Catalytic performance of organically templated nano nickel incorporated-rice husk silica in hydroconversion of cyclohexene and dehydrogenation of ethanol Academic research paper on "Chemical sciences"

Egyptian Journal of Petroleum (2013) xxx, xxx-xxx

Egyptian Petroleum Research Institute Egyptian Journal of Petroleum

www.elsevier.com/locate/egyjp www.sciencedirect.com

FULL LENGTH ARTICLE

Catalytic performance of organically templated nano nickel incorporated-rice husk silica in hydroconversion of cyclohexene and dehydrogenation of ethanol

Salah A. Hassan a, Ahmed M. Al-Sabagh b, Nasser H. Shalaby b *, Samia A. Hanafi b, Hamdi A. Hassan a

Department of Chemistry, Faculty of Science, Ain Shams University, Abbassia, Cairo, Egypt ' Egyptian Petroleum Research Institute, Nasr City 11727, Cairo, Egypt

KEYWORDS

Rice husk silica; Organically templated nano nickel-rice husk silica; Cyclohexene hydroconversion;

Dehydrogenation of ethanol

Abstract Rice husk silica (RHS) was extracted from local rice husk by acid digestion and burning at 650 °C. RHS-Ni catalyst was prepared by dissolving RHS in 1 N NaOH and titrating with 3 N HNO3 containing 10 wt.% Ni2+. The organic modifiers, eitherp-amino benzoic acid (A) orp-phen-ylenediamine (PDA) were incorporated in 5 wt.% and reduced in H2 flow. Investigation of the three catalysts, (RHS-Ni)R350, (RHS-Ni-A)R350 and (RHS-Ni-PDA)R350, confirmed good dispersion of Ni nanoparticles; all catalysts were amorphous. The BET surface areas increased in the order: (RHS-Ni)R350 < (RHS-Ni-A)R350 < (RHS-Ni-PDA)R350 with controlled pore sizes. The as-prepared catalysts were applied for both hydroconversion of cyclohexene with molecular H2 and eth-anol dehydrogenation, using a flow-type reactor, at different temperatures. The activity in cyclohexene hydroconversion and selectivity to cyclohexane depended upon the reaction temperature; at t < 150 °C, the increased hydrogenation activity was referred to the formed SiO2-Ni-amine complex, pore regulation as a prime requirement for H2 storage and homogeneous distribution of incorporated Ni nanoparticles. At t > 150 °C, the backward dehydrogenation pathway was more favored, due to unavailability of H2; the process became structure-sensitive. In ethanol conversion, the prevailing dehydrogenation activity of organically modified catalyst samples was encouraged by improved homogeneous distribution of Ni nanoparticles and created micropre system.

© 2012 Egyptian Petroleum Research Institute. Production and hosting by Elsevier B.V.

All rights reserved.

* Corresponding author.

E-mail address: chem.shalaby@gmail.com (N.H. Shalaby).

Peer review under responsibility of the Egyptian Petroleum Research

Institute.

1. Introduction

Rice husks (RH), removed as waste through rice processing in mills, are normally burned in the fields, in open air, causing serious environmental and health problems. They have become a promising source of high-grade amorphous silica, extracted from rice husk ash by a suitable alkali solution or acid digestion. The amorphous silica is commonly used as a support

1110-0621 © 2012 Egyptian Petroleum Research Institute. Production and hosting by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ejpe.2012.09.009

material due to its high surface area and narrow pore size distribution for metal dispersion [1,2]. Preparation of catalysts utilizing rice husk can be considered a cheaper and much lower energy consuming alternate to the current use of TEOS as a starting material for most silica-based catalysts. Organofunc-tionalization of porous silica has attracted much attention, because it combines a high surface area and a narrow pore size distribution with the advantage of organic ligand properties [3].

Introduction of several organic templates, e.g., amine-ter-minated molecules, has been used to synthesize well regulated pore structures in silica-based catalysts [4,5] and also to produce various types of nanoparticles [6]. For instance, 4-(methyl amino) benzoic acid incorporated iron-silica catalyst extracted from rice husk (RH) was investigated in benzylation of toluene, where a drastic reduction in the disubstituted products was shown [7]. The ca. 25% increase in specific surface area of RH-Fe (5% amine), compared to RH-Fe, was attributed to the templated formation of regular pores. However, incorporated 4-(methyl amino) benzoic acid into RH silica-ruthenium catalyst, utilized in the oxidation of 1-butanol, showed 40% loss in specific surface area, compared to RH-Ru, due to blocking of pores by the added amine [5]. The narrow pores could be referred to the formation of SiO2-Ru-amine complex that allowed controlling the pore size formation, e.g., calcination at 700 0C yielded particles with large and regular pores, encouraging the catalyst selectivity for the oxidation of 1-buta-nol to butanal.

On the other hand, catalytic conversion of cyclohexene (CHE), an intermediate in various industrial processes, has been extensively studied as an important probe reaction for several catalytic systems (e.g., [8,9]). Different products could be obtained depending on catalyst composition and operating conditions. In the absence of hydrogen, disproportionation and dehydrogenation took place [10]. Disproportionation of CHE yielded benzene and cyclohexane (CHA) (hydrogen transfer) by using metal catalysts, since CHE acts as hydrogen donor and acceptor. By using silica-supported copper catalyst, benzene and cyclohexane were formed together, due to the partial reactivation of the catalytic surface [10,11]. Recently, the effect of particle size on reaction selectivity for the conversion of cyclohexene in excess hydrogen was examined in the presence of transition metal nanoparticle SBA-15-supported catalysts [12]. It was demonstrated that in a regime where both cyclohexane and benzene form, the surface is depleted in hydrogen and the particle size influences reaction selectivity through a particle size dependent reactive hydrogen coverage. The hydrogenation of cyclohexene is structure insensitive under conditions of reversible hydrogen adsorption (low temperature) and becomes structure-sensitive with an apparent dependence on hydrogen pressure at higher temperatures.

Also, among the important chemical processes is the partial oxidation of ethanol to produce acetaldehyde and hydrogen. It can be used as a hydrogen source for various fuel cells [13,14]. Ethanol is considered promising of many candidates for the next generation of energy carriers; it can be easily obtained in large amounts by fermentation of biomass. Moreover, its transport, storage, and distribution are relatively unproblem-atic. Biomass-based ethanol is currently made from cornstarch in USA, wheat-, barley-, and rye-starch in Europe, and from sugar cane-derived sucrose in Brazil. Current market prices encourage the transformation of ethanol into acetaldehyde

but not into acetic acid [15]. This is because major industrial production for acetic acid is based on the carbonylation of methanol with an Rh complex catalyst, whereas acetaldehyde production needs ethylene as a source and PdCl2-CuCl2 as a catalyst. Owing to the gradual shift from fossil to renewable resources in chemical industry, transformation of ethanol into valuable chemical feed stocks has attracted growing concerns.

The aim of the present work was to extract active amorphous silica from local rice husk as a cheap source, fine-tuned through modification with incorporated nano sized nickel and adopting p-amino benzoic acid or p-phenylenediamine as the organic template. The finished catalytic systems were characterized by the aid of XRD, FTIR, N2-adsorption-desorption isotherms and TEM techniques. The catalytic performances of these systems in hydroconversion of cyclohexene with molecular H2 and in the dehydrogenation of ethanol, at different temperatures using a flow-type reactor were followed up with the aim of elucidating the roles played by the incorporated nano nickel and organo-functional moieties attached to the silica surface.

2. Experimental

2.1. Extraction of rice husk silica (RHS)

The rice husk used in the present study was collected from rice husk mills at Kafr El-Zayat, West Delta region, Egypt. It was washed well with distilled water to remove all the contaminants and dried at 110 0C for 3 h. The clean husk was refluxed in 3 N HNO3 for 4 h to reduce the metallic contents. The digested husk was washed with distilled water, till the washing was neutral, and then dried at 110 0C for 3 h. The acid-leached RH was burned in a muffle furnace at 650 0C for 4 h to produce the rice husk silica (RHS).

2.2. Preparation of silica xerogel gel (SG)

The RHS (12.0 g) was added to 600 ml of 1.0 N NaOH in a Teflon container and stirred overnight at room temperature to extract the pure silica as sodium silicate in which traces of contaminated metallic impurities precipitated as hydroxides. The solution was filtered and the extracted sodium silicate was titrated with 3.0 N HNO3 at a slow rate of ca. 1.0 mL min-1 with constant stirring. The titration was continued till the solution pH reached 5.0 [1]. The silica gel/precipitate was aged for 24 h. The silica gel was washed thoroughly with distilled water, recovered by centrifugation and dried at 100 0C overnight. The produced xerogel (SG) was ground and washed again with distilled water for complete removal of nitrate ions.

2.3. Preparation of nano nickel incorporated-rice husk silica

Six grams of RHS was added to 300 ml of 1 N NaOH in a Teflon container and stirred overnight. After filtration, the extracted sodium silicate solution was titrated with 3 N HNO3 containing 10% (w/w) Ni2+ (Ni (NO3)2) till it reached pH 5.0. The suspension was aged in a Teflon container for 2 days. The gel was separated by centrifugation, washed thoroughly with distilled water then with acetone (influences the pore characteristics [5]) and dried at 110 0C for 24 h. The product was

Table 1 BET surface parameters of SG, (RHS-Ni)R350, (RHS-Ni-A)R350 and (RHS-Ni-PDA)R350.

Sample

BET surface area (m2/g)

Vp (cc/g)

Pore diameter (nm)

Metal content (%)

(RHS-Ni)R350

(RHS-Ni-A)R350

(RHS-Ni-PDA)R350

702.78 296.5 301.13 364.13

0.423 0.578 0.233 0.153

1.63 2.16 1.89

12.4 11.8 11.3

labeled as RHS-Ni. Four grams of the prepared RHS-Ni was reduced with H2 flow, at a rate of 20 ml min-1, for 4 h at 350 0C. The obtained catalyst was denoted as (RHS-Ni)R350.

2.4. Preparation of amino benzoic acid templated nano nickel incorporated-rice husk silica

Figure 2 XRD patterns of various organically templated Ni incorporated-RHS samples.

20 ml min-1, for 4 h at 350 0C, labeled as (RHS-Ni-Ph)R350. Another sample was reduced for 4 h at 600 0C and labeled as (RHS-Ni-PDA)R600.

p-Amino benzoic acid (0.316 g) was added to 300 ml of sodium silicate solution extracted from RH and the mixture was titrated with 3 N HNO3 containing 10% (w/w)Ni2+ till it reached pH 5.0 [5]. The suspension was aged in a Teflon container for 2 days and the gel was separated, washed and dried according to the procedure mentioned above for RHS-Ni. The produced material was labeled as RHS-Ni-A (A: p-amino benzoic acid incorporated in 5wt.%). Similarly, 4 g of the prepared material, RHS-Ni-A, was reduced with H2 for 4 h at 350 0C and was denoted as (RHS-Ni-A)R350.

2.5. Preparation of p-phenylenediamine templated nano nickel incorporated-rice husk silica

Similar to RHS-Ni-A, RHS-Ni-PDA (PDA: p-phenylenedi-amine incorporated in 5wt.%) was prepared. A sample of RHS-Ni-PDA was reduced with H2, at a flow rate of

2.6. Catalyst characterization

Structural characteristics of the prepared samples were investigated through FTIR analysis using ATI Mattson WI 53717 model Genesis spectrometer, USA, with a resolution of 2 cm-1 and XRD using a X-ray diffractometer, PANalytical model X'Pert PRO, operated with a Cu-K radiation (k = 1.5418 A) at a scanning rate of 0.3 degree min-1. The metal content of the catalysts was determined using a flame atomic absorption spectrometer ZEEnit 700P-Analytik Jena -Germany (Table 1). Pyrolysis or thermal decomposition behavior of the precursors was studied in N2 atmosphere using a TA Instrument SDTQ 600 simultaneous TGA-DSC thermo-gravimetric analyzer with a heating rate of 5 0C min-1 and a nitrogen flow of 100 ml min-1.

The texture of various solid samples was investigated via BET surface area determination and pore analysis, based on

Figure 3 IR spectra of the prepared samples and organic modifiers (PDA and A).

N2-adsorption-desorption isotherms at —196 °C, recorded using a Quantachrome Autosorb-1 adsorption analyzer. The morphology as well as the average particle size of the active phase in the nanocomposite systems was examined by a JEM-1230 Transmission Electron Microscope (TEM) working at 120 kV with 0.2 nm resolution.

2.7. Activity tests for the catalysts

The hydrogenation of cyclohexene in vapor phase with hydrogen over the prepared catalysts was carried out in a quartz flow reactor (of 0.8 cm ID), placed in a tubular furnace equipped with a thermocouple (see Fig. 1). In each experiment, 0.4 g

ppo Relative Pressure PIP"

Figure 4 N2 adsorption-desorption isotherm for SG, (RHS-Ni)R350, (RHS-Ni-A)R350 and (RHS-Ni-PDA)R350.

Figure 5 The probable formation of pore structure involving RHS, Ni and the amine: RHS-Ni-A, (a) RH-Ru-A (from Adam and Sugiarmawan [5]) (b) RHS-Ni-PDA.

Figure 6 PSD curves of SG, (RHS-Ni)R350, (RHS-Ni-A)R350 and (RHS-Ni-PDA)R350.

of catalyst was mixed with 0.8 g of porcelain pieces, matched in particle size with the catalyst, and placed in the hot zone of the reactor. The reaction temperature was changed from 100 to

350 0C. Cyclohexene was introduced into the reactor by means of a dosing pump at a flow rate of 0.6 ml min-1. Hydrogen was injected at a flow rate of 20mlmin~1. The products and the

unreacted cyclohexene leaving the reactor were analyzed using a PerkinElmer chromatograph, model Clarus 500, using a stainless steel 6" x 1/8" OD SS column for auto system GC (NOC), packed with Chromosorb P-AW 80/100 mesh, 30% silicone DC-200 500 CSTK. The following operating conditions were applied: oven temperature: from 80 to 200 0C, heating rate: 5 0C min-1, hold time: 30 min and FID temperature: 200 0C. The same technique was used for ethanol dehydroge-nation, using Rtx®-Wax (Crossbond®-PEG) column (30 m x 0.25 mm ID) with the same operating conditions.

The ethanol conversion, fractional selectivity and yield (%) were calculated from chromatographic peaks according to:

Ethanol conversion (Total conversion), wt%

— y^ yield (%) of each component in the products

Acetaldehyde yield, wt% =

Quantity of acetaldehyde

Total quantity of the product components

Acetaldehyde yield, wt% — Acetaldehyde fractional selectivity (wt%)

— yield% acetaldehyde/total conversion x 100

[The % yield of each component is calculated by the mathematical program (Total Chrom Workstation Version 6.3) of GC)]. The same equations were applied for cyclohexene hydroconversion for selective production of cyclohexane.

3. Results and discussion

3.1. XRD analysis of different organically templated Ni incorporated-RHS samples

Fig. 2 shows the XRD patterns for the samples reduced with H2 at 350 0C, namely, (RHS-Ni)R350, (RHS-Ni-A)R350 and (RHS-Ni-PDA)R350. In all cases, a broad peak was observed in the 2h range between 150 and 300, being characteristic of amorphous silica. The characteristic peaks of Ni could not be detected, indicating its highly dispersed nature and/or its existence as nano sized particles, located most probably in the created silica mesopores. However, for (RHS-Ni-PDA)R600, the characteristic peaks of Ni metal were observed with d spacings: 2.036 A at 2h = 44.50 and 1.77 A at 2h = 51.50, ascribed mainly to (111) and (200) crystallite surfaces (International Center of Diffraction Data -ICDD-2010), i.e., the particle size seemed to increase considerably by rising the temperature. The average size of Ni particles, calculated by the aid of the Scherrer equation [16], based on the obtained full width at half maximum peak heights (FWHM) ranged between 5.7 and 7.7 nm (cf., evidences from TEM investigation, described below). The (RHS-Ni-PDA)R350 sample was also examined, after exploitation in the test reaction, denoted as (RHS-Ni-PDA)Re, where the pattern remained intact, ensuring robustness of the catalyst structure during different catalytic cycles at 350 0C.

3.2. FTIR analysis of investigated samples

Fig. 3 depicts the FTIR spectra of SG, various organically templated Ni-incorporated RHS catalysts and organic modifiers. It is evident that the IR spectrum of neat silica xerogel

(SG) is markedly different from that of Ni incorporated-RHS samples, with or without organic modifier, referring probably to some Ni-silica interactions.

The obtained spectra of organically templated Ni-incorpo-rated RHS catalysts are typical for hybrid materials with both components, organic and inorganic phases.

The inorganic component can be identified from the Si-O-Si modes observed below 1250 cm-1, the typical silica overtone bands at ca. 1846 cm-1 and the silanol stretching bands above 3300 cm-1. The weak band observed at 870 cm-1 for the entire catalysts may be attributed to Si-O-Ni [1]. For RHS-Ni-PDA, small bands at 1395,1434,1519and1543 cm-1 are observed corresponding to C=C benzene ring stretching which often occur in pairs. The band around 1635 cm-1 can be attributed to the bending vibration of water molecules bound to the silica matrix. It appears to be more intense in the organically modified precursors than in the case of unmodified precursor, due to overlapping with the band of N-H bending and/or C=O stretching vibration. The shoulder at 1220 cm-1 is due to C-N stretching. The C-H stretching of the benzene ring is observed at 37553819 cm-1. For RHS-Ni-A, the three bands at 1634, 1567 and 1507 cm-1 could be associated to the asymmetric stretching and symmetric stretching of the carboxylate ion.

The interaction between the carboxylate ion and the metal atom was classified into four types: monodentate, bridging (bidentate), chelating (bidentate) and ionic interaction. It can be identified by the wave number separation (D) between IR bands [5]. D lying in the range of 200-320 cm-1 refers to the monodentate interaction, while D < 110 cm-1 refers to the chelating bidentate. Between the two ranges, the separation of IR bands refers to the bridging bidentate. From the FTIR spectra obtained, two types could be distinguished: (i) chelat-ing bidentate having D (1634 - 1567 = 67 cm-1) or (1567 - 1507 = 60 cm-1); (ii) bridging bidentate having D (1634 - 1507 = 127 cm-1). In the IR spectra of three catalysts, the weak band at 870 cm-1 characteristic for Si-O-Ni disappeared due to the reduction of Ni2+ into Ni0. The bands characteristic for the organic part disappeared due to the pyro-lysis during the reduction of incorporated metal precursor although some bands are common in the organic modifier and the organically templated samples.

3.3. BET analysis

The obtained adsorption-desorption isotherms of N2 at -196 0C on the surface of the studied supported and organically modified RHS-Ni samples are illustrated in Fig. 4, being almost of type II in the IUPAC classification. The adsorption isotherm for RHS-Ni shows a wider hysteresis loop instead of nearly retracing the adsorption curve as in the hysteresis loops of SG, (RHS-Ni-A)R350 and (RHS-Ni-PDA)R350 samples; It is typically of mesoporous materials, i.e., those having pores with openings greater than 2 nm. Such pores are likely to have a wide range of sizes and shapes and also may interconnect with one another. The hysteresis loops in the cases of SG, (RHS-Ni-A)R350 and (RHS-Ni-PDA)R350 are closed in the pressure region near saturation revealing that these samples contain mesopores with an upper size restriction [17].

From the BET surface analysis in Table 1, it is found that the specific surface area decreases in the order of SG > (RHS-Ni-PDA)R350 > (RHS-Ni-A)R350 > (RHS-Ni)R350. Comparing

Figure 7 The TEM image of: (a) (RHS-Ni-A)R350, (b) RHS-Ni-A, (c) (RHS-Ni)R350, (d) (RHS-Ni-PDA)R350, (e) (RHS-Ni-PDA)R600, (f) Ni/TiO2 reduced at 600 °C (from Wu et al. [18]) (g) (RHS-Ni-PDA)Re.

the surface area of (RHS-Ni)R350 with that of the neat xerogel, SG, a dramatic decrease in surface area (more than 50%) can be observed; due to the incorporation of Ni into SG pores. However, the increase in average pore size and pore diameter for this catalyst sample may be linked with the number and some regulation of pores. The increased surface area of both organically modified samples, viz., (RHS-Ni-PDA)R350 and (RHS-Ni-A)R350, seems to be associated with the pore regulation and formation of much narrower pores through the formation of SiO2-Ni-amine complex as depicted in Fig. 5.

The pore size distribution (PSD) of (RHS-Ni-PDA)R350 is shown to be in a single narrow range of pore radii (13-15 A)

(Fig. 6). This may be due to the symmetric functionality of p-phenylenediamine that creates typical pores as suggested in Fig. 6. For (RHS-Ni-A)R350, three different pore ranges are evident. The first major range is from 5 to 16 A, the second range of wider pores is from 16 to 22 A, and the last mesopor-ous range, from 30 to 40 A, with least fraction. For (RHS-Ni)R350, five different pore ranges are observed, where pore creation results mainly through the reduction of Ni (NO3)2 into nickel metal leaving several pores, dimension of which is controlled by concentration of the used Ni (NO3)2 and diffusion of the produced gases (Fig. 6). For parent SG, two ranges of micro pore sizes are observed. Generally, the three catalysts

Figure 8 DSC and TGA curves of RHS-Ni-PDA catalyst precursor.

(RHS-Ni)R350, (RHS-Ni-A)R350 and (RHS-Ni-PDA)R350 are shown to be mesoporous materials, where the organically tem-plated samples are clearly of regulated pores (i.e., with generated micro pores).

3.4. Morphology of nanocatalysts

Fig. 7 shows the TEM images of (RHS-Ni)R350, (RHS-Ni-A)r350, (RHS-Ni-PDA)R350, (RHS-Ni-PDA)R600 and (RHS-Ni-PDA)Re catalyst samples as well as the RHS-Ni-A catalyst precursor (xerogel), before reduction with different magnification power. In the TEM images (a and b) for (RHS-Ni-A)R350 and RHS-Ni-A (xerogel), while well-shaped Ni nanoparticles are homogeneously distributed in the reduced sample, much less of the incorporated Ni particles can be detected before reduction as the majority of Ni species remains as ions (i.e., still in the nitrate form, Ni(NO3)2). The image (c) of organically unmodified (RHS-Ni)R350 shows agglomerated Ni particles cited on the silica surface. For the (RHS-Ni-PDA)R350 sample, the image (d) shows more randomly distributed Ni nanoparticles. Generally, for all the catalyst samples under study, of almost amorphous structure, the Ni particle size ranges from 3 to 7 nm, distribution of which depends on the method of preparation (cf., XRD results). By reduction at

600 0C, Ni particles tend to be more incorporated in silica matrix, aggregating in 5-10 nm sizes. They seem to crystallize in the pores, in homogenous dispersion still restricted by robust silica structure. Silica matrix remains unaffected, i.e., not undergoing destruction, in contrast to the Ni/TiO2 nanocom-posite prepared via the same technique (sol-gel) [18], where most of the pores were destroyed as a result of phase transformation and particle growth (image (f)).

Moreover, the image (g) of the exploited catalyst (RHS-Ni-PDA)Re, after hydroconversion of cyclohexene, reveals inho-mogeneous distribution of some enlarged Ni particles, yet still resistant to reaction conditions.

3.5. TGA-DSC analysis of RHS-Ni-PDA catalyst precursor

The pyrolysis of RHS-Ni-PDA was studied in N2 atmosphere. Fig. 8 shows the DSC and TGA curves of RHS-Ni-PDA catalyst precursor, where a broad endothermic peak at 100 0C with a steep weight loss up to 120 0C may be attributed to the volatilization of H2O (ca. 11.1% loss). Above this temperature up till 500 0C, the rate of weight loss decreases which could only be attributed to the decomposition of both the organic modifier and the Ni (NO3)2 (ca. 7.58% loss). The phase transformation from amorphous to crystalline Ni seems to be associated with a small endothermic peak, as revealed from the XRD pattern. Above this temperature, the weight loss may be attributed to the evaporation of the residual of the charred organic matter.

3.6. Catalytic activity

The cyclohexene (CHE) hydroconversion was carried out at temperatures ranging between 100 and 350 0C, in the presence of (RHS-Ni)R350, (RHS-Ni-A)R350 and (RHS-Ni-PDA)R350 samples. Maximum conversion could be obtained in the range between 100 and 150 0C over all the catalysts used, achieving almost 100% selectivity to cyclohexane (Fig. 9). Above these temperatures, the % conversion underwent continuous decrease, where benzene fraction showed a gradual increase. The hydrogenation activity of the studied catalyst systems at the mentioned reaction conditions can be arranged in the following order: (RHS-Ni-PDA)R350 > (RHS-Ni-A)R350 > (RHS-Ni)R350 (Table 2), running in harmony with BET surface areas (Table 1). It is evident that the activity of organically modified RHS-incorporated Ni samples exceeds that of the

Figure 9 Production and selectivity to cyclohexane as a function of reaction temperature in CHE hydroconversion in the presence of organically modified RHS-incorporated Ni catalysts.

Table 2 Effect of reaction temperature on catalytic hydroconversion of cyclohexene over the nanocomposite catalysts (C = cyclo-hexane, B = benzene and S = selectivity toward cyclohexane production).

T (°C) (RHS-Ni)R350 (RHS-Ni-A)R350 (RHS-Ni-PDA)R350

%C %B %S %C %B %S %C %B %S

100 89.1 0 100 94.3 0 100 98.1 0 100

150 95.8 0 100 96.8 0 100 98.6 0.503 99.5

200 88 6.1 93.5 89 6.3 93.4 84.1 12.3 87.24

250 72 15.4 82.3 72.5 20.1 78.3 69.8 24.7 73.86

300 57.6 23.2 71.3 59.2 28.6 67.4 51.9 33.7 60.63

350 41.4 34.6 54.47 51.2 30.1 63 38.9 43.8 47

Figure 10 Production and selectivity to acetaldehyde as a function of reaction temperature in ethanol dehydrogenation in the presence of organically modified RHS-incorporated Ni catalysts and the exploited catalyst.

Scheme 1 Ethanol catalytic conversion pathway for acetaldehyde and acetone production.

unmodified catalyst (RHS-Ni)R350. This may be referred to the formed SiO2-Ni-amine complex [5], associated with pore regulation as a prime requirement for H2 storage. This may be encouraged also by homogeneous distribution of incorporated Ni nanoparticles, i.e., the selectivity here may be owed to particle size dependent reactive hydrogen coverage [12]. At temperatures >150 °C, the disproportionation and consequently the backward dehydrogenation pathway become more favored, due to the unavailability of H2, being probably entrapped in the generated micropores, i.e., the hydroconversion becomes structure-sensitive with an apparent dependence on hydrogen pressure at higher temperatures [12]. The catalytic conversion of ethanol was also carried out over the investigated catalysts at different temperatures between 100 and 250 °C. The ethanol conversion was found to increase with

temperature in all cases (Fig. 10), yielding mainly acetaldehyde with a considerable amount of acetone. These results and previous literature reports [19,20] lead us to propose the reaction pathway described in Scheme 1. In this scheme, ethanol is converted on the surface of Ni nanoparticles predominately via dehydrogenation into acetaldehyde. Aldol intermediates formed from acetaldehyde condensation dehydrate to form crotonaldehyde which undergoes an intramolecular hydride shift leading to the keto form which decomposes to acetone. The dehydrogenation activity of the studied catalyst systems decreased at the mentioned reaction conditions in the following order: (RHS-Ni-A)R350 > (RHS-Ni-PDA)R350 > (RHS-Ni)R350 (Table 3). The prevailing activity of organically modified catalyst samples seems to be encouraged by the improved homogeneous distribution of Ni nanoparticles as well as by the

Table 3 Catalytic behavior of ethanol dehydrogenation over the nanocomposite catalysts at different temperatures (Ald = acet-aldehyde, Ac = acetone and S = selectivity toward acetaldehyde production).

T (°C) (RHS-Ni)R350 (RHS-Ni-A)R350 (RHS-Ni-PDA)R350 (RHS-Ni-PDA)R600 (RHS-Ni-PDA)Re

%Ald %Ac %S %Ald %Ac %S %Ald %Ac %S %Ald %Ac %S %Ald %Ac %S

100 4.94 9.3 34.7 34.4 5.62 85.96 17.65 23.03 43.38 14.5 0 100 22.7 0 100

150 6.5 11.1 37 36.35 6.61 84.6 20.1 26.2 43.41 6.3 0 100 14.1 0 100

200 7.6 9.51 44.42 44.9 6.1 88.04 21.21 27.1 43.9 9.67 0 100 11.2 0 100

250 9.1 10.61 46.17 47.11 9.21 83.65 24.47 26.35 48.15 11.94 0 100 7.97 0 100

created trapping micropre system. For the catalyst sample (RHS-Ni-PDA)R600, produced by the reduction of RHS-Ni-PDA precursor at 600 0C, a dramatic decrease in activity could be observed, yet with 100% selectivity toward acetaldehyde, being linked with the size enlargement of Ni nanoparticles, i.e., the process is mainly structure-sensitive. This may also confirm the importance of the homogeneous distribution of Ni nanoparticles and the created micropre system in the overall conversion of ethanol.

Of special interest from the application point of view, the exploited catalyst (RHS-Ni-PDA)Re in the CHE hydroconversion for ca. 5 h at temperatures up to 350 0C was tried again in the eth-anol dehydrogenation. The activity results were somewhat better than those of the sintered sample and even at lower temperatures (100-150 0C). The activity seems to be comparable with that of the original catalyst reduced at 350 0C, indicating the possibility of reuse of this composite system for several times.

4. Conclusion

Rice husk Silica (RHS) was extracted from the local rice husk by acid digestion and burning at 650 0C for 4 h. It was dissolved in 1 N NaOH, the solution was titrated with 3 N HNO3 containing 10 wt.% Ni2 +, and the organic modifiers, p-amino benzoic acid (A) orp-phenylenediamine (PDA) were incorporated in 5 wt.%. The produced pastes were reduced in H2 flow. Investigation of the three catalysts, confirmed good dispersion of Ni nanoparti-cles in (RHS-Ni-A)R350 and (RHS-Ni-PDA)R350 composites, in contrast to aggregated particles in the original (RHS-Ni)R350 supported catalyst. The activity of studied catalyst systems in hydroconversion of cyclohexene at 100-350 0C decreased in the order: (RHS-Ni-PDA)R350 > (RHS-Ni-A)R350 > (RHS-Ni)R350, in agreement with BET surface areas. At temperatures < 150 0C, the increased hydrogenation activity (maximized selectivity toward cyclohexane) of organically modified RHS-incorporated Ni samples, (RHS-Ni-PDA)R350 and (RHS-Ni-A)R350, compared with that of unmodified catalyst (RHS-Ni)R350, could be referred to the formed SiO2-Ni-amine complex, pore regulation as a prime requirement for H2 storage and homogeneous distribution of incorporated Ni nanoparti-cles. The selectivity was owed to the particle size dependent reactive hydrogen coverage. At temperatures > 150 0C, the disproportionation and consequently the backward dehydroge-nation pathway become more favored, due to the unavailability of H2, probably entrapped in the generated micropores. The hydroconversion becomes structure-sensitive with an apparent dependence on hydrogen pressure at higher temperatures.

The dehydrogenation activity of the studied catalytic composite systems in ethanol conversion decreased at temperatures be-

tween 100 and 250 0C in the following order: (RHS-Ni-A)R350 > (RHS-Ni-PDA)R350 > (RHS-Ni)R350 (Table 3). The prevailing activity of organically modified catalyst samples was encouraged by the improved homogeneous distribution of Ni nanoparticles and the created micropre system. For the catalyst sample (RHS-Ni-PDA)R600, a dramatic decrease in activity was observed, yet with 100% selectivity toward acetaldehyde, being linked with the size enlargement of Ni nanoparticles. The acet-aldehyde production was suggested to be mainly structure-sensitive. The activity results of the exploited (RHS-Ni-PDA)Re sample were comparable with those of the original catalyst reduced at 350 0C at even lower temperatures (100-150 0C), indicating the possibility of reuse of this composite system for several times.

References

[1] A.E. Ahmed, F. Adam, Microporous Mesoporous Mater. 103 (2007) 284-295.

[2] U. Kalapathy, A. Proctor, J. Shultz, Bioresour. Technol. 85 (3) (2002) 285-289.

[3] F. Adam, K. Kandasamy, S. Balakrishnan, J. Colloid Interface Sci. 304 (2006) 137-143.

[4] V.B. Valodkar, G.L. Tembe, M. Ravindranathan, H.S. Rama, J. Mol. Catal. A: Chem. 223 (2004) 31-38.

[5] F. Adam, I. Sugiarmawan, J. Porous Mater. 16 (2009) 321-329.

[6] M. Meilikhov, K. Yusenko, D. Esken, S. Turner, G.V. Tendeloo, R.A. Fischer, Eur. J. Inorg. Chem. 24 (2010) 3701-3714.

[7] F. Adam, J. Andas, J. Colloid Interface Sci. 311 (2007) 135-143.

[8] S.M. Aboul-Fotouh, A. K Aboul-Gheit, Appl. Catal., A 208 (2001) 55-61.

[9] A.K. Aboul-Gheit, S.M. Abdel-Hamid, S.M. Aboul-Fotouh, N.A.K. Aboul- Gheit, J. Chin. Chem. Soc. 53 (2006) 793-802.

[10] N.A.K. Aboul- Gheit, J. Chin. Chem. Soc. 54 (2007) 1211-1222.

[11] A. Fasi, I. Palinko, Catal. Lett. 58 (1999) 103-106.

[12] G.A. Somorjai, R.M. Rioux, Catal. Today 100 (2005) 201.

[13] N. Hebben, C. Diehm, O. Deutschmann, Appl. Catal., A 388 (2010) 225-231.

[14] A. Neramittagapong, W. Attaphaiboon, S. Neramittagapong, Chiang Mai J. Sci. 35 (1) (2008) 171-177.

[15] Ministry of Economy, Trade, and Industry of Japan, Chemical Industry Statics, 2009.

[16] B.D. Cullity, Elements of X-ray Diffraction, Addison-Wesley Publishing Co. Inc., 1976 (Chapter 14).

[17] P.A. Webb, C. Orr, Analytical Methods in Fine Particle Technology, Micromeritics Instrument Corp., Norcross, GA, 1997, p. 57.

[18] Yu Wu, X. Wu, L. Zhang, Chin. Particul. 1 (6) (2003) 262-265.

[19] J.I. Di Cosimo, C.R. Apesteguia, M.J.L. Gines, E. Iglesia, J. Catal. 190 (2000) 261-275.

[20] L. Mahdavian, Nat. Sci. 3 (2011) 471-477.