Zeolitized-pumice as a new support for hydrogenation catalysts

Catalysis Communications 9 (2008) 2085–2089

Contents lists available at ScienceDirect

Catalysis Communications journal homepage: www.elsevier.com/locate/catcom

Zeolitized-pumice as a new support for hydrogenation catalysts G. Neri a,*, G. Rizzo a, L. De Luca a, F. Corigliano a, I. Arrigo a, A. Donato b a b

Department of Industrial Chemistry and Materials Engineering, University of Messina, Salita Sperone 31, 98166 Messina, Italy Department of Mechanics and Materials, University of Reggio Calabria, Loc. Feo di Vito, 89100 Reggio Calabria, Italy

a r t i c l e

i n f o

Article history: Received 24 January 2008 Received in revised form 21 March 2008 Accepted 26 March 2008 Available online 3 April 2008 Keywords: Pumice Catalyst support Campholenic aldehyde Naturanol Selective hydrogenation

a b s t r a c t A zeolitized-pumice (Z-PM), obtained from a pumice mine waste by extraction of most of the silica contained, has been proposed as a new catalyst support. The as prepared Z-PM material was characterized by surface area and porosity measurements, XRD, SEM and TEM-EDX, showing a high content of crystalline zeolite Pc (about 60%) and a higher (33 m2/g) surface area compared to starting pumice. Pt and Pt–Sn supported on Z-PM were prepared by impregnation and their catalytic properties in the selective hydrogenation of campholenic aldehyde to naturanol investigated in detail. Results highlighted the better performance of Pt and Pt–Sn/Z-PM catalysts in the selective reduction of –C@O group in comparison to corresponding silica-supported catalysts. Ó 2008 Elsevier B.V. All rights reserved.

1. Introduction Pumice is a porous, natural glass formed from volcanic activity. It is largely used in place of sand in light concrete, as a filtering agent for water clarifying, as a softly abrasive material in the textile industry (jeans stone-washing) and in a number of other industrial applications [1,2]. Pumice has also been used as a support for metal catalysts [3–9]. Among various reactions, the hydrogenation of the unsaturated carbon-carbon bonds was the most investigated on pumice supported catalysts. Deganello and coworkers investigated in details the preparation, characterization and behaviour of these catalysts. They exhibited good activity and selectivity in the liquid phase hydrogenation of 1,3-cyclooctadiene [5,6] and phenylacetylene [7,8] and in the gas phase hydrogenation of acetylene in ethylene rich feedstocks [9]. At our knowledge, no pumice supported catalysts were investigated in the selective hydrogenation of unsaturated aldehydes, despite of the significant attention received by the synthesis of unsaturated alcohols from these substrates in the last years, owing to the importance of these compounds in the cosmetic, food and flavors industry [10–13]. The untreated pumice materials so far used as catalyst support have the disadvantage of a low surface area [3–9]. It is well known that a wide surface area is an important requisite for this application. Therefore, we are here proposing as catalyst support in place

* Corresponding author. Address: Dipartimento di Chimica Industriale e Ingegneria dei Materiali, Universita di Messina, Contrada di Dio, 98166 Messina, Italy. Tel.: +39 0903977297; fax: +39 0903977464. E-mail address: [email protected] (G. Neri). 1566-7367/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.catcom.2008.03.044

of untreated pumice a new pumice material (Z-PM) where a suitable treatment caused a plentiful crystallization to zeolite Pc and an increase of an order of magnitude of the surface area. The synthesis of Z-PM starting from pumice mine waste as raw material was described in a recent patent [14]. The resulting material has a surface area of about 30–40 m2/g and a high content (around 60%) of crystalline zeolite Pc which led to the development of some applications typical of zeolites [15–16] . In this paper, the use in catalysis – that is a further application field typical of zeolite [17] – is proposed. Zeolite catalysts have found extensive use in the petrochemical industry and show potential applications in fine chemical synthesis [18]. Nevertheless, the catalytic hydrogenation of unsaturated aldehydes was less investigated with zeolite materials and consequently the role of these supports in the promotion mechanism is still obscure. Blackmond et al. [19] reported the hydrogenation of 3-methyl crotonaldehyde over Pt, Rh or Ru metals supported on NaY and KY zeolites. They found that the selectivity to the unsaturated alcohol increased when support was less acidic due to a greater electron density on the metal particles. Other authors pointed out instead the benefit of geometrical effects of zeolite pores [20] or the peculiar catalytic properties of the smaller metallic particles located inside the zeolite cavities [21]. In this study, we report results obtained in the selective hydrogenation of campholenic aldehyde (CPA) to naturanol (NAT), a valuable fine chemical largely used in food and perfumery industry due to its particularly well defined sweet, natural and specific berry-like notes [22]. The reaction was carried out on either Pt or bimetallic Pt–Sn catalysts, both supported on Z-PM. A comparison with the corresponding Pt and Pt–Sn catalysts supported on silica is also reported.

2086

G. Neri et al. / Catalysis Communications 9 (2008) 2085–2089

2. Experimental 2.1. Z-PM preparation and characterization The synthesis of Z-PM was carried out by extracting from pumice mine waste as much silica as possible by a sodium alkali aqueous solution in hydrothermal conditions and by rinsing the unextracted residue with water up to neutral pH, according to a previous report [14]. BET surface area and porosity data were determined by adsorption–desorption of dinitrogen at 77 K using a Micromeritics ASAP 2010 instrument. The chemical composition was determined by an energy dispersive spectroscopy (EDS) microprobe (Oxford INCA ENERGY-400 with PENTAFET Si(Li) detector). X-ray diffraction (XRD) spectra were recorded on an Ital Structure mod. APD 2000 X-ray diffractometer using the Cu Ka radiation and mounting powder samples on a Plexiglass holder. Scanning electron microscopy (SEM) analyses were carried out with a JEOL 5600LV electron microscope. The samples were coated with a thin layer of gold to prevent charging of the samples. Micrographs of the samples were recorded in the 10–20 kV accelerating voltage. Transmission electron microscopy (TEM) studies of the catalysts were performed on a JEOL 2010 EX instrument operating at 200 kV and equipped with an EDX analyzer (Oxford). The specimens for electron microscopy were prepared by grinding the powder samples in an agate mortar, suspending and sonicating them in isopropanol, and placing a drop of the suspension on a holey coated carbon copper grid. After evaporation of the solvent, the specimens were introduced into the microscope column. 2.2. Catalysts preparation Pt and Pt–Sn/Z-PM catalysts were prepared by (co)impregnation by using the incipient wetness technique of the Z-PM support with H2PtCl6 (hydrogen hexachloroplatinate(IV) hydrate 99.9%, Fluka) and SnCl2 (Tin (II) chloride 99.99%, Aldrich) aqueous solutions having the appropriate metal concentration. Pt and Pt–Sn/SiO2 were also prepared in the same way, by using a commercial Silica support (SiO2 Grace, BET surface area 276 m2/g). After impregnation, the catalysts were dried at 393 K in air for 1 h followed by reduction under flowing H2 at 623 K for 2 h. On all catalysts Pt loading was 2 wt% and on bimetallic samples Sn was 0.48 wt%, corresponding to a Sn/Pt atomic ratio equal to 0.39. 2.3. Kinetic tests Catalytic activity tests were carried out in a 100 ml five-necked flask, equipped with a reflux condenser. Constant temperature (343 ± 0.5 K) was maintained by circulation of silicone oil in an external jacket connected with a thermostat. The reactions were carried out as follows. The catalyst (250 mg) was added to 50 ml of cyclohexane through one arm of the flask. Before introduction of the substrate (CPA), the catalyst was reduced ‘‘in situ” by a H2 flow (50 ml/min) for 30 min. at 343 K. Then 0.1 ml of the substrate and 0.1 ml of tetradecane were introduced (the latter was used as an internal standard for gas chromatographic analysis) in the reactor. The reaction mixture was stirred with a stirrer head at 600 rpm with permanent magnetic coupling which ensure a very efficient stirring under H2 at atmospheric pressure. The progress of the reactions was followed analyzing by GC samples withdrawn from the reaction mixture at different times. Details are reported in a previous work [23].

In order to check the absence of diffusional limitations, preliminary runs have been carried out at different stirring conditions, loading and catalyst grain size. The dependence of hydrogenation rate upon the stirring speed, revealed that above 400 rpm the rate is independent on the stirring speed. The trend of CPA conversion at different amounts of catalyst was linear, confirming the absence of external mass transfer limitations. Experimental results have shown that the activity remained nearly constant when the grain size of the support was changed, indicating that the inner diffusion limitation was negligible.

3. Results and discussion 3.1. Z-PM preparation and characterization Several previous studies report pumice powder as a silica source in the synthesis of zeolites by hydrothermal treatment in concentrated sodium hydroxide [24–26]. The aluminosilicate phase produced is dependent on the synthesis conditions, in particular temperature, NaOH concentration and Si/Al ratio [24,25,27]. The synthesis of Z-PM was carried out as reported above and described in detail in a recent patent [14]. The composition of pumice mine waste we used as starting material is shown in Table 1. Its Si/Al atomic ratio was 5.6–5.7. Chemical analysis of the product obtained after the silica extraction process is also reported in Table 1. Compared to pumice mine waste the resulting material contains much less silica, consequently the Si/Al atomic ratio is strongly reduced (about 2.2). On the other hand, Al2O3 and Na2O were significantly found in higher concentration in the lightened residue and both involved in the formation of zeolite. Accordingly to data reported in Table 1, the material obtained has a surface area of about 34 m2/g, much higher with respect to the raw material (1 m2/g). A certain degree of microporosity (micropore area 3.6 m2/g), absent in the starting material, was also developed in connection with zeolite formation. Morphological and microstructural modifications of the raw material after the silica extraction process are reported in Figs. 1–3. The starting pumice mine waste material is essentially amorphous as confirmed by its diffraction pattern reported in Fig. 1. The product obtained is instead highly crystalline, an approximate quantification of which gives a value around 60%. A comparison of the diffraction peaks with reference compounds indicates that reflections observed are related to the presence of zeolite Pc. Zeolite P indicates a series of synthetic zeolite phases, of which the more common are the cubic, Pc, and the tetragonal zeolite, Pt [28], having a Si:Al ratio from 1.0 to 2.5 [29]. The morphology of the raw material and product obtained was examined by scanning electron microscopy. SEM analysis of pumice mine waste (Fig. 2a) shows irregular particles in a wide range of size. After processing, the material was deeply re-structured and a remarkable morphology change and reduction of the particle size was observed (Fig. 2b). TEM micrographs of the Z-PM material taken at different magnification are also shown in Fig. 3a b. The aggregates observed at any magnification level are constituted by particles mainly homogeneous regarding both the size and, as evinced by EDX analysis, spatial distribution of the main components and impurities. Pt and Pt–Sn catalysts supported on Z-PM have been then prepared by the impregnation technique as reported in the experimental section. A detailed characterization of these catalysts will be reported in a forthcoming paper. 3.2. Catalytic activity The behaviour of Pt-based catalysts supported on Z-PM was investigated in the selective hydrogenation of CPA. This substrate

2087

G. Neri et al. / Catalysis Communications 9 (2008) 2085–2089 Table 1 Chemical composition, surface and micropore area of the mine waste pumice (PM) and the product obtained after the silica extraction process (Z-PM) Sample

Composition (wt%) SiO2

Al2O3

FeO

CaO

MgO

Na2O

K2O

PM Z-PM

70.90 50.68

12.76 25.65

0.64 4.36

1.36 2.11

0.60 0.90

3.23 15.35

3.83 1.85

P * zeolite SiO

Intensity (a. u.)

° *

Micropore area (m2/g)

1.2 34.4

<0.05 3.6

PM Z-PM

*

2

SA (m2/g)

*

* *

*

*

° 10

15

20

25

30

35

40

45

50

2ϑ Fig. 1. XRD patterns of the pumice mine waste (PM) and zeolitized material (Z-PM).

Fig. 3. TEM image of the Z-PM material. (a) Low magnification and (b) high magnification.

Fig. 2. SEM image of: (a) pumice mine waste (PM) and (b) Z-PM.

is characterized by the presence of two isolated reducible groups, a C@C double bond and a CH@O group (see Scheme 1). The hydroge-

nation reaction on silica supported platinum catalysts was described in detail elsewhere [23]. The hydrogenation of the double bond, affording the saturated campholenic aldehyde, ACS, occurred on the monometallic Pt catalyst. However, in the presence of suitable promoters (i.e. Sn or Fe) the reaction can be addressed towards the selective hydrogenation of the CH@O group to give the unsaturated campholenic alcohol. These intermediates can be further hydrogenated to give the saturated campholenic alcohol, NATS.

G. Neri et al. / Catalysis Communications 9 (2008) 2085–2089

3.2.1. Monometallic Pt catalysts The results of the hydrogenation of CPA under mild reducing conditions (T = 343 K, H2, 1 atm of partial pressure) on the monometallic Pt/Z-PM and Pt/SiO2 catalysts, are summarized in Table 2. According to previous reports, the reaction on the Pt/SiO2 catalyst involved the reduction of the olefinic double bond and led mainly to the formation of ACS. Naturanol was detected only in small amount (<2%), whereas the formation of the final product of hydrogenation, NATS, was observed already at low substrate conversion. This is however not desired because the fragrance of the saturated products, ACS and NATS, is somewhat weaker in strength [22]. The products distribution in the hydrogenation of campholenic aldehyde changed strongly when platinum supported on Z-PM was used. The products distribution was largely shifted towards the formation of naturanol with a very high selectivity (>90% at 90% conversion). Moreover, as a consequence of the strong poisoning of the C@C double bond, the further hydrogenation of naturanol into the saturated final product was much lower. It is also noteworthy that the catalytic activity of the Pt/Z-PM catalyst is only three times lower compared to the corresponding silica supported catalyst, despite of the larger surface area of the latter catalytic system. Several hypotheses can be formulated to explain the results obtained. First, it should be considered that the microstructure of the Z-PM support can influence the product distribution in the hydrogenation of unsaturated aldehyde. Gallezot et al. showed that the microporous structure of zeolite Y imposed molecular constraints on the mobility of cinnamaldehyde within pores containing metal clusters and favoured adsorption and reaction at the C@O bond [20], suggesting that the geometric effect of the zeolite pores predominate over the electronic ones. Blackmond and coworkers found that the selectivity to unsaturated alcohols increased with the support basicity due to the enhancement of the electron density on the metal particles [19]. Considering the high basicity of zeolite P, this should results in greater electron density on platinum, decreasing then the adsorption of double bond onto the metal surface and favouring correspondingly the C@O hydrogenation. However, it cannot be excluded that the impurities in the Z-PM support play a key role in the hydrogenation reaction. For simplicity we can group them into two categories: (1) the alkaline and alkaline-earth oxides (Na2O, K2O, MgO, CaO) and (2) iron oxides (FeO and Fe2O3). Deganello and co-workers reported, for Pd/pum-

ice catalysts, that the presence of sodium ions in the pumice structure increase the electron density of the supported metal, as evinced in the XPS spectrum by a negative shift in the binding energy of metallic Pd [5–9]. The alkaline and alkaline-earth oxides act then as expected on the basis of their basic properties, that is increase the electron density of the supported metallic particles and therefore, decreasing the adsorption of double bond onto the metal surface, the attack of hydrogen on the double bond is less favoured. A XPS investigation is actually in progress to verify if a similar behaviour occurs also for metallic Pt on the Pt/Z-PM catalysts. Regarding iron oxides, the role of Fe can be associated instead to the increase of adsorption of the substrate through the carbonyl group. This behaviour is related to the polarization of the –CH@O group by Fe ions. It is well known that Fe, when added to platinum supported catalysts, increase the selectivity towards unsaturated alcohols in the hydrogenation of several unsaturated aldehydes [11–13]. On the basis of data so far available, we cannot unambiguously discriminate what is the main promoting process. Most likely, the behaviour of the catalyst supported on Z-PM derive from a cooperation of the several promotion processes above mentioned. A detailed investigation is necessary in order to better clarify this point and the results will be communicated in a forthcoming paper. 3.2.2. Sn-doped Pt catalysts Previously we reported that the selective hydrogenation of campholenic aldehyde to naturanol can be effectively promoted by addition of tin to monometallic Pt catalysts, leading to the formation of the unsaturated alcohol in high yield [23]. This is in accordance with literature data indicating that bimetallic Pt–Sn catalysts are highly selective in the hydrogenation of several unsaturated aldehydes to unsaturated alcohols [30]. Therefore, we fo-

100

80

Composition (%)

2088

60

CPA NAT ACS NATS

40

20 OH

(NAT)

0 0

O

OH

(CPA)

100

200

300

400

500

Time (min.)

(NATS) Fig. 4. Composition of the reaction mixture vs reaction time in the hydrogenation of campholenic aldehyde on the Sn-Pt/Z-PM catalyst. CPA = campholenic aldehyde; ACS = saturated campholenic aldehyde; NAT = naturanol; NATS = saturated naturanol. Reaction conditions: [CPA] = 1.23  10 2 mol/l; Solvent = cyclohexane; T = 343 K; Cat. weight. = 250 mg.

O

(ACS) Scheme 1. Hydrogenation of CPA on Pt- and Pt–Sn/Z-PM catalysts.

Table 2 Comparison of the catalytic activity and selectivity to reaction products in the hydrogenation of campholenic aldehyde on monometallic Pt catalysts Catalyst

Pt/SiO2 Pt/Z-PM

Activity (min

1.07 0.29

1

/gcat  10)

Selectivity (50% conversion)

Selectivity (80% conversion)

ACS

NAT

NATS

ACS

NAT

NATS

93.2 0

1.9 99.2

4.9 0.8

87.9 0.9

0 93.9

12.1 5.2

G. Neri et al. / Catalysis Communications 9 (2008) 2085–2089

the C@O group is significantly better compared to performance of the corresponding silica-supported catalysts. Its use as a catalyst support, combined to other uses already enlightened [15,16], allows advantages in the extractive process from both environmental and economic points of view, because an useful co-product is ever a better option for the environment than to be disposed off as a waste and, moreover, disposal charge is saved. Consequently, the novel support could be produced at a low price, likely less than many conventional catalytic supports.

Selectivity to naturanol (%)

100

80

60

Pt-Sn/Z-PM Pt/Z-PM Pt-Sn/SiO2

40

2089

Pt/SiO2

References

20

0 0

20

40

60

80

100

Conversion (%)

[1] [2] [3] [4] [5]

Fig. 5. Selectivity to naturanol vs. substrate conversion profiles for the investigated catalysts.

[6]

cused our attention on the effect of tin addition on the Pt/Z-PM system. Fig. 4 show the composition of the reaction mixture vs. reaction time in the hydrogenation of campholenic aldehyde on the Pt–Sn/ Z-PM catalyst. The hydrogenation reaction led almost exclusively to the formation of naturanol and only little NATS was formed at a conversion approaching 100%. Data reporting the selectivity to naturanol vs. substrate conversion for the catalysts investigated are summarised in Fig. 5. The Sn-Pt/Z-PM catalyst results the most selective towards the unsaturated alcohol at any conversion. It is also noteworthy that whereas on the other catalyst the selectivity to NAT tend to decrease, particularly at the higher conversion level, on the Sn–Pt/Z-PM catalyst naturanol is obtained with a 96% or more selectivity up to 95% substrate conversion.

[7] [8] [9] [10]

4. Conclusions

[22]

[11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]

[23]

A highly crystalline zeolitized-pumice (Z-PM) having a larger surface area compared to usual pumice materials can be easily prepared by a silica extraction process from a pumice mine waste as a co-product of sodium silicate. In this paper it is proposed as a new catalyst support. The catalytic properties of Pt and Pt–Sn catalysts supported on Z-PM were investigated in detail in the hydrogenation of campholenic aldehyde. The results reported show that the behaviour of these novel catalysts in the selective reduction of

[24] [25] [26] [27] [28] [29] [30]

U. Kelma, V. Sanhueza, C. Guzman, Appl. Clay Sci. 24 (2003) 35. O.E. Kuzugudenli, Key Eng. Mater. 264–268 (2004) 1427. G. Bond, J. Turkev, Trans. Faraday Soc. 50 (1954) 1335. K. Venkata, S. Rao, A. Rachel, M. Subrahmanyam, P. Boule, Appl. Catal. B 46 (2003) 77. G. Deganello, D. Duca, A. Martorana, G. Fagherazzi, A. Benedetti, J. Catal. 150 (1994) 127. G. Deganello, D. Duca, A. Martorana, A.M. Venezia, A. Benedetti, G. Fagherazzi, J. Catal. 151 (1995) 125. D. Duca, L.F. Liotta, G. Deganello, J. Catal. 154 (1995) 69. D. Duca, L.F. Liotta, G. Deganello, Catal. Today 24 (1995) 15. D. Duca, F. Frusteri, A. Parmaliana, G. Deganello, Appl. Catal. A 146 (1996) 269. Kirk-Othmer, Concise Encyclopedia of Chemical Technology, WileyInterscience, New York, 1979. P. Maki-Arvela, J. Hajek, T. Salmi, D. Yu. Murzin, Appl. Catal. A 292 (2005) 1. P. Gallezot, D. Richard, Catal. Rev. Sci. Eng. 40 (1998) 81. G. Neri, S. Galvagno, Rec. Res. Dev. Catal. 2 (2003) 121. F. Corigliano, I. Arrigo, P. Catalfamo, S. Di Pasquale, P. Primerano, Patent Pending No. MI2004A 1766, 2004. I. Arrigo, P. Catalfamo, L. Cavallari, S. Di Pasquale, J. Hazard. Mater. 147 (2007) 513. P. Catalfamo, I. Arrigo, P. Primerano, F. Corigliano, J. Hazard. Mater. 134 (2006) 40. A.P. Jansen, R.A. van Santen, J. Phys. Chem. 94 (1990) 6764. E.J. Creyghton, J. Mol. Catal. A 134 (1998) 47. D.G. Blackmond, R. Oukaci, B. Blanc, P. Gallezot, J. Catal. 131 (1991) 401. P. Gallezot, A. Giroir-Fendler, D. Richard, Catal. Lett. 4 (1990) 169. J. Álvarez-Rodríguez, A. Guerrero-Ruiz, I. Rodríguez-Ramos, A. Arcoya-Martín, Catal. Today 107–108 (2005) 302–309. US Patent number 4,766,002 assigned to Givaudan Corporation, Clifton, NJ, 1988. G. Neri, G. Rizzo, A. Aricò, C. Crisafulli, L. De Luca, A. Donato, M.G. Musolino, R. Pietropaolo, Appl. Catal. A 325 (2007) 25. R. Aiello, A. Nastro, F. Crea, C. Colella, Zeolites 2 (1982) 290. R. Ottana, L.M. Saija, N. Burriesci, N. Giordano, Zeolites 2 (1982) 295. N. Burriesci, M. Luisa Crisafulli, N. Giordano, J.C.J. Bart, G. Polizzotti, Zeolites 4 (1984) 384. C.D. Johnson, F. Worrall, J. Mater. Chem. 17 (2007) 476. D.W. Breck, Zeolite Molecular Sieves, Robert E Krieger Publ. Co., Florida, 1984. William C. Beard, in: Molecular Sieve Zeolites, Adv. Chem Ser., vol. 101, 1971, p. 237. J.A. Ryndin, C.C. Santini, D. Prat, J.M. Basset, J. Catal. 190 (2000) 364.