Catalytic hydrogenation of soybean oil promoted by

Available online at www.derpharmachemica.com

Scholars Research Library Der Pharma Chemica, 2013, 5(2):118-126 (http://derpharmachemica.com/archive.html)

ISSN 0975-413X CODEN (USA): PCHHAX

Catalytic hydrogenation of soybean oil promoted by synthesized Ni/SiO2 nanocatalyst Sohail Ahmeda, Chandrappa K. G.b*, Bakhtiar Muhammada and Jayant Indurkarb# a

Department of Chemistry, Hazara University, Mansehra KPK Pakistan Centre for Research in Nanotechnology and Catalysis (NANOCAT), University of Malaya, Lembah Pantai, Kuala Lumpur, Malaysia

b

_____________________________________________________________________________________________ ABSTRACT In this contribution we are presenting a simple impregnation method to synthesize 25 wt% of nanocrystalline Ni/SiO2 catalyst. The obtained powder was structurally characterized by using powder-XRD, TGA, DSC, SEMEDAX and particle size analyzer (BET) techniques. The synthesized catalyst was utilized for catalytic hydrogenation of soybean oil using Parr reactor and also comparative study of generated catalyst with commercially available Ni catalyst. Results indicate the generated 25 wt% of Ni/SiO2 catalyst exhibited good catalytic activity, thermal stability and selectivity for catalytic hydrogenation of soybean oil as compared to commercially available catalyst by reducing the formation of trans fat by 20-25%. Key words: Hydrogenation, Nanocatalyst, Impregnation, Soybean oil, Trans fat

_____________________________________________________________________________________________ INTRODUCTION Supported metal catalysts are of significant importance in chemical industries for the synthesis of valuable products used in gas storage, purification and separation purposes [1]. Nickel (Ni) based supported catalysts on silica or alumina is usually used in the hydrogenation of oils [2]. Precious metal catalysts based on supported gold have been widely used for low temperature oxidation of CO with the valuable metal deposited as nanocatalyst on metal oxides [3-6]. It is well known fact that, the catalytic performance is affected by changing the charge, size and shape of metal particles [7,8], crystallographic structure [9] and by forming the specific active sites at the metal support boundary [10]. Large amount of work has been reported on elucidating the effects of support on the structure and subsequently the catalytic performance of supported metal catalysts [11-13]. The partial hydrogenation of natural oils to margarines, shortenings, salad oils, toppings and various other edible products are major applications of hydrogenation [14]. The hardening of vegetable oils through hydrogenation of double bonds is a major activity in food industries. The process not only results in a physical change but also it extends the range of usage and gives the stable product against oxidation to rancid smelling derivatives with air. The primary issue in hydrogenation of soybean oil lies in the fact that, the conventional hydrogenation methods produce hydrogenated oils containing 4045 % of trans fats that have unfavorable effect on blood cholesterol level which may increase the risk for coronary heart diseases and have come under intense scrutiny with regard to human health. American people consume average of about 22 g/day of soybean oil and this provides 3.6 g of trans fatty acids. Furthermore, it has become a law in US to suggest consumers to reduce their consumption of trans fats as much as possible [15]. The main focus in the commercial process is to reduce trans fatty acids formed from their natural cis-isomers during conventional

118 www.scholarsresearchlibrary.com

Chandrappa K. G.et al Der Pharma Chemica, 2013, 5 (2):118-126 _____________________________________________________________________________ hydrogenation results in a net hardening of soybean oil. In the finished product it is desired to maximize the amount of oleic acid residues (cis-9-octadecanoic acid) as far as eliminate linolenic acid residues and to reduce the content of linoleic acid residues to a substantial extent without going too far towards producing the fully saturated stearic acid chains, since these are not easily digested as foodstuffs [16]. Food industries are facing several major challenges that have been emphasized to improve the characteristics of supported metal catalysts, but the majority of researchers suffer a lot of disadvantages in terms of decrease in surface area, difficulty of separating fine particles due to particles agglomeration. So, the purpose of our study is to develop a novel catalyst which should be useful in enhancing the catalytic activity and to achieve maximum yield for industrial applications [17,18]. In the present study, we mainly focus on the improvement of particle size and morphology of synthesized nickel-silica nanocatalyst. In addition, the modified catalyst with improved properties is used in hydrogenation of soybean oil. During hydrogenation process the naturally occurring cis-unsaturated fatty acids are partly converted to un-natural trans isomers. MATERIALS AND METHODS Starting Materials and Synthesis of Ni/SiO2 catalyst Nickel nitrate hexahydrate (AR grade; 99 %) purchased from Acrose chemicals, ethylene glycol (AR grade; 99.5%) and sodium hydroxide (AR grade; 99 %) were purchased from Riedel de-Haen chemicals, hydrazinium hydroxide (AR grade; 80%), 1-propanol and ethanol were purchased from Merck chemicals and used without further purification. Double distilled water was used to prepare NaOH solution. In a typical synthesis, about 4.94 gm of Ni(NO3)2. 6H2O and 3.75 gm of silica (SiO2) powder was dissolved in 200 ml of ethylene glycol in round bottom flask. The resultant solution was then stirred continuously on hot plate stirrer at a temperature of about 90 °C for 30 min. To this add 3 ml of hydrazinium hydroxide as a reducing agent in the solution at a rate of 0.1 ml/min with HPLC pump followed by the drop wise addition of 1M NaOH solution for maintaining the pH from 10.2 to 10.5 at a temperature of 60 °C on hot plate with continues stirring for 3 hrs to obtain a homogeneous slurry. The resultant slurry was evaporated and washed three times to remove the impurities and dried at 80 °C overnight. The dried compound was grind to a homogeneous powder and then calcined at 400 °C for 6 hrs. Catalyst characterization General morphology, structure, crystallite size and compositional analysis of synthesized nanocatalysts were performed using powder X-ray diffraction (XRD), Thermal analysis (TGA-DSC), Scanning electron microscopy (SEM), energy dispersive X-ray analyzer (EDAX) and BET surface area. X-ray diffraction analysis was carried out by Roentgen Diffractometer System (Model; PW 3040/60 X-Pert PRO) using nickel- filtered Cu Kα radiation source operated at 40 KV and 100 mA. Thermogravimetric analyzer (TGA) and differential scanning calorimeter (DSC) were performed in the temperature range 50–600 °C at a heating rate of 5 °C/min under nitrogen atmosphere using a Model Q 50 for TGA and Q 100 for DSC by TA Instruments USA. Morphology and compositional analysis were carried out in a scanning electron microscope (SEM, Philips XL 30, Model No. 5432) fitted with an energy dispersive X-ray analyzer (EDAX). Average particle size and particle size distribution was determined using Malvern Laser particle size analyzer Model No. 2000 Analyzer. The hydrogenation of soybean oil was carried out in a batch slurry reactor system (Parr Instruments Inc. USA, Model No.4841). Catalytic hydrogenation of soybean oil The as-synthesized Ni/SiO2 (25wt%) nanocatalyst was subjected to catalytic hydrogenation reaction using par reactor system at atmospheric pressure. Figure 1 shows the diagram of par reactor system. About, 200 ml of soybean oil and 1.0 gm 25 wt% of Ni/SiO2 catalyst were loaded in to the reactor. Prior to the experiment, Argon gas was purged to remove any unwanted air initially present in the reactor and also to supply an inert atmosphere in to the reactor system. The inlet and outlet of the reactor were then closed. The reaction contents were heated from preselected temperature of 100 °C to 110 °C and this range of temperature was selected for two reasons. Firstly, the reaction rate was sufficiently fast to allow the experiments to be completed within 20 min. Secondly, the pressure developed within this range could be reproducibly determined with good accuracy using the present experimental equipment. The temperature range of 100 °C to 120 °C was then selected for more detailed investigations, because the significant amount of reaction product was produced in this range. The reaction mixture was stirred at 750 rpm for 30 min. The selectivity and conversion were calculated from GC/TCD analytical results. After the reaction, the obtained product was filtered and the catalyst was recovered, dried and it could be reused for further reaction.


Chandrappa K. G.et al Der Pharma Chemica, 2013, 5 (2):118-126 _____________________________________________________________________________

Fig 1. Parr reactor for hydrogenation reaction.

RESULTS AND DISCUSSION X-ray diffraction studies The phase purity and crystalline structure of all synthesized catalyst were investigated by using powder X-ray diffraction technique. The powder XRD patterns of as-prepared Ni/SiO2 (25 wt%) catalyst are shown in Fig. 2. It can be seen from Fig. 2. that, the three main peaks appeared at 2θ range of 44.5°, 51.9° and 76.3° shows highly crystalline characteristic peaks of nickel nanoparticles corresponding to Miller indices (111), (200) and (220) in the XRD spectrum. This result confirms that, the resultant nickel nanoparticles are composed of pure face-centered cubic (fcc) with particle size of 9.2 nm calculated from Debye Scherrer equation. The phase analysis shows that, there is no signal to be assigned for silica (SiO2) indicating that the SiO2 is in amorphous state. Thermal studies The thermal behavior of synthesized catalyst was studied with TGA-DSC measurements. The catalyst was heated in nitrogen atmosphere on platinum pan at the temperature scanning rate of 5°/min. Figure 3 shows the TGA and DSC plot of synthesized Ni/SiO2 catalyst. It can be seen from Fig. 3(a). that, the prepared Ni/SiO2 catalyst shows the smallest amount of mass loss of about 7 % in the temperature range of 280 °C and 500°C, revealing the dehydration process of surface adsorbed water molecules. Figure 3(b) shows the DSC curve of synthesized catalyst, it evaluate that up to 275 °C the heat flow of the sample exhibit constant behavior along with temperature. After that, the melting was observed at 425 °C. This result shows that, indigenously prepared Ni/SiO2 nanocatalyst has higher thermal stability for catalytic application of soybean oil.


Chandrappa K. G.et al Der Pharma Chemica, 2013, 5 (2):118-126 _____________________________________________________________________________

Fig 2. Powder X-ray diffraction pattern for 25 wt% Ni/SiO2 catalyst (uncalcined).

Fig 3. TGA-DSC analysis of 25 wt% Ni/SiO2 catalyst, (a) TGA-DSC curves and (b) DSC curve (Dual run).

SEM/EDAX analysis The SEM photomicrographs of synthesized Ni/SiO2 (25 wt%) catalyst has shown in Figs. 4(a) and 4(b). In case of 25 wt % nickel nanoparticles supported on silica surface, the crystallite size was found to be 50 nm and it can be clearly seen from SEM micrographs (Fig. 4(a)). It indicates that, the formation of irregular spongy shaped surface where the nickel nanoparticles are well dispersed within the pores of silica support and does not form aggregates (Fig. 4(b)). Hence, by providing large surface area for fat molecules to reach the center of metal particles and ultimately decrease the diffusivity time. Moreover, by controlling the texture and uniform distribution of these metal particles results in the improvement of accessibility of triglycerides molecules to the center of metal particles. The surface composition and elemental analysis of synthesized catalyst is shown in Fig. 4(c), which indicates the presence of only nickel, silicon, oxygen and small amount of carbon components with their appropriate proportions. The actual metal loading during the catalyst preparation is also shown in the Table 1. It was found that the nickel concentration is about 25.92 % in 25 wt% Ni/SiO2 catalyst. This increases the percentage of nickel within the specified composition and it is due to the calcinations of catalyst at 400 °C. The high surface concentration of 25 wt% Ni/SiO2 might be attributed to the dispersion of nickel on high surface silica support and very small particle


Chandrappa K. G.et al Der Pharma Chemica, 2013, 5 (2):118-126 _____________________________________________________________________________ size (9.2 nm) of nickel supported on silica. This indicates that, the reduced nanoparticles played a very important role in promoting the nickel dispersion uniformly on the surface of the catalyst.

Fig 4. Scanning electron photomicrographs of 25 wt% Ni/SiO2 catalyst (a) as-prepared (b) calcined at 400 °C for 6 hr and (c) EDAX spectrum for 25 wt% Ni/SiO2 catalyst. Table 1 EDAX analysis (elemental composition) of 25 wt% Ni/SiO2 catalyst. Catalyst 25 wt% Ni/SiO2

Components Oxygen Silicon Nickel

Wt % 47.47 26.61 25.92

Atomic % 60.11 21.75 18.14

Actual % ----25.00

BET measurement BET analysis shows the surface area and pore volume for silica supported nickel nanocatalyst. Table 2 shows the BET measurement and pore volume of both 25 wt% of Ni/SiO2 and commercial Ni catalyst. The surface area was decreased up to a value of 157.69 m2g-1 in case of 25 wt% Ni/SiO2 modified catalyst due to uniform dispersion of catalytically active nickel nanoparticles within the pores of silica support followed by increase in pore volume up to 6.58 cm3g-1 and thus provides a relatively accessible path for H2 and triglycerides molecules to diffuse spontaneously from the outer surface of the catalyst powder to the mesopores and finally to the metal crystallites surface forming chemisorbed H atoms compared to use of unmodified catalyst.


Chandrappa K. G.et al Der Pharma Chemica, 2013, 5 (2):118-126 _____________________________________________________________________________ Table 2 BET measurement and pore volume of both 25 wt% of Ni/SiO2 and commercial Ni catalyst. Sample 25 wt% Ni/SiO2 catalyst 25 wt% commercial Ni catalyst

BET Surface area (m2g-1) 157.69 4.06

Pore Volume (cm3/gm) 6.58 0.65

Particle size distribution (nm) 42.89 17.92

Particle size distribution The calculation of particle size distribution for 25 wt% Ni/SiO2 catalyst shows that 50 % of the particles are in the range of 42.89 nm and the distribution of particles from 41.43 to 48.27 nm is 54.07 % which shows a good agreement with BET results, thus providing large catalytic active surface area for the reactants, where as in case of 25 wt% of commercial Ni catalyst the particle size distribution is 16.57 to 19.31nm with 4.68 % in sample. It shows that the particle size of 25 wt% commercial Ni catalyst is 17.92 nm as compared to 25 wt% Ni/SiO2 catalyst where the particle size is 42.89 nm. Catalytic performance of synthesized Ni/SiO2 (25 wt%) nanocatalyst Modification of 25 wt% Ni/SiO2 catalyst Surface area and porosity is a very important characteristic of solid materials that determines the properties and performance of catalyst. The catalyst supports or carriers are greatly influenced the activity and selectivity of synthesized catalyst. The method for improving the characteristics of catalytic materials like particle size, surface area, pore volume and particle size distribution is concerned. In this study, we have tried to enhance the catalytic performance of nickel nanoparticles by dispersing on high surface area, high purity amorphous silica by directly adding to the nickel nitrate solution by keeping the desire amount at 25 wt% prepared from AR grade Ni(NO3)2. 6H2O was dissolved in 200 ml ethylene glycol where the nickel was reduced to 9.2 nm by hydrazinium hydroxide. Although different methods have been followed show an increase in surface area with corresponding decrease in particle size, but our method is much more effective in terms of major increase in surface area and pore volume with decrease in particle size as compared to the other methods. The increase in surface area with controlled particle size is might be due to ethylene glycol used as a solvent and it forms a protective layer around the particle surface during reduction process and also preventing the particles agglomeration. This causes uniform distribution of nickel nanoparticles on silica support. Selective hydrogenation of soybean oil The catalytic properties of 25 wt% Ni/SiO2 catalyst were evaluated for the selective hydrogenation of soybean oil. The hydrogenation of soybean oil was carried out in a batch slurry reactor system shown is Fig. 1. Selective hydrogenation process is often carried out in stirred autoclaves, since most hydrogenation reactions are exothermic and the outlet temperature is commonly about 200 °C. So, the care must be taken to control the temperature of autoclaves and is required to achieve the desired selectivity and prevent temperature runway. The reactor was a high-pressure (SS-316 reactor) with a volume of 320 ml. A J-typed thermocouple was immersed into the liquid phase to measure the temperature and was connected to a heater/controller to maintain the reactor temperature within ±1 k. Liquid sample were collected using a stainless steel tube (outer diameter: 1/8 inch) connected to a sampling valve. A pressure transducer (Foxboro electronic transmitter, Model 84 GM-D) was used to measure the pressure of reactor with an accuracy of ±4.0 kPa. Using Parr reactor system (Fig. 1), the following operation conditions were set for the hydrogenation of soybean oil. About 1 gm of catalyst and 200 ml of soybean oil were loading in a reactor under argon atmosphere to remove unwanted air. Hydrogen flow rate should be 50 ml/min and heating rate was 100 °C/min. Finally, the obtained product stream was analyzed by using GC/TCD measurements. Hydrogenation mechanism Nickel catalyst is usually used in the hydrogenation of edible oils. The soybean oil consists of unsaturated 18-carbon long chain fatty acids. The partial hydrogenation reduces the triene linolenic acid (cis-9, cis-12, cis-15octadecatrienoic acid) and diene linoleic acid (cis-9, cis-12-octadecadienoic acid) contents of the oil. The sequence illustrates the complexity of the catalytic surface reactions operating in the conversion of soybean oil to more valuable product. First soybean oil is chemisorbed onto a nickel supported catalyst (active site) rapidly. All chemisorption reactions are exothermic that is, the chemisorption occurs spontaneously and with the increase in ordering of the phase molecules as they adsorb. Similarly, diatomic H2 (g) chemisorbs dissociatively onto nickel surface through macro pores network of silica surface provides a relatively accessible path for H2 to diffuse from the outer surface of the catalyst powder to the mesopores and finally to the metal crystallites surface again rapidly,


Chandrappa K. G.et al Der Pharma Chemica, 2013, 5 (2):118-126 _____________________________________________________________________________ reversibly on to nickel surface sites forming chemisorbed hydrogen atoms. Chemisorbed soybean oil and hydrogen [Oil-catalyst complex and H (ads)] reacts relatively slow and essentially irreversible forming chemisorbed hydrogenated oil plus catalyst which subsequently desorbs rapidly. However, the reaction can be broken down into a number of simple or elementary steps, each of which occurs as written at the molecular level and which cannot be further simplified. The following is one of several possible mechanistic sequences, H2 (g) + catalyst → 2H (ads) R – CH = CH – R(1) → R – CH = CH – R (ads) H (ads) + R – CH = CH – R → R – CH2 – CH – R (ads) H (ads) + R –CH2 –CH – R (ads) → R – CH2 –CH2 – R(1)

(1) (2) (3) (4)

The last step is rate-determining step, Oil + catalyst → Oil-catalyst complex Oil-catalyst complex + H → hydrogenated oil + catalyst The experimental studies shows that the slow rate-determining step is the reaction between adsorbed soybean oil and hydrogen, and thus this reaction controls the overall reaction rate. In the absence of a nickel catalyst, H2 decomposition is the rate-limiting step, and the thermal reaction becomes important at about 200 °C. The presence of nickel catalyst, however, decreases the activation energy for H2 dissociation reaction allowing partial reduction to occur at 120 °C. Investigation of catalyst activity Gas chromatography and FT-IR techniques are commonly used for the determination of trans fatty acids contents in the edible oils and fats [19-21]. FT-IR spectroscopy provides rapid and reproducible measurements for the trans fat containing more than 5% therefore gas chromatography is desired to measure low trans fat level. The GC method offer both identification and quantification of individual fatty acids, but the accuracy and reliability of the method depends on the resolution and identification of all trans fatty acids [22]. The catalytic activity of the catalysts was analyzed using GC/TCD analysis. The results obtained from GC/TCD chromatogram data under different reaction conditions are given in the Table 3. Table 3 Shows trans fats reduction obtained under the different conditions. Catalyst designation 25 wt% Ni/SiO2 25 wt% commercial Ni

Reaction Temp (°C)

Reaction time (min)

Stirrer speed (RPM)

H2-flow mil/min

Reaction completion time (min)

Degree of SBO reduction (%)

Trans fats (%)

120

20

750

30

15

55-58

13.8

160

60

750

30

40

45-48

21.6

The activity and selectivity trends of the catalyst depend strongly on specific surface area and pore size. In case of 25 wt% Ni/SiO2 system, the surface area is 157.69 m2g-1 with average pore volume 6.58 cm3/gm and hydrogenation was carried out at 120 °C and 1 atm pressure for 20 min. Diatomic H2 (g) chemisorbs dissociatively onto nickel surface through macro pores network of silica surface provides a relatively accessible path for H2 to chemisorbs without much hindrance favors early low temperature 120 °C and efficient hydrogenation that is 55 to 58 % of soybean oil to soybean oil product in first 15 min. The increased activity observed over 25 wt% Ni/SiO2 catalyst might be due to following reasons. (i) High surface area and porous structure provides large accessibility to reacting molecules to active catalytic sites which are dispersed throughout the internal pore structure of the catalyst. In case of 25 wt% Ni/SiO2 system, as the surface area is decreased up to a value of 157.69 m2g-1 possessing mean pore volume up to is 6.58 cm3/gm as noted from characterization studies, thus exposing a large fraction of the supported material to the reacting molecules. (ii) The number of reactant molecules converting to product in a given time interval is directly related to number of active sites available. It is therefore to maximize the number of active sites available to the reactant molecules were achieved by two ways. (a) In order to increase the surface area per gram of nickel the mean particle size was reduced up to 9.2 nm. (b) Also increase the weight percentage of nickel metal up to 25 % by dispersing on to a silica support to acquire the desired results. On the other hand, in case of 25 wt% commercial Ni catalyst (Table 2), the surface area is 4.06 m2g-1 with average pore volume 0.65 cm3/gm operates at commercial reaction conditions that is higher catalyst concentration and reaction temperature (above 170 °C) may


Chandrappa K. G.et al Der Pharma Chemica, 2013, 5 (2):118-126 _____________________________________________________________________________ attributed to the poor reactivity of 25 wt% commercial Ni catalyst from the small surface area and pore size. The activity trends of both catalysts are shown in Fig. 5(a). Another issue of major importance is the catalyst ability to effect cis/trans isomerization. During conventional hydrogenation, nickel catalyst isomerizes the natural cis double bond of triglyceride molecules to unnatural trans double bond when triglycerides molecules desorbs from the surface of catalyst without being hydrogenated. Indigenously prepared 25 wt% Ni/SiO2 catalyst operates at optimal reaction conditions that is 120 °C and at normal pressure could reduce trans fats from 20–25 % on complete reduction of soybean oil as shown in Fig. 5(b). Typical 25 wt% commercial Ni catalyst operates under commercial reaction conditions that is high reaction temperature and catalyst concentration is quite encourage at converting Cis double bond to trans double bond up to 45 % on complete reduction as shown in the Fig. 5(b). Take too long diffusion time to reach the nickel surface due to small surface area and pore size provides less accessibility to triglycerides molecules which ultimately desorbs without being hydrogenation and trans isomers and it increased significantly during the hydrogenation of soybean oil.

Fig 5. Catalytic hydrogenation of soybean oil for 25 wt% Ni/SiO2 with commercial Ni catalysts (a) activity vs. temperature and (b) effect of catalyst vs. temperature on trans isomers.


Chandrappa K. G.et al Der Pharma Chemica, 2013, 5 (2):118-126 _____________________________________________________________________________ CONCLUSION Impregnation method has been proved to be a useful method for increasing the surface area and pore volume supports by decrease in mean particle size, uniform size distribution and calcination temperature. The synthesized nanocatalyst has strong metal-support due to the formation of well dispersed nickel nanoparticles on silica support. The conventional hydrogenation reaction is carried out at 200-250 °C and 5 atm pressure for 2-4 hr that depends on the amount of catalyst and batch preparation. This type of reaction do not proceed well and thus forming 25 - 45 % trans fatty acids whereas in 25 wt% Ni/SiO2 catalyst is much more effective in hydrogenation of soybean oil with less time and operates at low temperature. The catalytic hydrogenation reaction provides more stable with low trans fatty acid contents and thus giving 23 % of trans fat concentration compared to commercial catalyst. Thus, the synthesized 25 wt% Ni/SiO2 nanocatalyst could be economically beneficial for catalytic applications than commercial Ni catalyst. Further, the studies are essential to investigate the viability of this catalyst on commercial scale for selective hydrogenation of soybean oil. Acknowledgment The authors thank to Department of Chemistry, Hazara University, Mansehra, Pakistan and NANOCAT research center, University of Malaya, Malaysia for providing the lab facilities to bring about this work. REFERENCES [1] M. Gabrovska, J. Krstic, R. Edreva-Kardjieva, M. Stankovic and D. Jovanovic, Appl. Catal. A: General, 2006, 299, 73–83. [2] M.M. Selim and Islam Hamdy Abd El-Maksoud, Micropor. Mesopor. Mater 2005, 85, 273–278. [3] H. L. Zhang, L. H. Ren, A. H. Lu and W. C. Li, Chin J Catal, 2012, 33, 1125. [4] M. R. Benjaram, Gode Thrimurthulu and Katta Lakshmi. Chin J Catal, 2011, 32, 800. [5] Q. L. Li, Y. H. Zhang, G. X. Chen, J. Q. Fan, H. Q. Lan and Y. Q. Yang. J Catal, 2010, 273, 167. [6] I. Dobrosz-Gomez, I. Kocemba and J. M. Rynkowski. Appl Catal B, 2008, 83, 240. [7] T.J. McCarthy, C.M.P. Marques, H. Trevino and W.M.H. Sachtler, Catal. Lett, 1997, 43, 11- 18. [8] L. Guczi, Z.Konya, Zs. Koppany, G. Stefler and I. Kiricsi, Catal. Lett, 1997, 44, 7-10. [9] M. Vaarkamp, J.T. Miller, F.S. Modica and D.C. Koningsberger, J. Catal, 1996, 163, 294- 305. [10] P. V. Menacherry, M. Fernandez-Garcia and G.L. Haller, J. Catal, 1997, 166, 75–88. [11] M. Che and C.O. Bennett, Adv. Catal, 1989, 36, 55-172. [12] J. P. Boitiaux, Rev. Inst. Franc. Petrol, 1993, 48, 527-531. [13] R. Burch and Z. Paal, Appl. Catal, 1994, 114, 9-33. [14] P. N. Rylander, J. R. Anderson and M. Boudart, Catalytic Processes in Organic Conversions, Catalysis: Science and Technology, Springer-Verlag, Berlin, 1983. [15] M.A. DiRienzo, J.D. Astwood, B.J. Petersen and K.M. Smith, Lipids, 2006, 41(2), 149-157. [16] B. Nohair, C. Especel, G. Lafaye, P. Marecota, LeChien Hoang and J. Barbier, J. Mol. Catal. A: Chem, 2005, 229, 117–126. [17] A. Le Bail and D. Louer, J. Appl. Crystal, 1978, 11(1), 50–55. [18] S.R. Sashital, J.B. Cohen, R.L. Burwell Jr. and J.B. Butt, J. Catal, 1977, 50(3), 479–493. [19] F. Priego-Capote, J. Ruiz-Jimenez and M.D. Luque de Castro, Food Chemistry, 2007, 100, 859–867. [20] A. Baylin, X. Siles, A. Donovan-Palmer, X. Fernandez and H. Campos, J. Food Compos. Anal, 2007, 20, 182192 [21] Y. Ismail and G. Umit, Food Control, 2007, 18, 635. [22] P. Delmonte and J. I. Rader, Analytical and Bioanalytical Chemistry, 2007, 389(1), 77 – 85.
