Applied Clay Science 33 (2006) 1 6 www.elsevier.

com/locate/clay

A new application of clay-supported vanadium oxide catalyst to selective hydroxylation of benzene to phenol
Xiaohan Gao, Jie Xu
State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian, 116023, PR China Received 11 October 2005; received in revised form 13 December 2005; accepted 19 December 2005 Available online 27 January 2006

Abstract The highly selective direct hydroxylation of benzene with hydrogen peroxide to phenol was observed on a clay-supported vanadium oxide catalyst. Under mild reaction conditions at 313 K, high selectivity to phenol of 94% was realized on the claysupported vanadium oxide catalyst at 14% conversion of benzene, comparable with the well-known TS-1 catalyst. On the other hand, vanadium on other supports and other metals on the clay support gave inferior catalytic performances. VOAl and VOSi bridges in the catalyst may be responsible for the reaction. 2005 Elsevier B.V. All rights reserved.
Keywords: Clay; Benzene; Phenol; Vanadium

1. Introduction Developing new applications of the natural resources is an interesting field. Clay that is a very versatile material has attracted more attention owing to its wide applications in various field for more than a century (Shimizu et al., 2002; Murray, 2000). The application of clay for catalysts in the organic chemical commenced at the 1970s last century, Pinnavaia et al. reported the hydrogenation of 1-hexene catalyzed by rhodium complexes immobilized in smectites (Pinnavaia and Welty, 1975). Many papers and monographs were published in recent years about the use of clay in the organic syntheses (Campanati and Vaccai, 2001). Clays have a great potential as host compounds for fancy catalysts such as shape selective and asymmetric
Corresponding author. Tel.: +86 411 84379245; fax: +86 411 84379245. E-mail address: xujie@dicp.ac.cn (J. Xu). 0169-1317/$ - see front matter 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.clay.2005.12.002

reactions due to the characteristic layer structure and the facile intercalation properties. Some important results in this aspect are indicated (Kun et al., 2001; Bartok et al., 1999; Shimazu et al., 1996). Additional advantage of using clay as support of catalysts is the composition of clay. Clay is mainly made up of tetrahedral silicon dioxide and octahedral alumina and the ratio of the Si / Al is relatively steady (Campanati and Vaccai, 2001), which is similar to molecular sieves. The hydroxylation of alkanes and aromatic compounds is an attractive and challenging subject, especially for the direct hydroxylation of benzene to phenol from economical and environmental point of view, which is an important intermediate for the manufacture of petrochemicals, agrochemicals and plastics (Miyahara et al., 2001). The investigations of direct oxidation of benzene to phenol with various oxidants, such as nitrous oxide (Sobolev et al., 1993; Panov et al., 1992), hydrogen peroxide (Bianchi et al., 2000; Bengoa et al., 1998; Zhang et al., 1996;

X. Gao, J. Xu / Applied Clay Science 33 (2006) 16

Kuznetsova et al., 1996), molecular oxygen (Miyahara et al., 2001; Ohtani et al., 1995), or a mixture of oxygen and hydrogen (Niwa et al., 2002) have been reported. The direct hydroxylation of benzene with hydrogen peroxide to produce phenol was widely attempted as a green process (Clerici, 2001; Walling, 1975), which proceeds at the mild conditions. Vanadium is a most widely used element in catalysts, especially for the hydroxylation of alkanes and aromatic compounds with hydrogen peroxide (Gao and Hua, 2004). There were several reports for the benzene to phenol reaction on vanadium-containing catalysts (Zhang et al., 2005). However, the selectivity to phenol was less than 70% (Lemke et al., 2003; Chen and Lu, 1999) since phenol is more reactive than benzene in the reaction. Therefore we used clay-supported metal oxide catalysts for the hydroxylation of benzene and optimized the reaction conditions, as will be shown in this paper. Further, we report the clay-supported vanadium oxide catalyst exhibits the high selectivity in hydroxylation of benzene to phenol, which is higher or comparable to the known best catalyst, TS-1.
2. Experimental 2.1. Catalyst preparation Clay used in these experiments is composed of chlorite, illite, attapulgite, etc., and were produced in Inner Mongolia, northwest of China. The clay consists of Si and Al (ca. Si / Al = 4 in atomic ratio) as major components, together with Na, K, Fe, Mn, and Ti as minor components. The catalysts were prepared by the following procedures, 100 g of clay was added to a 300 mL of 0.25 M aqueous sulfuric acid solution, stirred for 24 h at room temperature, and filtered. The wet clay was carefully washed until no sulfate ion can be detected. Thus, all soluble basic ions were removed. The washed clay was dried at 393 K. Vanadium oxides were introduced to the dried clay by impregnation method from an aqueous ammonium metavanadate solution. The clay-supported vanadium oxide was dried at 393 K and calcined at 973 K for 5 h. The other metal oxides such as copper, iron, manganese, chromium, molybdenum, and tungsten, were supported on clay, as similarly as vanadium case, except that copper, iron, and tungsten/clay were calcined at 773 K for 5 h because the precursor of copper, iron, and tungsten could decompose under low temperature. The TS-1 was prepared using the similar procedure as has been reported (Taramaso et al., 1983). In brief, tetraethylorthosilicate (TEOS) was added slowly to a solution of titanium n-butoxide (TBOT) and 25% w / w aqueous solution of tetra-n-propylamminumn hydroxide (TPAOH). The final compositions of TEOS : TBOT : TPAOH : H 2 O were 1 : 0.25 : 0.25 : 68 in mole ratio. The above mixture was heated

at 348 K to remove alcohols. TS-1 was statically crystallized in a Teflon-lined stainless-steel autoclave at 443 K for 5 days. The crystalline solid was centrifuged, washed, dried, and calcined at 773 K for 5 h. 2.2. Hydroxylation reaction and products analysis The hydroxylation of benzene to phenol with hydrogen peroxide was performed in a 100 mL four-neck flask. Catalyst, benzene and acetic acid were added into the flask. The acetic acid addition was meant to enhance the activity of hydrogen peroxide, to improve the solubility of the reactants and to minimize the phase transfer problems. After heating the mixture to 313 K under stirring, hydrogen peroxide (30%) was added dropwise through a dropping funnel. The addition rate was controlled in 0.1 mL per min. After the complete addition of hydrogen peroxide, the solution was further stirred at 313 K for 4 h and then cooled down to room temperature. The GC analyses (Agilent 4890D) of products were performed with an OV-1 capillary column (35 m 0.32 mm 0.8 m). Peaks of GC chromatogram of the reaction products were assigned to phenol, 1,4-benzoquinone, catechol, biphenyl and hydroquinone from their retention times of reference samples. The conversion and the selectivity to phenol were obtained from following equations, taking into account of sensitivity factor of each compound. Conversion% phenol benzoquinone others 100 benzene phenol benzoquinone others Selectivity% phenol 100 phenol benzoquinone others

2.3. Characterization The XRD measurement was performed using miniflex (Rigaku) with a CuK radiation at 0.02 steps per second in angle range 2 = 5 90. The interaction between metal oxide and clay was estimated from the FT-IR spectra recorded by Tensor 27 (Bruker) at the resolution of 4 cm 1.

3. Results and discussion 3.1. Hydroxylation of benzene with hydrogen peroxide on various catalysts Table 1 shows the catalytic performances in the hydroxylation of benzene on clay-supported several metal oxide catalysts, together with clay without metal oxide. Clearly, the clay-supported vanadium oxide gave the best conversion and the best yield in these catalysts under the same reaction conditions. While some metal

X. Gao, J. Xu / Applied Clay Science 33 (2006) 16 Table 1 Hydroxylation of benzene on various clay-supported metal oxides Precursor of metal oxide Clay alone (NH4)6Mo7O244H2O Cr(NO3)39H2O Cu(NO3)3H2O Fe(NO3)9H2O Mn(CH3COO)24H2O H2WO4 NH4VO3 Metal in clay (wt.%) 0 10.0 10.0 10.0 10.0 10.0 10.0 10.0 Conversion (%) 3.3 0 1.3 3.4 3.7 4.8 5.0 8.3 Selectivity to phenol (%) 100 0 96 89 100 84 100 91

Reaction conditions: 0.056 mol of benzene and 0.098 mol of H2O2 (in 30% aqueous solution) in 30 mL of acetic acid, 313 K for 4 h, 1.0 g of catalyst weight.

Fig. 1. Effect of the amount of vanadium; the reaction conditions are same as those of Table 1.

oxide supported on the clay showed lower conversion of benzene, such as molybdenum and chromium, which may be poison for the hydroxylation of benzene to phenol with hydrogen peroxide. Table 2 represents the catalytic performances of Vbased catalysts in the oxidation of benzene to phenol, as compared with that on TS-1 catalysts as a reference. Vanadium oxide (entry 5) and its physical mixture with clay (entry 4) showed lower conversion and selectivity to phenol than the vanadium oxide/clay catalyst (entry 1). This shows that the vanadium oxides/clay is not equal to the physical mixture of vanadium and clay. A new matter may be formed in the vanadium oxide/clay catalyst owing to the interaction between vanadium and clay, which is responsible for higher conversion of benzene. Although V/MCM-41 (entries 23) showed somewhat higher conversions of benzene than vanadium oxide/clay catalyst, their selectivities to phenol were far lower. Under the similar reaction conditions, the TS1 catalyst gave the conversion of 5.3% and selectivity of

97% (entry 6) while the clay-supported vanadium oxide gave far higher conversion of 11.7%. Even reported result on TS-1 shows lower selectivity (82%) than the clay-supported vanadium oxide at similar conversion level (entry 8). We conclude from above results that the new matter is formed in clay-supported vanadium oxide due to the interaction between vanadium and clay, which promotes the hydroxylation of benzene with hydrogen peroxide at better or similar catalytic performances than TS-1, the best known catalyst for the reaction. 3.2. Optimization of reaction conditions Reaction conditions were optimized over the claysupported vanadium catalyst. Fig. 1 shows plots of vanadium metal wt.% in clay against the conversion and the selectivity to phenol in the hydroxylation of benzene. The conversion of benzene reached at a maximum value of 13.3% at 2.5 wt.% vanadium loaded

Table 2 Catalytic performances of benzene hydroxylation on vanadium-based catalysts Entry 1 2 3 4 5 6 7 8 Catalyst V/clay V/MCM-41 a V/MCM-41 b V2O5 + clay c V2O5 TS-1 TS-1 d TS-1 e Metal in support (wt.%) 5 Si / V = 25 1.1 5.6 56 Conversion (%) 11.7 13.0 10.8 7.1 5.9 5.3 31.0 11.6 Selectivity to phenol (%) 92 48 67 87 87 97 N95 82 Reference This work Chen and Lu (1999) Lemke et al. (2003) This work This work This work Tanev et al. (1994) Bhanmik et al. (1998)

The reaction conditions are the same as that of Table 1. a The reaction conditions:10 mmol benzene and 30 mmol H2O2 in 6 mL acetonitrile, at 333 K for 3 h. b The reaction conditions: 4 mmol benzene and 4 mmol H2O2 in acetic acid, at 295 K for 20 h. c The mixture of 0.10 g V2O5 and 0.90 g clay. d The reaction conditions: 10 mmol benzene and 29 mmol H2O2 in 0.13 mol acetone at 335 K for 2 h. e The reaction condition: benzene : acetonitrile : TS-1 = 1 : 5 : 0.15 (wt) and benzene : H2O2 = 1.0 (mol), at 333 K for 8 h.

X. Gao, J. Xu / Applied Clay Science 33 (2006) 16 Table 3 Effect of the ratio of addition of hydrogen peroxide on benzene Hydrogen peroxide/benzene (mol / mol) 2.2 1.8 1.4 1.0 0.6 Conversion (%) 13.5 13.3 14.1 11.2 5.7 Selectivity (%) 96.8 94.8 93.9 94.1 90.0

Reaction conditions: the different ratio of hydrogen peroxide on benzene in 30 mL of acetic acid, 313 K for 4 h, 0.25 g of catalyst weight, 2.5% of vanadium in clay.

Fig. 2. Effect of the reaction temperature. Reaction conditions: 0.056 mol of benzene and 0.098 mol of H2O2 (30%) in 30 mL of acetic acid, at different temperature for 4 h, 1.0 g of catalyst weight, 2.5% of vanadium in clay.

in clay and decreased with further increase of vanadium wt.%. As the vanadium wt.% increased beyond 2.5%, the excessive vanadium oxide decomposes more hydrogen peroxide to oxygen and water, which is responsible for the decrease of the conversion of reactant and the selectivity of product. The optimum vanadium content is about 2.5%. The reactions were conducted in the temperature range from 293 to 333 K, as shown in Fig. 2. As is clearly seen, the best yield of phenol was obtained at temperature of about 313 K. At higher temperature, the fall of the selectivity is owing to a possible decomposition of hydrogen peroxide that led to the by-product formation; at lower temperature the reason is that phenol has a more activity than benzene. Fig. 3 illustrated the influence of the amount of the catalyst from 0.05 to 1.00 g. It can be seen that the conversion of the substrate reached a plateau when the

amount of the catalyst is 0.25 g or larger. When the amount of the catalyst is over 0.25 g, the excessive catalyst decomposes more hydrogen peroxide to oxygen and water. The effect of the variation in the molar ratio between hydrogen peroxide and benzene is described in Table 3. The volume of benzene remained while that of H2O2 changed. The conversion of benzene increased as the increasing ratio of hydrogen peroxide on benzene and reached the best value at the ratio of hydrogen peroxide on benzene about 1.4. 3.3. Characterization of the support and catalysts The samples of catalyst were characterized by FT-IR to understand the interaction between vanadium and clay. Fig. 4 shows the IR spectra of the samples between 1400 and 900 cm 1. There are two bands in the spectra of the mechanical mixture of V2O5 and the clay (Fig. 4C). The band at 1154 cm 1 is attributed to the clay (Fig. 4B) and the band at 1024 cm 1 is assigned to

Fig. 3. Effect of the amount of the catalyst. Reaction conditions: 0.056 mol of benzene and 0.098 mol of H2O2 (30%) in 30 mL of acetic acid, 313 K for 4 h, 2.5% of vanadium in clay.

Fig. 4. IR spectra of sample A V/clay (2.5%), B clay, C the mixture of clay and V2O5, D V2O5, respectively.

X. Gao, J. Xu / Applied Clay Science 33 (2006) 16

Fig. 6. The XRD of support and catalyst. Fig. 5. IR spectra of sample A V/clay (2.5%), B clay, C the mixture of clay and V2O5, D V2O5.

V _ O stretching mode, which is very close to the band at 1026 cm 1 of V2O5 (Fig. 4D) (Concepcion et al., 2002). When vanadium was supported on the clay (Fig. 4A), the bands of clay at 1154 cm 1 and of V _ O at 1024 cm 1 shifted from 1127 to 1013 cm 1, respectively. In addition, both of them weakened and broadened. It is suggested that the changes in peak position of the V _ O stretching mode were attributed to the interaction between vanadium and clay. Fig. 5 displays the IR spectra of the samples at low wavenumber between 400 and 900 cm 1. There are two new bands at 591 and 424 cm 1 that appeared in the spectra of the clay-supported vanadium oxides (Fig. 5A). The band about 812 cm 1 (Fig. 5C and D) is assigned to the vibration of VO bond in the bridge of VOV (Scheme 1a) (Botto et al., 1997) whereas the band of the vibration of VO bond disappears in the IR spectra of the clay-supported vanadium oxides (Fig. 5A). Owing to the interaction between the vanadium and the clay, vanadium in the bridge of VOV could be replaced by other metal of the support to form the new bridges of VOM (Scheme 1b). The vibration of

the VO bond in the bridge VOM is not IR active (Wachs, 1996). For the similar reason, the band of the MO bond in the bridge VOM would shift and change under the influence of vanadium. The band of the AlO bond in the stretching model of AlO6 octahedra appears about 604 cm 1 (Morterro and Magnacca, 1996) (Fig. 5B and C), whereas the band of AlO bond in the bridge VOAl (Scheme 1c) shifts to 591 cm 1 and becomes sharper. The band of the SiO rocking vibrations appears about 472 cm 1 (Handke and Mozgawa, 1995) (Fig. 5B and C), whereas the band of the SiO bond in the bridge VOSi (Scheme 1d) shifts to 424 cm 1. It is believed that the new bridges of the VOAl and VOSi were formed in the claysupported catalysts. We envisioned the VOAl and VOSi bonds structure may be the active centre for the oxidation of benzene to phenol. The results of XRD of support and catalyst are shown in Fig. 6. There are no significant changes in XRD spectra between the original clay, and acid-treated and calcined clay, showing stability of clay to heat- and acidtreatments. XRD spectra of the clay-supported vanadium change a little from that of original clay, suggesting vanadium stayed at outside of layer. 4. Conclusions In conclusion, the catalyst of clay-supported vanadium oxides is effective for the direct hydroxylation of benzene to phenol and shows 14% conversion of benzene and 94% selectivity to phenol. The bridge of VOAl and VOSi may be formed in the catalysts and be the active centre for the reaction. The lower cost and less pollution of clay as support of catalyst may be positive and competitive for the reaction of hydroxylation in the future.

Scheme 1.

X. Gao, J. Xu / Applied Clay Science 33 (2006) 16 Lemke, K., Ehrich, H., Lohse, U., Beindt, H., Jahnisch, K., 2003. Selective hydroxylation of 12 benzene to phenol over supported vanadium oxide catalysts. Appl. Catal., A Gen. 243, 4151. Miyahara, T., Kanzaki, H., Hamada, R., Kuroiwa, S., Nishiyama, S., Tsuruya, S., 2001. Liquid-phase oxidation of benzene to phenol by CuOAl2O3 catalysts prepared by co-precipitation method. J. Mol. Catal., A Chem. 176, 141150. Morterro, C., Magnacca, G., 1996. A case study: surface chemistry and surface structure of catalytic aluminas, as studied by vibrational spectroscopy of adsorbed species. Catal. Today 27, 497532. Murray, H.H., 2000. Traditional and new applications for kaolin, smectite, and palygorskite: a general overview. Appl. Clay Sci. 17, 207221. Niwa, S., Eswaramoorthy, M., Nair, J., Raj, A., Itoh, N., Shoji, H., Namba, T., Mizukami, F., 2002. A one-step conversion of benzene to phenol with a palladium membrane. Science 295, 105107. Ohtani, T., Nishiyama, S., Tsuruya, S., Masai, M., 1995. Liquid-phase benzene oxidation to phenol with molecular oxygen catalyzed by Cuzeolties. J. Catal. 155, 158162. Panov, G.I., Sheveleva, G.A., Kharitonov, A.S., Romannikov, V.N., 1992. Oxidation of benzene to phenol by nitrous oxide over Fe ZSM-5 zeolites. Appl. Catal., A Gen. 82, 3136. Pinnavaia, T.J., Welty, P.K., 1975. Catalytic hydrogenation of 1hexene by rhodium 13 complexes in intracrystal space of a swelling layer lattice silicate. J. Am. Chem. Soc. 97, 38193820. Shimazu, S., Ro, K., Sento, T., Ichikuni, N., Uematsu, T., 1996. Asymmetric hydrogenation of , -unsaturated carboxylic acid esters by rhodium(I)-phosphine complexes supported on smectites. J. Mol. Catal., A Chem. 107, 297303. Shimizu, K.I., Kaneko, T., Fujishima, T., Kodama, T., Yoshida, H., Kitayama, Y., 2002. Selective oxidation of liquid hydrocarbons over photoirradiated TiO2 pillared clays. Appl. Catal., A Gen. 225, 185195. Sobolev, V.I.G., Panov, I., Khatitonov, A.S., Romannikov, V.N., Volodin, V.M., Ione, K.G., 1993. Catalytic properties of ZSM-5 zeolites in N2O decomposition: the role of iron. J. Catal. 139, 435443. Tanev, P.T., Chibwe, M., Pinnavaia, T.J., 1994. Titanium-containing mesoporous molecular sieves for catalytic oxidation of aromatic compounds. Nature 368, 321323. Taramaso, M., Perego, G., Notri, B., 1983. Preparation of Porous Crystalling Synthetic Material Comprised of Silicon and Titanium Oxides. US 4 410 501. Walling, G., 1975. Fenton's reagent revisited. Acc. Chem. Res. 8, 125131. Wachs, I.E., 1996. Raman and IR studies of surface metal oxide species on oxide supports: supported metal oxide catalysts. Catal. Today 27, 437455. Zhang, W.Z., Wang, J.L., Tanev, P.T., Pinnavaia, T.J., 1996. Catalytic hydroxylation of benzene over transition-metal substituted hexagonal mesoporous silicas. Chem. Commun. 979980. Zhang, J., Tang, Y., Li, G.Y., Hu, C.W., 2005. Room temperature direct oxidation of benzene 14 to phenol using hydrogen peroxide in the presence of vanadium-substituted heteropolymolybdates. Appl. Catal., A Gen. 278, 251261.

Acknowledgements The work was supported by the National Natural Science Foundation of China (20233040) and the National Hi-Tech Research and Development Program of China (2002AA321020). References
Bartok, M., Szollosi, G., Mastalir, A., Dekany, I., 1999. Hydrogenation reactions on heterogenized Wilkinson complexes. J. Mol. Catal. A Chem. 139, 227234. Bengoa, J.F., Gallegos, N.G., Marchetti, S.G., Alvarez, A.M., Cagnoli, M.V., Yeramian, A.A., 1998. Influence of TS-1 structural properties and operation conditions on benzene catalytic oxidation with H2O2. Microporous Mesoporous Mater. 24, 163172. Bhanmik, A., Mukherjee, P., Kumar, P., 1998. Triphase catalysis over titaniumsilicate molecular sieves under solvent-free conditions. 1 Direct hydroxylation of benzene. J. Catal. 178, 101107. Bianchi, D., Bortolo, R., Tassinari, R., Ricci, M., Vignola, R., 2000. A novel iron-based catalyst for biphasic oxidation of benzene with hydrogen peroxide. Angew. Chem., Int. Ed. 39, 43214323. Botto, I.L., Vassallo, M.B., Baran, E.J., Minelli, G., 1997. IR spectra of VO2 and V2O3. Mater. Chem. Phys. 50, 267270. Campanati, M., Vaccai, A., 2001. Acidic clays. In: Sheldon, R.A., Bekkum, H.V. (Eds.), Fine Chemicals through Heterogeneous Catalysis. Wiley-VCH, Weinheim, pp. 66120. Federal Republic of 11 Germany. Chen, Y.W., Lu, Y.H., 1999. Characteristics of V-MCM-41 and its catalytic properties in oxidation of benzene. Ind. Eng. Chem. Res. 38, 18931903. Clerici, M.G., 2001. Aromatic ring hydroxylation. In: Sheldon, R.A., Bekkum, H.V. (Eds.), Fine Chemicals through Heterogeneous Catalysis. Wiley-VCH, Weinheim, pp. 538549. Federal Republic of Germany. Concepcion, P., Kno1zinger, H., Lopez Nieto, J.M., Martinez-Arias, A., 2002. Characterization of supported vanadium oxide catalysts. Nature of the vanadium species in reduced catalysts. J. Phys. Chem., B 106, 25742582. Gao, F.X., Hua, R.M., 2004. Highly efficient K7NiV13O38-catalyzed hydroxylation of aromatics with aqueous hydrogen peroxide (30%). Appl. Catal., A:Gen. 270, 223226. Handke, M., Mozgawa, W., 1995. Model quasi-molecule Si2O as an approach in IR spectra description glassy and crystalline framework silicates. J. Mol. Struct. 348, 341344. Kun, I., Szollosi, G., Bartok, M., 2001. Crotonaldehyde hydrogenation over clay-supported platinum catalysts. J. Mol. Catal., A Chem. 169, 235246. Kuznetsova, L.I., Detusheva, L.G., Fedotov, M.A., Likholobov, V.A., 1996. Catalytic properties of heteropoly complexes containing Fe (III) ions in benzene oxidation by hydrogen peroxide. J. Mol. Catal., A Chem. 111, 8190.