation of ëé2 possibly occurs as a result of charge transfer complex excitation [9]. An analysis of the table shows that, irrespective of the method for increasing the concentration of 1é2, this increase (as a first approximation) results in an increase in the amount of deep oxidation products (CO and CO2). There was one exception: the yield of acrolein on Bi2O3/SiO2 became higher by almost an order of magnitude as the conversion increased by a factor of 1.5. The X-ray patterns of the Bi2O3/SiO2 sample at a 6 at % Bi2O3 concentration were indicative of the formation of the SiBiO5 mobile phase, which might be responsible for the catalytic process. The superequilibrium singlet oxygen generation at elevated temperatures is likely also related to this phase [10]. All the above assumptions concerning the concentration of singlet oxygen participating in the reaction being valid, its concentration is several orders of magnitude lower than that of the main reagents. The influence of singlet oxygen observed can therefore be explained by either the modification of catalytic centers with it or the release of dioxethane decomposition products into the volume and volume reactions with

their participation [11]. A more thorough analysis of the reaction requires performing it at comparable singlet oxygen and main reagent concentrations. ACKNOWLEDGMENTS This work was financially supported by the Russian Foundation for Basic Research (project no. 04-0332513). REFERENCES 1. M. V. Yakunichev, I. A. Myasnikov, and V. I. Tsivenko, Zh. Fiz. Khim. 60 (4), 1017 (1986). 2. M. V. Yakunicheva, Candidate’s Dissertation in Physics and Mathematics (Karpov Research Institute of Phys. Chem., Moscow, 1986). 3. O. V. Udalova, E. V. Khaula, M. Ya. Bykhovskii, et al., Zh. Fiz. Khim. 77 (6), 1018 (2003) [Russ. J. Phys. Chem. 77 (6), 912 (2003)]. 4. M. V. Vishnetskaya, A. N. Emel’yanov, N. V. Shcherbakov, et al., Zh. Fiz. Khim. 78 (12), 2152 (2004) [Russ. J. Phys. Chem. 78 (12), 1918 (2004)]. 5. M. V. Vishnetskaya, A. N. Emel’yanov, and N. V. Shcherbakov, Khim. Fiz. 23, 40 (2004). 6. H. Yoshida, C. Murata, and T. Hattori, J. Catal. 194, 364 (2000). 7. A. N. Romanov and Yu. N. Rufov, Zh. Fiz. Khim. 72 (11), 2094 (1998) [Russ. J. Phys. Chem. 72 (11), 1908 (1998)]. 8. M. E. Ryskin, T. Y. Kurenyova, V. G. Kustarev, et al., Abstracts of Papers, 8th International Congress on Catalysis (West Berlin, 1984), Vol. 3, p. 311. 9. L. D. Kreuzke and G. W. Keulke, J. Catal. 64, 295 (1980). 10. N. V. Shcherbakov, A. N. Emel’yanov, E. V. Khaula, et al., Zh. Fiz. Khim. (in press). 11. M. J. S. Dewar and W. Thiel, J. Am. Chem. Soc. 97, 3978 (1975).

RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A

Vol. 81

No. 9

2007