The benzylation of benzene using aluminium, gallium and iron incorporated silica from rice husk ash

The benzylation of benzene using aluminium, gallium and iron incorporated silica from rice husk ash

Microporous and Mesoporous Materials 118 (2009) 35–43 Contents lists available at ScienceDirect Microporous and Mesoporous Materials journal homepag...

488KB Sizes 0 Downloads 21 Views

Microporous and Mesoporous Materials 118 (2009) 35–43

Contents lists available at ScienceDirect

Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso

The benzylation of benzene using aluminium, gallium and iron incorporated silica from rice husk ash Adil Elhag Ahmed, Farook Adam * School of Chemical Sciences, Universiti Sains Malaysia, 11800 Penang, Malaysia

a r t i c l e

i n f o

Article history: Received 9 June 2007 Received in revised form 16 July 2008 Accepted 6 August 2008 Available online 20 August 2008 Keywords: Benzylation Benzene Gallium–silica catalyst Iron–silica catalyst Heterogeneous catalyst

a b s t r a c t Al, Ga and Fe catalysts supported on silica obtained from rice husk ash (RHA) were synthesized using the sol–gel technique at room temperature and were denoted as RHA–Al, RHA–Ga and RHA–Fe, respectively. The XRD and SEM results revealed that the amorphous and porous structure of RHA silica was retained after the incorporation of Al3+, Ga3+ and Fe3+ ions. EDX analysis confirmed Al3+, Ga3+ and Fe3+ ions were attached to the silica matrix and were homogeneously distributed. 27Al MAS NMR results of RHA–Al revealed the Al3+ ions were incorporated into the silica network with tetrahedral coordination. The benzylation of benzene (Bz) and substituted benzenes with benzyl chloride (BC) were studied over the prepared catalysts. RHA–Fe showed excellent activity for the benzylation of benzene, whereas, RHA–Ga gave good selectivity towards diphenylmethane (DPM). Almost complete BC conversion and about 86% and 80% selectivity to DPM were obtained after 90 and 13 min over RHA–Ga and RHA–Fe, respectively. However, RHA–Al was found to be inactive under the reaction conditions studied. Catalytic activity of RHA–Ga with substituted benzenes showed a reverse trend to that of the conventional Friedel–Crafts reaction, i.e. benzene > toluene > ethyl benzene > anisole. The catalysts could be reused for the benzylation reaction several times without significant loss in their activity and selectivity. Kinetic studies showed that the activation energy for the benzylation of benzene was 26.4 and 17.6 kcal mol1 over RHA–Ga and RHA– Fe. The frequency factor for RHA–Ga and RHA–Fe was found to be 5.3  1014 and 1.7  1010 min1, respectively. Ó 2008 Elsevier Inc. All rights reserved.

1. Introduction South and South East Asia account for over 90% of the world’s rice production [1]. The rice husk has no commercial value. However, many researchers have reported that rice husk is an excellent source of high-grade amorphous silica [2–5]. Real et al. [6] concluded that a homogeneous size distribution of nanometric silica particles could be obtained by burning rice husk at 600–800 °C in a pure oxygen atmosphere. Della et al. [7] found that active silica with a high specific surface area could be produced from rice husk ash (RHA) after heat-treatment at 700 °C in air. Kalapathy et al. [8] investigated an improved method to produce silica with lower sodium content by adding alkaline silicate solution to acidic solution of different acids at pH 1.5 until pH 4.0. Kamath and Proctor [9], however, prepared silica by first dissolving the RHA in sodium hydroxide solution, and then titrated with acid to obtain silica gel. These previous studies focused on the process of producing silica from rice husk. However, much less

* Corresponding author. Tel.: +60 4 6533888; fax: +60 4 6574854. E-mail addresses: [email protected], [email protected], [email protected] (F. Adam). 1387-1811/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.micromeso.2008.08.024

attention has been given to heterogeneous catalysts based on silica from rice husk – particularly for alkylation reactions. Friedel–Crafts type alkylation of aromatic compounds using homogeneous Brønsted and Lewis acid catalysts are well-known in organic synthesis as an important means for attaching alkyl chains to aromatic rings [10]. Recently, considerable efforts have been directed towards the heterogenization of Lewis acids (for the alkylation of benzene and substituted benzenes) onto different supports. These supports include mesoporous molecular sieves [11–15], zeolites e.g. ZSM-5 [16], Hb [17], clays [18,19], alumina [20] and poly acid salts [21]. Few papers have reported the use of Ga containing catalysts, but many reports of aluminum and iron supported catalysts had been published. Gallium catalysts were found very active although not comparable to those of iron supported catalysts, but their high selectivity towards diphenylmethane (DPM) and little or no moisture sensitivity gave an extra advantage [14,17,22]. We had reported the catalytic activity of iron incorporated silica from RHA, i.e. RHA–Fe, for the Friedel–Crafts benzylation of toluene [23]. The selectivity for mono benzyl toluene was further improved by incorporation of amino benzoic acid molecules into the silica– iron catalyst [24]. In a recent investigation we reported the preparation of indium-modified silica from RHA, i.e. RHA–In [25]. This

A.E. Ahmed, F. Adam / Microporous and Mesoporous Materials 118 (2009) 35–43

catalyst showed very good catalytic activity and product selectivity for the benzylation of benzene. This work is a continued systematic study of Friedel–Crafts reactions by RHA modified catalyst for the benzylation reactions of benzene and substituted benzenes. Herein we report the incorporation of aluminum, gallium and iron into the RHA matrix by using a simple sol–gel technique to form the RHA–Al, RHA–Ga and RHA–Fe catalyst. The synthesized catalysts were tested for the benzylation reactions and the results are discussed. 2. Experimental 2.1. Raw material The following chemicals were used without further purification: aluminum nitrate nonahydrate (Riedel–De Haen AG, 99%), gallium nitrate monohydrate (Alfa Aesar, 99.99%), iron granules (Aldrich, 99.999%), sodium hydroxide (Systerm, 99%), nitric acid (Systerm, 65%), benzene (Merck, 99.7%), benzyl chloride (Fluka, 99%), toluene (J.T. Baker, 99.8%), ethyl benzene (Fluka, P98%) and anisole (Fluka, 99%). Rice husk was obtained from a rice mill in Penang, Malaysia. 2.2. Methods 2.2.1. Extraction and modification of silica from RHA The catalysts were prepared following the pathway previously reported in [25]. In a typical synthesis 5.0 g of the treated RHA sample was dissolved in 250 mL 1.0 M NaOH in a plastic container and then filtered. The filtrate was titrated with 3.0 M HNO3 solution (or containing 0.25 g of Al3+, Ga3+ or Fe3+ metal ions) at a slow rate of 1.0 mL min1 with constant stirring to pH 5. The gel formed was aged for 24 h then filtered, washed thoroughly with distilled water and dried at 100 °C for 18 h. The silica or metal modified silica xerogel was ground to powder and washed again with distilled water several times then filtered and dried at 100 °C for 18 h. This extra washing enabled the nitrate ions present to be washed off completely. The materials obtained were labeled as RHA–SiO2, RHA–Al, RHA–Ga and RHA–Fe. 2.2.2. Samples characterization The prepared samples were characterized by FT-IR spectroscopy (Perkin Elmer System 2000), N2 sorption analysis (Micromeritics Instrument Corporation Model ASAP 2000, Norcross), powder X-ray Diffractometry (Siemens Diffractometer D5000, Kristalloflex), Scanning Electron Microscopy (SEM) (Leica Cambridge S360), Energy Dispersive Spectrometry (EDX) (EDX Falcon System), Inductively Coupled Plasma Mass Spectroscopy (ICP-MS) (PE SCIEX ELAN 6100) and 27Al MAS NMR Solid-state spectroscopy (400 MHz Bruker AVANCE solid-state NMR). Gas Chromatograms of the benzylation products were obtained using a Shimadzu GC-17A with a VB-1 non polar capillary column, 30 m length and 0.32 mm inner diameter. The chromatograph was equipped with a recorder model C-R8A Chromatopac Shimadzu and FID detector. Product identification was achieved using a GC–MS model Trace GC 2000, Thermo Finnigan coupled to a Trace MS detector (Thermo Finnigan). 2.2.3. Catalytic benzylation reaction Liquid phase Friedel–Crafts benzylation reactions of benzene and substituted benzenes over the prepared catalysts were carried out in a magnetically stirred round-bottom flask (600 rpm) equipped with a reflux condenser. The temperature of the reaction vessel was maintained precisely using an oil bath. The reactions were carried out under a continuous flow of dry argon

following the same experimental procedures reported earlier [25]. 3. Results and discussion 3.1. Catalysts characterization Fig. 1 shows the X-ray diffractograms of RHA–SiO2 support as well as RHA–Al, RHA–Ga and RHA–Fe catalysts. The absence of sharp well defined peaks indicate the absence of any ordered crystalline structure. It also indicates the high dispersion of the metal ions in the silica matrix or surface. All the catalysts and RHA silica (support) had similar diffraction patterns: a broad peak centered at 2h angle of around 22° conforming to the amorphous nature of the samples. Nitrogen adsorption–desorption isotherms of the silica and metal modified silica samples exhibited almost the same shapes as observed before [25], indicating the presence of porous materials of almost similar textural properties, i.e. type (IV) isotherms according to BET classification and type H1 hysteresis loops according to IUPAC classification (Fig. 2a–c). The isotherms of the porous solids showed slightly sloping H1 hysteresis loops indicating a wide range of pores rather than a narrow pore size distribution (Fig. 2 (insets)) as stated in the IUPAC report [26]. Similar observation was also made by El Shafei et al. [27]. The adsorption–desorption isotherm of RHA–Al displayed a slightly distorted hysteresis loop, indicating the presence of pores of quite different size as shown in Table 1. This might be due to the formation of Al tetrahedral structure within the silica network as indicated in the solid state 27Al NMR in Fig. 3. However, the average pore diameter was found to be ca. 5.2 nm which was smaller than the pores in RHA–Ga and RHA–Fe. Similar changes in textural properties of supported aluminum and iron compared with the support was also reported by Szegedi et al. [28] for aluminum and iron supported onto mesoporous MCM-41. Fig. 4 shows the FT-IR spectra of RHA–SiO2, RHA–Al, RHA–Ga and RHA–Fe. The absorption bands at around 470, 800 and 1100 cm1 are attributed to the consequence of stretching and bending vibrations of SiO4 tetrahedra [29,30]. The peak at 972 cm1 is attributed to silanol Si–OH stretching vibration. The intensity of this peak decreases in RHA–Ga and RHA–Fe samples, whereas, this peak is diminished in the case of RHA–Al. This might indicate that these metal ions were attached on the surface of RHA silica replacing the silanol hydrogen. However, the possibility of the formation of trigonal environment around Al on the silica matrix is expected since 27Al NMR results in Fig. 3 confirmed the absence of octagonal aluminum environment. Moreover, the shift in

200

a b Ln (Cps)

36

100

c d

0 10

20

30

40

50

60

70

2-Theta - scale Fig. 1. The X-ray diffraction patterns of (a) RHA–SiO2, (b) RHA–Ga, (c) RHA–Al and (d) RHA–Fe.

37

A.E. Ahmed, F. Adam / Microporous and Mesoporous Materials 118 (2009) 35–43

a

500

0.6

dV/dLogD

Volume (cm3 g-1 at STP)

0.5

400

300

a

0.4 0.3 0.2 0.1 0 0

200

400

600

800 1000 1200

Pore diameter A

200

100 Adsorption Desorption

b

0 0

b

0.2

0.4

P/P0

0.6

0.8

1

600 1.2

80

dV/dLogD

Volume (cm3 g-1 at STP)

500

400

20

0

-20

Fig. 3. The 27Al MAS NMR spectra of RHA–Al: (a) as synthesized and (b) calcined at 500 °C for 5 h, indicating the presence of Al tetrahedral species only.

0.6 0.4 0.2 0 0

300

200

400

600

800

1000

1200

Pore Diameter A

200

Adsorption Desorption

0 0

0.2

0.4

0.6

0.8

1

P/P 0 800

2 1.8

700

1.6 1.4

600

dV/dlogD

Volume (cm3 g-1)

40

0.8

100

c

60

ppm

1

500

1.2 1 0.8 0.6

400

0.4 0.2

300

0 -50

100

250

400

550

700

850

1000

pore diameter A

200 100

Adsorption Desorption

0 0

0.2

0.4

0.6

0.8

1

P/P0 Fig. 2. The nitrogen adsorption–desorption isotherms of (a) RHA–Al, (b) RHA–Ga and (c) RHA–Fe. Insets are the pore size distribution graphs of the respective catalysts.

Table 1 The surface analysis parameters and metallic contents for RHA–SiO2, RHA–Al, RHA– Ga and RHA–Fe Sample

RHA–SiO2 RHA–Al RHA–Ga RHA–Fe

BET surface area S (m2 g1)

BJH average pore volume (cc g1)

BET average pore diameter D (nm)

Average content of metal ions (w/w)% EDX

ICP–MS

347.3 384.3 332.4 284.0

0.872 0.452 0.824 1.080

10.4 5.2 10.4 15.2

– 3.9 (Al) 2.0 (Ga) 3.8 (Fe)

– 5.1 (Al) 2.5 (Ga) 3.7 (Fe)

the band around 1100 which is assigned to asymmetric vibration of Si–O–Si towards lower values of wave numbers for all the modified silica samples might indicate that some metal ions were incorporated into the silica tetrahedral framework. The shifting is significant in the case of RHA–Al. This might indicate that most of the Al3+ ions were occupying the framework positions. This result is quite consistent with the conclusion obtained from 27Al NMR for RHA–Al. The 27Al MAS NMR results for RHA–Al as synthesized and RHA– Al(C) calcined at 500 °C for 5 h, Fig. 3b, showed only one peak at about 50 ppm indicating Al3+ ions have been bonded into the network silica forming a tetrahedral framework. The absence of a chemical shift around 0 ppm showed that no octahedral Al was available and the aluminum ions were stabilized within the tetrahedral environment even after it had been calcined at 500 °C. However, the presence of other coordination environment symmetries of Al that was difficult to be detected may be possible i.e. the formation of a severely distorted tetrahedral site or a trigonal structure [31,32]. SEM images of RHA–SiO2, RHA–Al, RHA–Ga and RHA–Fe are given in Fig. 5. Spherical particles are evident in RHA–SiO2, RHA–Ga and RHA–Fe. These spherical particles are between 100 and 200 nm diameters. However, RHA–Al exhibits a very porous surface. The presence of the nano-sized spherical particles leads to the highly porous structure of this catalyst. This results in a higher specific surface area as shown in Table 1. The elemental analysis in Table 1 for the metal modified silica samples showed the presence of Al3+, Ga3+ and Fe3+ on the silica surface. This together with the shifting of the Si–O–Si vibration frequency at ca. 1100 cm1 in Fig. 4 suggests the formation of M–O–Si (M = Al, Ga or Fe) bonds as a result of metal ion incorporation during the synthesis. A comparison between EDX (surface content) and ICP-MS (total content) elemental analysis data for each catalyst could give a logical explanation on how the metallic species were incorporated in the silica support. Unlike the case of RHA– Fe, the amounts of Al3+ and Ga3+ species detected by EDX on the surface of RHA–Al and RHA–Ga, respectively, were lower than that determined by ICP-MS. This could be due to the presence of some Al3+ and Ga3+ ions at the framework positions (Fig. 3). However, the extremely lower concentration of Ga3+ in the catalyst (2.5%) compared to that introduced in the preparation (5%) could be due to the high tension generated in the silicate structure between Si4+ and Ga 3+ ions at this concentration level of Ga3+ ions. This tension,

38

A.E. Ahmed, F. Adam / Microporous and Mesoporous Materials 118 (2009) 35–43 70.4

a

a

65

b

b

60 55

801.09 1636.99

c

50

972.59

d

45

c

%T

40

d

3457.70

35

793.79 469.57

30

1638.05

25 20 1099.59

15 3464.32

10

462.99 1077.79 1094.79

5 1.6 4000.0

3600

3200

2800

2400

2000

1800

1600

1400

1200

1000

cm-1 Fig. 4. The FT-IR spectra of (a) RHA–SiO2, (b) RHA–Fe, (c) RHA–Al and (d) RHA–Ga.

Fig. 5. The SEM micrographs of (a) RHA–SiO2, (b) RHA–Al, (c) RHA–Ga and (d) RHA–Fe.

800

600

400.0

39

A.E. Ahmed, F. Adam / Microporous and Mesoporous Materials 118 (2009) 35–43

3.2. Catalytic reactions The Friedel–Crafts benzylation of benzene (Bz) with benzyl chloride (BC) was carried out using different experimental conditions, namely varying molar ratios of reactants, reaction temperature, reaction time, catalyst type, different benzene substituents, catalyst weight, recyclability, and leaching effect. It is important to highlight the fact that for all the reactions the only products formed were the mono benzylated product (main product) along with the dibenzylated products (side products) of the respective aromatic compounds.

BC conversion %

3.2.1. Influence of molar ratio of the reactants The benzylation of Bz was carried out at 80 °C with various Bz:BC molar ratios using 0.1 g of RHA–Ga catalyst. Fig. 6 shows the catalytic activity of RHA–Ga increases with decreasing Bz/BC molar ratio. This could be explained by the fact that more carbocations were formed due to high concentration of BC being in contact with Ga3+ on the catalyst surface. Fig. 6 also shows the percent selectivity for DPM (i.e. the mono-substituted product) with respect to reaction time. In all cases the selectivity of DPM decreased as the reaction time increased. The intersection points of the respective graphs give the minimum time required for a particular reaction to yield the maximum amount of DPM for a given molar ratio of the reactants. Based on this, it can be concluded that the maximum yield of DPM was obtained at about 72 min for the 20:1 molar ratio, and this yield was determined to be ca. 90%. This was followed closely by the 15:1 molar ratio mixture which gave a DPM yield of ca. 88% in about 66 min. The optimum yield of DPM for the 10:1 molar ratio was only ca. 80% which was achieved in about 50 min. However, the 20:1 and 15:1 mixtures reached selectivity of 88% and 86%, respectively, at the time of complete conversion.

3.2.2. Influence of reaction temperature The benzylation of Bz was also carried out at different reaction temperatures using RHA–Ga catalyst and Bz/ BC molar ratio of 15:1. The catalytic performance of the catalyst changed with temperature. It is clear from Fig. 7 that the conversion and the selectivity towards DPM can be easily controlled by simply adjusting the reaction temperature. The conversion of BC increased by increasing the reaction temperature while the percentage selectivity to DPM decreased. No reaction was observed without heating or when the catalyst was omitted from the reaction mixture. The optimum temperature was found to be 80 °C which gave 88% yield of DPM in about 66 min. However, at 75 °C, the reaction only gave ca. 88% yield of DPM but in a longer time of 88 min. 3.2.3. The effect of catalyst mass Table 2 shows the conversion of BC increased when the mass of the catalyst was increased. There was a marked increase in conversion when the mass of the catalyst was increased from 0.05 to 0.10 g. Further increase of the mass of the catalyst to 0.15 g did not significantly change its overall catalytic performance, although it was registered in a shorter time of 60 min. However, the values of turn over frequency (TOF) for the reactions of different catalyst mass clearly illustrated that the most efficient conversion occurred with 0.05 g of the catalyst. However, by doubling the catalyst mass two or three folds, the TOF were not increased, but it rather decreased and almost stabilized at the value of 312 h1. This indicates there might be an influence of interphase mass transfer effect. It was reported by Madon and Boudar [33] that the reaction rate should be doubled by doubling the catalyst weight if the interphase mass transfer effect is absent. Table 2 also shows the catalyst mass of 0.1 g gave the highest selectivity for DPM but when using a mass of 0.15 g, there was an observable decrease in the selectivity. Since relatively better catalytic performance was obtained by introducing a catalyst mass of 0.1 g, this weight was considered as an optimum weight for studying the other parameters.

100

100

90

90

80

80

70

70

60

60

50

50

40

40

30

30 Conversion, molar ratio 10:1 Conversion, molar ratio 15:1 Conversion, molar ratio 20:1 Selectivity, molar ratio 10:1 Selectivity, molar ratio 15:1 Selectivity, molar ratio 20:1

20

10

0 0

10

20

30

40

50

60

70

80

90

100

Selectivity to DPM %

as reported by Szegedi et al. [28], may limit the concentration of metal ions taken up within the silica structure.

20

10

0 110

time - min Fig. 6. Percentage conversion and selectivity profile versus time for the reaction between Bz and BC at 80 °C using different molar ratios of Bz/BC and RHA–Ga as a catalyst.

A.E. Ahmed, F. Adam / Microporous and Mesoporous Materials 118 (2009) 35–43

100

100

90

90

80

80

70

70

60

60

50

50

40

40 Conversion at 85 C Conversion at 80 C Conversion at 75 C Selectivity at 85 C Selectivity at 80 C Selectivity at 75 C

30 20 10

30 20 10

0 0

10

20

30

40

50

60

70

80

90

100

110

Selectivity to DPM %

BC conversion %

40

120

130

0 140

time - min Fig. 7. Percentage conversion and selectivity profile versus time for the reaction between Bz and BC at different temperatures using Bz/BC molar ratio of 15:1 and RHA–Ga as a catalyst.

Table 2 The product distribution with varying RHA–Ga catalyst mass for the benzylation of benzene at 80 °C using Bz/BC molar ratio of 15:1 Catalyst mass (g)

Time (min)

BC conversion (%)

TOF (h1)a

DPM (%)

Di-substituted products (%) 1,2-DBB

1,3-DBB

1,4-DBB

0.05 0.10 0.15

90 90 60

84.2 97.8 98.6

538 312 315

84.9 86.0 83.3

5.4 5.1 5.8

2.3 1.8 2.4

7.4 7.1 8.5

The turn over frequency describes the number of mmol of BC converted by mmol of supported gallium per hour.

BC conversion %

3.2.4. Influence of catalyst type The effect of catalyst type was investigated at the optimum conditions (i.e. at 80 °C, 90 min and Bz/ BC molar ratio of 15:1 and cat-

alyst mass of 0.1 g) using either RHA–Al, RHA–Ga or RHA–Fe as a catalyst. Fig. 8 presents the catalytic behavior of all the catalysts studied.

100

100

90

90

80

80

70

70

60

60

50

50 40

40

Conversion, RHA-Al Conversion, RHA-Ga

30

Selectivity to DPM %

a

30

Conversion, RHA-Fe Selectivity, RHA-Ga

20

20

Selectivity, RHA-Fe 10

10 0 0

10

20

30

40

50

60

70

80

90

0 100

time - min Fig. 8. Percentage conversion and selectivity profile versus time for the reaction between Bz and BC at 80 °C over the catalytic activity of different catalysts using Bz/BC ratio of 15:1. Note that there was no reaction with RHA–Al.

A.E. Ahmed, F. Adam / Microporous and Mesoporous Materials 118 (2009) 35–43

It is interesting to note that, no reaction was seen over RHA–Al in the benzylation reaction of benzene at this condition. No reaction took place with RHA–SiO2 (not shown). Obviously, the activity of RHA–Ga in this reaction was high, since 98% conversion of BC was observed within 90 min. Much higher activity was observed with RHA–Fe since the same percentage conversion was achieved within 12.5 min. The optimum yield of DPM with RHA–Ga was found to be ca. 87% achieved in about 65 min. However, the optimum yield of DPM with RHA–Fe was ca. 85% which was achieved within a shorter time of 10 min. While no products were observed over RHA–Al catalyst under optimum reaction conditions, however, after 5 h at 80 °C, an 8% conversion of BC was detected. It can therefore be concluded that the catalytic activity of RHA–Al was negligible compared to that of RHA–Ga and RHA–Fe. The catalytic activity of RHA–Ga and RHA–Fe were attributed to the presence of Ga3+ and Fe3+ ions, respectively, on the surface of the catalysts. These ions were expected to be much more available to the reactant molecules than in the case of Al3+ ions in RHA–Al catalyst. The Al3+ ions were probably embedded within the tetrahedral structure of the silica framework (replacing Si4+ ion) as indicated by the 27Al solid state NMR shown in Fig. 3. Moreover, the redox properties of Ga3+ (0.44 V) and Fe3+ (+0.77 V) metal ions species which are higher than that of Al3+ (1.662 V) might play an important role in initiating the BC carbocation for the benzylation reaction as suggested by Choudhary et al. [17,22]. Poor activity of directly incorporated Al3+ ions was observed by Vinu et al. [12]. Only 20% conversion was registered over Al–MCM41 after 10 h for the same reaction at 80 °C. No reaction over anhydrous AlCl3 impregnated Si-MCM-41 was observed by Choudhary et al. [34] at the same reaction conditions. However, they reported good activity [35] over anhydrous AlCl3 grafted to the same support (Si-MCM-41) and complete conversion was observed within 15 min. It was however, found that the activity of this catalyst became completely inactive when exposed to the atmosphere for more than 3 min. Only moderate activity was obtained over Al supported on SBA-15 at the same reaction condition and about 55% conversion was obtained after 3 h [12]. The increased activity of Table 3 The kinetic parameters for the benzylation of benzene over RHA–Ga and RHA–Fe catalysts using Bz/BC molar ratio of 15:1

RHA–Ga RHA–Fe

ka (min1) 348 K

353 K

358 K

14.4  103 158  103

24.5  103 206  103

41.8  103 321  103

Ea (kcal mol1)

A (min1)

26.4 17.6

5.3  1014 1.7  1010

this catalyst, however, was attributed to its high acidity and high pore diameter. It can thus be concluded that the activity of the metal modified silica from RHA catalyst for the benzylation of benzene with BC depends more on the redox property of the supported metal rather than its acidity. It is very clear that the order of the activity of these catalysts follow the redox properties of the loaded metal; i.e.

RHA—Fe > RHA—Ga  RHA—Al 3.2.5. Reaction kinetics It was observed that as the reaction time was increased the selectivity to DPM decreased, whereas, the selectivity to dibenzyl benzene (DBB) isomers (side products) increased. The selectivity to DBB follows the trend, 1,4-DBB > 1,2-DBB > 1,3-DBB. A similar trend was observed when the same reaction was studied over RHA–In [25] and also reported by Vinu et al. [12]. The data obtained from this study was used to determine the reaction kinetics for the catalytic reaction of RHA–Ga and RHA– Fe. The experimental data was found to fit well to the pseudo-first order [32] rate law shown:

log

  1 ka ¼ ðt  t0 Þ; 1x 2:303

0.00279

0.0028

0.00281

0.00282

3.2.6. Influence of electron donating substituent The effect of electron donating substituents was investigated at the optimum reaction conditions using RHA–Ga as the catalyst.

0.00283 0.00284

0.00285 0.00286

0.00287 0.00288

-1 -1.5

2

R = 0.9863

RHA-Fe

-2 -2.5 -3 -3.5

2

R = 0.9999

ð1Þ

where ka is apparent rate constant, x is fractional conversion of benzyl chloride, t is reaction time and t0 is the induction period. From Table 3, it is very clear that the apparent rate constant for the catalytic benzylation reaction increased ca. 7 to 11-folds when the catalyst was changed from RHA–Ga to RHA–Fe. The calculated first order rate constants gave linear Arrhenius plots for RHA–Ga and RHA–Fe as shown in Fig. 9. From Table 3, the activation energy, (Ea), for RHA–Ga was 26.4 kcal mol1while the activation energy for RHA–Fe was found to be 17.6 kcal mol1. The low activation energy for RHA–Fe is showing that it is a more efficient catalyst than RHA–Ga. It should be noted that the frequency factor for RHA–Fe was much smaller than RHA–Ga. This correlates well with the fact that the specific surface area of RHA–Ga was much larger than RHA–Fe (Table 1). Interestingly, the activity of RHA–Fe for this reaction was also superior to that of RHA–In catalyst previously reported [25], possibly due to the higher reduction potential of RHA–Fe. However, RHA–Ga exhibited lower activity when compared to RHA–In.

1/T (K—1) 0 0.00278 -0.5

ln ka

Catalyst

41

RHA-Ga

-4 -4.5

Fig. 9. The Arrhenius plots for the benzylation of benzene over RHA–Ga and RHA–Fe catalysts.

42

A.E. Ahmed, F. Adam / Microporous and Mesoporous Materials 118 (2009) 35–43

The results are shown graphically in Fig. 10. For the aromatic substrates studied, the BC conversion decreased in the order shown in Scheme 1. The rate of benzylation was retarded due to the presence of the electron donating groups such as methyl, ethyl and methoxy in the aromatic substrate. Although this observation was found to be opposite to the classical mechanism of the Friedel–Crafts type acid catalyzed benzylation reaction, but the result of this study was consistent with the previously reported data over RHA–In [25] and also with those observed by other researchers such as Vinu et al. [12] and Choudhary et al. [17]. To explain this reversal behavior we proposed the hypothesis that the carbocation species was adsorbed on the catalyst surface [25]. A similar mechanism can be said to be operative here as well. The poor reactivity of RHA–Ga for the benzylation of anisole at the optimal reaction conditions used in this study could be due to the blocking of the catalyst active sites by anisole as explained in [25]. This might be achieved either by adsorption of anisole on the catalyst surface or by coordinating directly to the gallium ions. Similar conclusions were also made by Barlow et al. [36] and He et al. [37]. With respect to toluene and ethyl benzene (EB), the percentage conversion dropped drastically compared to benzene. Toluene gave higher conversion compared to EB, possibly due to slightly strong adsorption of EB within the pores or coordinating with Ga3+ ions at the catalyst surface. It could also be due to the increased bulk of EB compared to toluene. However, the selectivity to mono benzylated products was observed to follow the order: anisole = ethyl benzene (100%) > toluene (98%) > benzene (86%). This could be due to the steric bulk of the mono-substituted products from anisole and ethyl benzene as well as toluene being too large to approach the carbocation that was immobilized within the mesopores of the catalyst surface for further substitution to take place.

100

97.8

90

100

97.5

100

86

Percentage (%)

70 60 43

40

34.8

30 20 10

3.7

0 Benzene

Toluene

Ethylbenzene

Anisole

Substrate BC conversion (%)

Mono sub. Product (%)

Fig. 10. The effect of electron donating substituents for the benzylation of aromatics (Ar) over RHA–Ga at 80 °C and 90 min reaction time using Ar/BC molar ratio of 15:1.

CH3

CH3

>

>

O

Catalyst

BC conversion (%)

DPM (%)

Di-substituted products (%) 1,2-DBB

1,3-DBB

1,4-DBB

Fresh 1st reuse 2nd reuse

100 100 100

86.9 86.0 80.0

4.8 5.0 6.9

1.7 1.9 3.0

6.6 7.1 10.1

Table 5 Leaching test for RHA-Ga catalyst in the benzylation of benzene at 80 °C using Bz/BC molar ratio of 15:1 Time (min)

50 90a 120a a

BC conversion (%)

63.1 66.0 68.9

DPM (%)

87.6 86.7 85.6

Di-substituted products (%) 1,2-DBB

1,3-DBB

1,4-DBB

4.4 4.7 5.2

1.7 1.8 1.9

6.3 6.8 7.3

Reaction after catalyst has been removed.

3.2.7. Recyclability The most important advantage of using heterogeneous catalysts is the possibility of their regeneration and reusability. The recyclability of the catalysts had been studied by running the reaction successively (under the optimum reaction conditions but with a reaction time of 100 min) with the same catalyst (RHA–Ga) in the same reaction vessel according to published method [22,25]. The results are shown in Table 4. These results indicate the catalyst could be used several times in the benzylation process without any significant change in its percentage conversion and selectivity. 3.2.8. Leaching effect It was concluded by Arends et al. [38] that leaching is expected for heterogeneous catalysts especially in a reaction involving oxidation catalysts, owing to the solvolysis of the metal-oxygen bonds through which the metal is attached to the silica support. In order to study the heterogeneous nature of the reaction, a test reaction using RHA–Ga was conducted. The catalyst was removed after the first 50 min of the reaction as explained elsewhere [25], and the reaction was allowed to proceed further without the catalyst. Table 5 shows the results of this study. Neither significant increase in the BC conversion nor any differences in the product distribution was observed after the removal of the catalyst. This indicated that there was no leaching and the catalysis was purely heterogeneous. These results also emphasized the fact that the heterogeneous RHA–Ga catalyst was necessary for the reaction to take place. Even if the gallium had leached out into the solution (not tested), it had not catalyzed the reaction further. It is therefore safe to conclude that a purely heterogeneous catalytic reaction was taking place over RHA–Ga catalyst.

80

50

Table 4 The recyclability of RHA–Ga catalyst in the benzylation of benzene at 80 °C using Bz/ BC molar ratio of 15:1

CH3

>

Scheme 1. The series showing the decrease in the reaction rate for benzylation of benzene and substituted benzenes with benzyl chloride.

4. Conclusion The catalytic activity of RHA–Al, RHA–Ga and RHA–Fe were tested in the benzylation of benzene employing benzyl chloride as the alkylating agent. RHA–Fe showed the highest catalytic activity whereas, RHA–Ga gave the highest selectivity to DPM. However, RHA–Al was almost inactive and this was attributed to many reasons including the presence of the Al3+ ions in a tetrahedral orientation on the silica network making them inaccessible to the reactants and its low redox property. For the benzylation of benzene and substituted benzenes, RHA–Ga shows the trend as in Scheme 1. This was opposite to that observed for the classical acid

A.E. Ahmed, F. Adam / Microporous and Mesoporous Materials 118 (2009) 35–43

catalyzed Friedel–Crafts type benzylation reactions. The extremely low activity of RHA–Ga in the reaction with anisole was suggested to be due to the strong adsorption of the anisole on the catalyst surface. The catalyst was reused for the benzylation reaction several times without any significant change in its activity and selectivity. Other advantages of the RHA–Ga and RHA–Fe catalysts over previously reported Ga and Fe supported catalysts are: (a) easily prepared at room temperature from rice husk ash which is an agricultural waste material, (b) no complicated procedures or additional chemicals or surfactant template molecules were required for the preparation of the catalyst, (c) no need for calcination after catalyst preparation which is the normal procedure in most papers published previously, (d) comparable catalytic activity and reusability with previously published catalysts and (e) RHA–Ga and RHA–Fe catalysts are not moisture sensitive and can be handled and stored normally. Acknowledgments The authors would like to express their thanks to Leong Guan Rice Mill Sdn. Bhd, Penang, Malaysia for providing the rice husk. We would also like to thank the Malaysian Government for the FRGS Grant (A/c No. 203/PKIMIA/671021) for partly supporting this work. We thank the Sudan University of Science and Technology, Sudan for a scholarship to Adil Elhag Ahmed. References [1] T.B. Reed, Biomass Gasification Principles and Technology, Noyes Data, Park Ridge, NJ, 1981. [2] N. Yalçin, V. Sevinç, Ceram. Int. 27 (2001) 219. [3] R. Conradt, P. Pimkhaokham, U. Leela-Adison, J. Non-Cryst. Solids 145 (1992) 75. [4] U. Kalapathy, A. Proctor, J. Shultz, Bioresour. Technol. 73 (2000) 257. [5] T.-H. Liou, Mater. Sci. Eng. A 364 (2004) 313.

[6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38]

43

C. Real, M.D. Alcala, J.M. Criado, J. Am. Ceram. Soc. 79 (1996) 2012. V.P. Della, I. Kühn, D. Hotza, Mater. Lett. 57 (2002) 818. U. Kalapathy, A. Proctor, J. Shultz, Bioresour. Technol. 85 (2002) 285. S.R. Kamath, A. Proctor, Cereal Chem. 75 (1998) 484. G.A. Olah, Friedel–Crafts and related reactions, Wiley Interscience, New York, 1963. Y. Sun, S. Walspurger, J.P. Tessonnier, B. Louis, J. Sommer, Appl. Catal. A 300 (2006) 1. A. Vinu, D.P. Sawant, K. Ariga, M. Hartmann, S.B. Halligudi, Micropor. Mesopor. Mater. 80 (2005) 195. M. Selvaraj, T.G. Lee, J. Mol. Catal. A 243 (2006) 176. K. Okumura, K. Nishigaki, M. Niwa, Micropor. Mesopor. Mater. 44–45 (2001) 509. M.S. Hamdy, G. Mul, J.C. Jansen, A. Ebaid, Z. Shan, A.R. Overweg, Th. Maschmeyer, Catal. Today 100 (2005) 255. E. Diaz, S. Ordonez, A. Vega, A. Auroux, J. Coca, Appl. Catal. A 295 (2005) 106. V.R. Choudhary, S.K. Jana, N.S. Patil, S.K. Bhargava, Micropor. Mesopor. Mater. 57 (2003) 21. A.B. Deshpande, A.R. Bajapai, S.D. Samant, Appl. Catal. A 209 (2001) 229. T. Cseri, S. Békássy, F. Figueras, S. Rizner, J. Mol. Catal. A 98 (1995) 101. M.S. Niasari, J. Hasanalian, H. Najafian, J. Mol. Catal. A 209 (2004) 209. Y. Izumi, M. Ogawa, K. Urabe, Appl. Catal. A 132 (1995) 127. V.R. Choudhary, S.K. Jana, Appl. Catal. A 224 (2002) 51. F. Adam, K. Kalaivani, B. Saraswathy, J. Colloid Interface Sci. 304 (2006) 137. F. Adam, J. Andas, J. Colloid Inter. Sci. 311 (2007) 135. A.E. Ahmed, F. Adam, Micropor. Mesopor. Mater. 103 (2007) 284. K.S.W. Sing, IUPAC Reporting Physisorption Data for Gas/Solid Systems, Pure Appl. Chem. 57 (1985) 603. G.M.S. El Shafei, N.A. Moussa, Z.A. Omran, Powder Technol. 107 (2000) 118. Á. Szegedi, Z. Kónya, D. Méhnb, E. Solymár, G. Pál-Borbély, Z.E. Horváth, L.P. Biró, I. Kiricsi, Appl. Catal. A 272 (2004) 257. D.A. Maria, Z.H. Luan, J. Klinowski, J. Phys. Chem. 100 (1996) 2178. R.F.S. Lenza, W.L. Vasconcelos, Mater. Res. 4 (2001) 189. J.M. Fraile, J.I. Garcia, J.A. Mayoral, E. Pires, L. Salvatella, M. Ten, J. Phys. Chem. B 103 (1999) 1664. I. Jaymes, A. Douy, J. Eur. Ceram. Soc. 16 (1996) 155. R.J. Madon, M. Boudar, Ind. Eng. Chem. Fundam. 21 (1902) 438. V.R. Choudhary, S.K. Jana, B.P. Kiran, Catal. Lett. 64 (2000) 223. V.R. Choudhary, K. Mantri, Micropor. Mesopor. Mater. 56 (2002) 317. S.J. Barlow, T.W. Bastock, J.U. Clark, S.R. Cullen, Tetrahedron Lett. 34 (1993) 3339. N. He, S. Bao, Q. Xu, Appl. Catal. A 169 (1998) 29. I.W.C.E. Arends, R.A. Sheldon, Appl. Catal. A 212 (2001) 175.