SciELO - Scientific Electronic Library Online

 
vol.92 issue1-3O-O bond dissociation enthalpy in perfluoromethyl fluorocarbonyl peroxide, CF3OOC(O)F. A density functional quantum chemical analysisSolvent effect on the 1H NMR spectra of a pyridinocalix(4)arene derivative and its protonation constants in methanol author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

  • Have no cited articlesCited by SciELO

Related links

  • Have no similar articlesSimilars in SciELO

Share


Anales de la Asociación Química Argentina

Print version ISSN 0365-0375

An. Asoc. Quím. Argent. vol.92 no.1-3 Buenos Aires Jan./July 2004

 

REGULAR PAPERS

Environmental Application Of Natural Microporous Aluminosilicates: NOx Reduction By Propane Over Modified Clinoptilolite Zeolite

Botto, I.L.1; Canafoglia M.E.1; Lick, I.D.2; Cabello, C.I.2; Schalamuk, I.B.3; Minelli, G.4; Ferraris, G.4

1Centro de Química Inorgánica (CEQUINOR-CONICET), Facultad de Ciencias Exactas, Universidad Nacional de La Plata, 47 y 115, La Plata (1900), Argentina.
FAX :+(54-221)4259485; E-mail: botto@quimica.unlp.edu.ar;
2Centro de Investigación y Desarrollo en Ciencias Aplicadas (CINDECA-CONICET), Facultad de Ciencias Exactas, Universidad Nacional de La Plata, 47 No 257 , La Plata (1900), Argentina.
3 Instituto de Recursos Minerales (INREMI-CICPBA), Facultad de Ciencias Naturales y Museo, 60 y 122, La Plata (1900), Argentina
4 Istituto di Metodologia Inorganica e di Plasma (IMIP), Universidad “La Sapienza”, Roma (00185), Italia

Received March 11, 2004. In Final form April 15, 2004.
Dedicated to Prof. Pedro J.Aymonino on the occasion of his 75th birthday

Abstract
Synthetic exchanged zeolites have proved to be active catalysts for the selective catalytic reduction  (SCR) NOx by hydrocarbons as reductors in the presence of O2. The abundance and potentiality of argentine  zeolite mineral resources encouraged us to analyze the possibitity of taking advantage of these minerals as starting materials for catalyts preparation. In this sense, clinoptilolite-rich tuffaceous material was selected. Clinoptilolite (named CLI), belonging to the heulandite–type zeolite (HEU), was treated with Co(II) and Co(II)-Rh(III) solutions to obtain the monometallic and bimetallic zeolite systems, precursors of the catalysts. The mineral, the precursors and the catalysts were characterized by means of several techniques such as XRD, XRF, FTIR and DRS spectroscopies, thermal studies  (DTGA, TPR), SEM-EDAX microscopy and textural analysis. The steps of the chemical treatment were carried out, in part, in solution, involving the mineral transformation into the NH4-CLI, Co(II)-CLI and Co-Rh-CLI precursor phases. Catalysts were obtained by thermal treatment at 500ºC. TPR measurements provided information about the interactions and characteristics of the active metals with the aluminosilicate framework of the catalysts. Two types of catalysts were evaluated for the NOx reduction reaction using propane in O2 presence: 0.5%(w/w)Co-CLI and 0.25%(w/w)Rh--0.5%(w/w)Co-CLI respectively. The performance of catalysts obtained from natural zeolites revealed that they were active for the reaction, despite the very low metallic content.

Resumen
Las zeolitas sintéticas intercambiadas con metales han demostrado ser activos catalizadores para la reducción catalítica selectiva (SCR) de óxidos de nitrógeno (NOx) mediante  el uso de hidrocarburos como reductores, en presencia de O2. La abundancia y potencialidad de los recursos naturales argentinos en zeolitas, nos impulsó a analizar la posibilidad de aprovechamiento de esos minerales para su uso como materiales de partida en la preparación de catalizadores. En tal sentido, se seleccionó material tobáceo, rico en clinoptilolita. La clinoptilolita (denominada CLI) fue tratada con soluciones de Co(II) y de Co(II)-Rh(III) para obtener sistemas zeolíticos mono- y bimetálicos, precursores de los catalizadores. El mineral, los precursores y los catalizadores fueron  caracterizados mediante distintas técnicas, tales como XRD, XRF, espectroscopías FTIR y de reflectancia difusa, estudios térmicos (DTGA, TPR), microscopía electrónica de barrido (SEM-EDAX) y análisis texturales. Las etapas del proceso  químico fueron, en parte,  en solución acuosa,   implicando la transformación del mineral en las fases precursoras NH4-CLI, Co-CLI y Rh-Co-CLI. Los catalizadores fueron obtenidos por tratamiento térmico a 500ºC. Las medidas de TPR permitieron  obtener información acerca de la interacción y características del metal activo en el esqueleto aluminosilícico de los catalizadores. Se evaluaron dos tipos de catalizadores para la reacción de reducción de NOx a Nmediante el empleo de propano, en presencia de O2 : 0,5 % (en peso) Co-CLI y 0,25 % (en peso) Rh- 0,5 % Co-CLI.  El comportamiento de los catalizadores obtenidos a partir de zeolita natural reveló que ellos son activos para la reacción, no obstante el muy bajo contenido metálico.

Introduction

     Natural zeolites are aluminosilicates with a 3D framework, building from TO4 tetrahedra (T=Al, Si), delimiting channels and cages of different dimensions. Molecular sizes of 6, 8, 10 and 12 member ring openings are ranged between 3 to 7.5 Å [1].
     Zeolites can be distinguishes from some other tectosilicates by the framework density (FD) expressed as number of T atoms per 1000 Å 3 [1,2]. This parameter (ranged between 12 and 20 for zeolites and higher than 20 for non-zeolitic framework structures) is related to the pore volume but it does not reflect the size of the pore openings. Hence, the application of zeolites in adsorption and ion exchange processes is derived from this structural feature [2].
     The second half of 20th century was transcendent in the zeolites development. From the geological point of view, enormous deposits of zeolite bearing rocks were discovered in many parts of the world. Likewise, the interest in inorganic ion exchangers was connected with the growth in the nuclear industry and the need of suitable ion exchangers for processing radionuclide-containing waters. The stimulated fundamental research on inorganic ion exchangers points out the selectivity and specificity for certain ions, opening the way for applications in the treatment of industrial exhausted solutions, in wastewater treatments and pollution control [3-5].
     On the other hand, the synthesis of zeolite-analogues and the preparation of new materials from the bare mineral species have been topics of considerable interest. So, the exceptional physicochemical properties of these microporous materials offered new and interesting perspectives for basic and applied studies in the chemistry field. Hence, at the present time, a wide range of applications in petrochemistry, fine chemistry, environmental protection, agronomy and more recently in the pharmaceutical industry and medicine are based on the functionalization of these materials [2,6-8]. In all cases, inorganic chemistry plays a crucial role in the scope of the zeolite development: not only in designing and discovering new zeolite-like materials, but also in the chemistry routes to transform natural species in useful compounds as well as in the study of structural, thermal and spectroscopic properties. 
     Although the natural microporous aluminosilicates were extensively used in environmental protection (waste managements) from processes related to ion-exchange in solution, the usefulness of zeolite tuffs in catalysis is limited and recent [9-12]. In this field the pioneering works were directed by the synthetic zeolites.
     The increasing of toxic NOx emissions from anthropogenic activities and the interest in improving the environmental control promoted the formulation of novel materials as active/selective catalysts for the nitrogen oxides reduction. In this sense, several synthetic exchanged Metal-zeolites (M= transition element)  showed to be attractive catalytic materials for the selective catalytic reduction (SCR) of NOx by hydrocarbons in oxygen presence [13,14].
     The aim of this work is to study applications of argentine zeolitic tuffs taking into account the abundance, availability and low cost of these natural resources in our country. The structural, thermal and chemical characterization of natural zeolites was made with the purpose of using them in the preparation of monometallic (Co) and bimetallic (Co/Rh)-zeolite catalysts. The catalytic performance of these materials for the model reaction of NOx- reduction to N2  by means of propane (SCR) in presence of O2 was evaluated.  Clinoptilolite-rich tuff was selected for the transformation in catalytic materials. Bare and treated minerals by means of inorganic methods (ion-exchange precursors) were analyzed by SEM-EDAX microscopy, FTIR and DRS spectroscopical techniques, XRD analysis and thermal and textural methods. The catalytic performance of the NOx-SCR reaction resulted from the comparison with the Co/Al2O3 catalytic testing data providing supplementary evidence for the potentiality of mineral species in the preparation of useful materials.

Experimental

     Samples of clinoptilolite from different deposits of our country were characterized by means of X-ray diffraction (XRD), atomic absorption spectroscopy and X-ray fluorescence analysis (XRF) (University of Geneve), SEM-EDAX microscopy and thermal (DTGA) and textural studies. The surface data were obtained by N2 adsorption-desorption at 77 K using a Micromeritics ASAP 2010 analyzer. The samples were preheated under vacuum in two steps of 1h at 100oC and 1 h at 200oC. BET specific surface area, were obtained from adsorption data in the relative pressure range 0.05-0.2. BJH pore size distribution [15], were obtained from data of the desorption branch of the isotherm micropore analysis by t-test [16], adopting the Harkins & Jura reference curve [17] and total pore volume by Gurvitsch rule [18]. Original sample showed a specific area of ~10 m2/g whereas the specific areas of exchanged samples were between 11 and 14 m2/g, depending on the treatment.
     The thermal stability was also analyzed by means of “in situ” XRD analysis (in air and in water vapour atmospheres) up to ~700ºC.
     From the preliminary study, a clinoptilolite-rich sample with the chemical composition given in table 1 was selected to use as parent tuffaceous material in the preparation of  the precursor phases leading to the catalysts by thermal activation. A small proportion of SiO2 (quartz form) was observed in all samples as a subordinated mineral. The presence of vitrous phase (rhyolitic glass) was observed by petrographical mycroscopy. Volcaniclastic material is the precursor of sedimentary zeolite from reaction in an open hydrological system. From the geochemical point of view, the general hydration sequence: Glass + H2O → clinoptilolite + ions, left a small proportion of SiO2 that crystalized as a late alteration product.

Table 1:Chemical analysis of bare clinoptilolite by XRF analysis Majority elements (as % oxides)

SiO2

TiO2

Al2O3

Fe2O3

FeO

MnO

MgO

CaO

Na2O

K2O

P2O5

H2O

66.99

0.16

13.39

0.87

0.00

0.01

0.99

2.54

3.53

1.48

0.04

10.08

Minority elements (in ppm )

Nb

Zr

Y

Sr

U

Rb

Th

Pb

Ga

Zn

Cu

Ni

25      

79

28

1421

<2

37

10

10

11

22

13

<2

Co

Cr

V

Ce

Nd

Ba

La

S

Hf

Sc

As

<2

9

16

43

16

381

21

5768

6

7

<3

     Although clinoptilolite is characterized by a K>Na content, the studied sample showed an inverse trend, which can be explained by the zeolite genesis from the hydrolysis of the volcaniclastic material in a Na-rich saline solution.
     In a first step, CLI was treated with ammonium chloride solution to exchange alkali and alkaline-earth elements by NH4. Treatments were done at different experimental conditions: i) from RT up to 120ºC, ii) from 24 to 240 h., iii) at different ammonium concentration iv) with and without stirring. These preliminary treatments permitted to select the conditions to obtain NH4-CLI. So, 10g of (Na,K)CLI were treated with excess of NH4Cl 0.5 M at RT for 24 h. with constant stirring. The sample was heated up to 300ºC for 24 h to eliminate partially the NH4. The complete evolution of NH3 occurred at ~500ºC to give the acid zeolite, (named H-CLI).
     The monometallic Co-CLI precursor of catalysts was obtained by treating NH4-CLI with cobalt acetate solutions (variable concentration). Selected conditions to obtain Co-CLI were as follows: 5 g of NH4-CLI (after thermal treatment at 300ºC) were treated with solution of cobalt acetate 0.01M at RT for 12h (with stirring). The profile of the ion exchange isotherm in the experimental conditions led to a 0.5% (w/w) Co. The isotherm reports the equivalent fraction of the incoming cation, present at equilibrium in the liquid phase vs. the equivalent fraction of the cation in the zeolite. The chemical analysis of cobalt was made by AAS considering the initial and final concentration of the cobalt acetate solution used in the treatment. Standards of cobalt acetate were used.
     The bimetallic Rh-Co-CLI precursor was obtained from treating 2 g of Co/CLI (0.5 % Co) with a Rh(III) chloride solution ( 10-4 M)  until dryness ( final content of Rh = 0.25% (w/w)).
     The catalysts were obtained from the precursors by treatment at 500ºC in flowing air for 3 h.
     FTIR spectroscopy was particularly used to analyze the NH4 exchanging and its elimination by heating. Measurements were carried out in a Bruker  (IFS-66) Spectrophotometer with the KBr pellet technique.
     The DRS spectra were registered in a Beckman DK1 spectrophotometer between 2500 and 210 nm, by using MgO as reference.
     DTGA were registered on a Shimadzu 50 equipment under inert atmosphere.
     Thermal studies in  a reducing atmosphere (H2/N2 =10/90) were made in a home-made temperature programmed reduction equipment (TPR). This technique enables the determination  of interactions of the cobalt species with the zeolitic framework.
     Catalytic reactions were evaluated in a fixed bed reactor, heated electrically and the temperature measured by means of a thermocouple of the K type in contact with the catalytic bed. The reaction mixture was obtained from four feed lines: NO-He, methane or propane /He, O2/He and He as balance. Before reaction the catalysts were activated at 500ºC for 1 h.. The reaction was carried out between 250 and 700ºC. The reactant mixture had a composition 1500ppm of NO, variable concentration of O2 (0.8-2.5 %), 2000ppm of C3H8 and space velocity (GHSV) of 8600 h-1. The catalyst mass used was 0.400 g and the total flow rate was 50 ml/min. Reaction products were analyzed by using a gas chromatograph, Shimadzu GC-8a, provided with a thermal conductivity detector. The separation of products was performed with a concentric column CTRI (Altech).
     Catalytic NO-N2 conversion is calculated as XN2= 2[N2]/[NO] whereas the propane conversion (to give CO2), as the secondary reaction of combustion, is expressed as XCO2= 1/3 [CO2]/[C3H8].

Results And Discussion

     Clinoptilolite (CLI) is a sedimentary zeolite with a typical chemical composition (Na,K)6(Al6Si30O72).20H2O. This is the most abundant of natural zeolites and can be classified as a silica-rich member ((Si/Al) ratio ~5 and FD=17). The framework topology is characterized by a two dimensional pore system with the presence of 8-ring channels that intersect two parallel 10-ring and 8-ring channels respectively ([100] 8 2.6 x 4.7 ← →{[001 10 3.0 x 7.6 +8 3.3x4.6}) [2,19]. This zeolite presents good adsorption properties for CO2, SO2, NH3 and NOx gases [2]. The relatively open structure of heulandite-type zeolites enhances the sorption properties. The chemical composition of the aluminosilicate framework, modulates the ion exchange property. This is expressed as cation exchange capacity (CEC) and it can theoretically reach  330 meq/100g [20].
     
CLI is highly selective for the ammonium cation. The selectivity sequence at RT is NH4~K >Na>Ca>Mg.  According to these data, it is expected a low incorporation of Co. However, this is much more effective when the CLI was previously exchanged with NH4. The NH4 exchange is an indirect method to obtain the H-CLI that avoids the use of an acid treatment. Likewise, the protonic acid site seems to play a synergistic role interacting with the metallic specie in the metal modified synthetic zeolite catalysts [9].
     There are different sorption mechanisms including the so-called ion exchange. The sorption of metallic ions seems to be affected by a combination of factors, corresponding to the physicochemical characteristics of both, the solution and the solid. Ion exchange depends on the temperature, the pH and concentration of the solution and particularly on the hydrated ionic radii of the ions as correlated to the zeolitic channels. The temperature enhances the ion exchange while the pH is also critical [21]. So, the most optimum conditions for successful metal loading should be related to exchange cations of low valence and large size (small hydrated radious) by using increased temperature and rather high pH (in order to minimize the effect of mobile H+ ions). The high specific surface area as well as the long treatment time (by a stepwise process) favour the ion exchange [21] In HEU-type zeolites the ion exchange is easy for alcali and alcaline earth elements but it is not so clear for transition ion elements. In this case, the process is strongly dependent on the aqueous chemistry of the “d” species and thus the hydrolisis reactions can give soluble and/or insoluble products that interact with the microporous zeolite.  As a consequence, the sorption mechanisms do not only comprise the ion exchange process (absorption into the channels in the aluminosilicate framework) also adsorption (specific and/or non-specific regarding inner-sphere and/or outer-sphere surface complexation) and surface precipitation and co-precipitation processes. This last reactions can generate non-stoichiometric phases, composite materials, metal ions supported on the surface,  etc) [22,23].
     The synthesis of HEU-type zeolites modified by Co was reported in the literature [21], although controversial opinions were expressed respect to the Co-content. The highest reported value was 7.5%(w/w). With this concentration the simultaneous presence of Co species as supported phases, segregated oxides and exchanged zeolite were suggested by SEM-EDAX as well as XPS results [21]. The presence of Co(III) in the form of composite materials or segregated phases are strongly dependent on the Co loading. Another information about Co-HEU type zeolites was available by Kim et al. [24]. They reported the synthesis of a exchanged Co-CLI zeolite with the Co1.9H1.5Al5.3Si30.7O72 composition (Si/Al ~5.8) [24]. However even at  the highest surface area of synthetic zeolite (114 m2/g) and under drastic experimental conditions, the full exchange by cobalt was not reached. The reported exchange corresponds to a cobalt content of ~5 % (w/w) [24].
     It is common that natural zeolites present a very low surface area. This characteristic can be particularly attributed to a pore blockage of the crystalline zeolite phase by the vitreous zeolite-precursor. Because of a great affinity for the ammonia ion, it is possible to get, by ion exchange, the monoionic ammonia form of CLI. Thus, we have observed by EDAX that all ions, except a part of potassium were exchanged by NH4. However, a mild conditions employed for the Co exchange (RT) led to the very low content of this metal (only 0.5 % (w/w)).


Figure. 1 : Nitrogen adsorption-desorption isotherms at 77 K for Clinoptinolite (CLI) and Co-containing Clinoptinolite (Co-CLI)

     Fig 1 shows the nitrogen adsorption-desorption isotherms at 77 K for CLI and Co-CLI respectively. In order to evaluate the total pore volume, the Gurvitsch rule has been applied to the adsorption data at P/Po = 0.99 (instead of the usual value of P/Po = 0.95 adopted when a definite plateau is present). The hysteresis loop of  the desorption branch reveals the existence of large pores (meso and macroporous) and a limited amount of micropores in the samples. It is evident that the two samples do not possess the typical porous structure expected for synthetic zeolites as silicalite and ZSM-5, which give rise to another type of adsorption isotherms according to the classification of Brunauer, Deming, Deming and Teller (BDDT) [25].
     Results given in Table 2 indicate that the preparation method to introduce cobalt does not significantly modify the textural parameters of the host structure. However, it is known that an increase of micropore and external/macropore area is clearly observed after ion exchange of natutral HEU with ammonium [9]. That is, the preparative method for the incorporation of Cobalt restored part of the original mesoporosity of the host matrix.

Table 2. Textural features of CLI and Co-CLI precursor

 SAMPLES

BET aream2 g-1

Vtot ML g-1

Sext m2 g-1

Vμ mL g-1

Dp Å

CLI

11.0

0.043

8.4

0.001

n.d.

Co-CLI

14.5

0.052

11.8

0.001

n.d.

     The DTGA plots of CLI, NH4-CLI and Co-CLI, shown in Fig 2, reveal the water loss up to ~250ºC.


Figure 2: DTGA plots of: a) NH4-CLI, b) Co-CLI, c) CLI

     The NH3 loss from the NH4-CLI (curve a) is observed at ~380ºC, a situation that leads to the formation of Brönsted acid sites (H-CLI). The Co-CLI (curve b) obtained from NH4-CLI partially transformed in the acid form at 300ºC, clearly reveals the presence of NH4 ions. In this case the NH3 evolution occurs at higher temperature than that observed for the NH4-CLI (420ºC). The  temperature shifting can be attributed to the selective adsorption of this gas as well as the interaction between Co(II) and NH3. This is not unexpected because the intrazeolitic or surface cobalt ions are Lewis acid centres that can retain easily the NH3 Lewis base.
     Figure 3 shows the FTIR spectra of NH4-CLI. The splitting of the NH4 bending-mode (~1400 cm-1) is attributed to the lowering of the Td symmetry by interaction with the aluminosilicate framework. A higher intensity of the band at 1640 cm-1(in comparison with the band of water of the untreated CLI), corresponds to the overlapping of the bending mode of water with  some component of the NH4 splitting. The presence of aducts between the ion and the acid sites of the zeolite can be suggested [26].
     The interaction between the zeolite framework and NH4 ions and the further alteration by heating can be drawn as:

  ------Falta Imagen-------

species can be observed  at temperatures higher than 650oC [2].

Figure 3: FTIR spectrum of NH4-CLI ( between 4000 and 450 cm-1)

X-ray diffraction lines of the parent CLI remained intact during the NH4 or Co(II) ion exchange, revealing the low effect of the extra-framework ions. Likewise, the structure of Co-CLI was preserved by heating up to ~700oC. Fig. 4 shows the Co-CLI “in situ” X-ray diffraction pattern by heating up to 800ºC.

 

Figure 4: (in situ)  XRD patterns  of CoCLI (air atmosphere)

It is interesting to note that the thermal treatment is useful to differentiate clinoptilolite from heulandite [2]. Although heulandite and clinoptilolite are isotypic, the different stability can be related to the lower Si/Al ratio of the former. This is reliable for a different distribution of the extraframework cations and consequently heulandite show a complete structural collapse at ~450ºC [27].  The identity of the zeolite and the thermal stability of exchanged clinoptilolite was so corroborated.
     The thermal behaviour in water vapour atmosphere is shown in Fig 5. The crystal structure remains without changes and only a slight loss of crystallinity is observed by heating up to ~700 ºC. No clear signal of dealumination was registered by EDAX analysis.
     Fig 6 shows the DRS spectra of Co-CLI precursor heated at 300oC. Although the Co content is low, weak signals of Co- d-d transitions are observed. The two broad bands in the 700 and 1400 nm seem to be related to the presence of Co(II) and Co(III) species, whereas the band at ~410 nm can be attributed to Co(III)-Co(II) CT [28]. In this sense, a re-arrangement of the original site of cobalt (associated to O atoms and water molecules) is observed by heating. The elimination of water suggests a lowering of local

Figure 5 :(in situ)  XRD patterns  of CoCLI (water vapour  atmosphere)

 

Figure 6: DRS of Co-CLI precursor ( heated at 300ºC)

symmetry of the metal site. A strong interaction effect is governed by the Co redox chemistry and by diffusional aspects. So, the redox properties and the metal site-symmetry are factors used to explain the efficient enzimatic catalysis trough the pre-formation of a transition state so called “entatic state” [29]. It is expected that the redox chemistry of Co(II)-Co(III) is energetically favoured by a d7 Co(II) distorted tetrahedral environment (with completely filled bonding eg orbitals) and by d6 low spin Co(III) in octahedral coordination ( with the completely filled bonding t2g orbitals). These facts, as well as the low content of Co in the precursor and the chemistry of this element in solid state decrease, still more, the resolution of the DRS spectrum of catalysts (500ºC). So, the TPR technique, described below, will help to analyze the possible Co-interaction and/or segregation from the aluminosilicate framework.
     It is evident that the zeolite framework (in the presence or absence of water) functions as a huge ligand to coordinate and stabilize the metal cations. The nature and extent of interactions between the cations and zeolite framework, depending on the size and shape of the pores where the metals can be located, are useful to analyze the catalytic properties for the SCR reaction of NOx by hydrocarbons. In general, it was suggested that the Co(II) sites in 8-ring channels are thermodynamically more stable than those in 10-ring channels for this reaction and consequently the  CLI , with two types of 8-ring channels may play an important role in achieving the activity for NOx reduction by CH4 [24]. On the other hand, the migration of Co(II) ions to the internal sites of zeolite by temperature effect can be considered in the activation of the precursor to give the catalysts, without causing diffusional problems with the reactive molecules [24].
     It is generally accepted that the reduction of NOx by hydrocarbons in oxygen excess proceeds through  a series of steps. So it is expected that adequate combination of metal species (redox properties) may give more active catalysts. In this sense, bifunctional catalysts containing transition metals of the first series “d” and some element of the Pt group may improve the catalytic activity and the tolerance of water vapour, preventing severe dealumination process after hydrothermal aging. Hence, the acid-base, redox and structural properties are related to the potentiality of H-Zeol, the synergetic effect and the topology of the aluminosilicate matrix toward the geometry and characteristics of reductor hydrocarbon. In this way, we direct the attention to the Rh-Co-CLI system. The presence of Rh does not affect the structural feature of clinoptilolite, according to XRD analysis.
     An adequate form to analyze the interaction of metals with the aluminosilicate framework seems to be the temperature programmed technique (TPR). This method allows to identify the presence of   segregated phases, intrazeolitic metal cations as well as metal ions supported on the surface. It is possible to know about the activation of the catalyst from the availability of metals for their reduction by H2 [30]. The thermal behaviour under reduction conditions is observed in Fig. 7. The TPR pattern of zeolite parent clinoptilolite does not show signals of reducible species up to 950ºC. In this sense, it is interesting to denote that only one weak signal of reduction appears at 980oC. This is subsequent to the aluminosilicate framework collapse, which left available iron species. This fact is very interesting because the TPR technique is revealed as a useful technique to show the existence of structural Fe(III)  (by isomorphous replacement of Al(III)). In this sense, we are analyzing the reducing behaviour of different natural aluminosilicates (clays and zeolites), in which it is usual the presence of two types of iron species, according to the mineral genesis: those species related to the associated minerals as impurities (amorphous or crystalline Fe(III)-oxide-hydroxide such as goethite, limonite, haematite, etc) and the structural Fe(III), in the covalent  framework.
     The TPR pattern of the Co-CLI catalyst is characterized by two signals: the first one at ~580ºC and the second one at ~900ºC. This last shows a higher intensity than that observed in the bare CLI pattern and also appears at lower temperature. It is well known that binary cobalt oxides (CoO and Co3O4) as bulk present TPR signals in the 400 oC region, corresponding to the Co(III)-Co(II) and Co(II)-Co reduction processes [30]. The reducibility of cobalt species decreases when it is supported [30]. From the different Co-species in activated Co-Zeol systems the reduction temperature of extraframework cobalt (including free oxides and surface cations) is lower than that of intrazeolitic type. It can be established the following order: free Co-oxides ( ~400ºC), surface cobalt (500-600ºC) intrazeolitc and deeply anchored cobalt (temperature higher than 700ºC) [31].

Figure 7: TPR patterns of : a) CLI, b) Co-CLI, c) Rh-Co-CLI

     Hence, it can be clearly observed the presence of Co(II) and Co(III)  of intermediate interaction and Co(II) species with a strong interaction but it can be disregarded the presence of free binary Co-O oxides (from segregated crystalline oxides or amorphous composites). These phases have been observed in related systems by heating the precursors at temperature higher  than 650ºC [30,32].
     The incorporation of Rh(III) affects the Co(II) reducibility. Free Rh(III) oxide reduces at very low temperature (~135ºC), according to experimental conditions [30]. It is also known that the reducibility of this metal can be affected in a supported system. Rh(III) supported on SiO2 or TiO2 reduces at lower temperature whereas when it is supported on more basic oxides, the reduction temperature is increased up to 250oC [30]. Therefore, we are in conditions to analyze the TPR pattern of the bimetallic catalyst Co(0.5%)-Rh(0.25%)-CLI. The first reduction zone, located between 50 and 300ºC, can be associated to the Rh(III) –Rh process. From this reaction, the H2
2H activation favours the intrazeolitic/surface Co(II) reduction, showing  differences with  the monometallic system. A continuous reduction process between 300 and ~650ºC can be assigned to Co(II) species of different interaction.  Likewise, the strong and asymmetric signal between 780 and 900ºC is attributed to the Rh activation effect on the reducibility of Co(II) in the oxide matrix after the zeolite collapse and  also on that of the structural iron.
     
Fig 7 gives the NO-N2 conversion for the Co-CLI and Rh-Co-CLI catalysts measured as a function of reaction temperature at a GHSV of 8600 h-1, according to the experimental conditions. Although the bare clinoptilolite (activated at 500o C) is slightly active for the studied reaction (conversion of 17% at 600oC), the Co-CLI catalyst shows an activity increase by metallic effect (maximun of conversion close to 31 % at 600 oC). The Rh-Co-CLI catalyst does not involve an increase of the activity (conversion of 40 %) but the reaction occurs at lower temperature (400oC).

Figure 8: NO–N2 conversion  of   CLI (s), Co-CLI (+) and Rh-Co-CLI (l). Corresponding data forthe C3H8–CO2 conversion are indicated by empty simbols.

     Table 3 sums up data of NO-N2 and of propane-CO2 conversions by using the prepared Co-CLI and Rh-Co-CLI catalysts. Catalytic data of the Co(1.5%)/Al2O3 reference system, previously treated at 1000 oC (in similar experimental testing conditions[33]) and those of natural CLI activated at 500 C are included as reference. The propane-CO2 reaction is associated to the total process where the hydrocarbon reduces the NO but also reacts with O2 as a secondary process, delimiting the selectivity of the catalysts. In this sense, the catalyst promoted with Rh favours the secondary reaction to the detriment of the primary process. Thus, the diminution of the activity for the NO-N2 reaction by the use of bimetallic catalyst is attributed to the competency between the primary and secondary reactions. The selectivity is thus very low for the assayed NOx reduction process.

Table 3: Catalytic performance of studied catalysts (as conversion %) for NO-N2 reduction (principal reaction) and C3H8-CO2 oxidation (secondary reaction). Data of Co(1.5%)/Al2O3 catalyst and activated CLI, in similar testing conditions, are included for comparison

 

Conversion reaction

Co(1.5%)/Al2O3
Tmax= 650ºC

Co0.5%CLI
Tmax= 620oC

Activated-CLI
at 500oC
Tmax= 650ºC

Rh-Co-CLI
Tmax= 400 oC

NO-N2

 

       60

       35

     17

     40

Propane-CO2

 

       23

       30

    100

      82

Conclusions

     On the basis of structural, thermal and spectroscopic properties, the studied mineral is a good raw material to be used in catalysis. However, it is possible to improve the performance of these alternative catalytic materials, profiting, more adequately the mineral clinoptilolite as starting material. The Co-CLI catalyst shows good activity, in spite of the very low metallic content.
     Likewise, an appreciable decrease of the reaction temperature is observed by using the Rh-Co-CLI catalyst, although the activity is similar to that of a monometallic system. These results are promising for the studied process and open up new perspectives for the industrial application of natural zeolites.

Acknowledgements

     The authors thank the financial support of CONICET,  CICPBA (Argentina) and CNR(Italy). 

References

[1]       Meier, W. M.; Olson, D. H.; Baerlocher, Ch, Atlas of Zeolite Structure Types, IZA, Elsevier 1996        [ Links ]

[2]       Giannetto Pace, G.; Montes Rendon, A.; Rodriguez Fuentes, G,  Zeolitas, EDIT, Caracas 2000        [ Links ]

[3]      Pansini, M., Mineralium Deposita, 1996, 31, 451        [ Links ]

[4]       Bremner, P. R.; Schultze, L. E., Natural Zeolites ’93, Ming, D. W., Mumpton, F. A. (eds), N.Y. 1995        [ Links ]

[5]       Malion, E.; Loizidou, M.; Spyrellis, N, Sci. Total Environ. 1994, 149, 139        [ Links ]

[6]       Mumpton, F. A., Proc. Nat. Acad. Sci. USA’96, 1999, 3463        [ Links ]

[7]       Farias, T; Ruiz Salvador, A.R.; Rivera, A., Microporous & mesoporous materials, 2003, 61, 117        [ Links ]

[8]       Tsishishvili, G. T.; Andronikahshvili, T. G.; Kirov, G. N.; Filizova, L. D., Natural Zeolites, Springer Verlag , 1992        [ Links ]

[9]       Katranas, T. G.; Vlessidis, A. G; Tsiatouras; V. A; Triantafyllidis, K. S.; Evmiridis , N. P., Microporous and mesoporous materials, 2003, 61, 189        [ Links ]

[10]     Lee, H; Woo, H; Tyoo, R; Lee, K; Lee, J, Appl. Catal. A: Gen. , 2000, 196, 135        [ Links ]

[11]     Long, R.; Yang, R., J. Catal., 2001, 201, 145        [ Links ]

[12]     Mishimo, H.; Hashimoto, K.; Ono, T.; Anpo, M, Appl. Catal.B Environ., 1998, 19, 119        [ Links ]

[13]     Li, H., Armor, J. N., Appl. Catal., B1, 1992, L31; B2, 1993, 239; B3, 1993, L1        [ Links ]

[14]     Iwamoto, M.; Yahiro, H.; Shundo, S.; Yu, Y; Mizuno, N., Appl. Catal., 1991, 69, L15.         [ Links ]

[15]     Barrett, E.P.; Joyner, L.G.; Halenda, P.P., J. Amer. Chem. Soc. 1951, 73 , 373        [ Links ]

[16]     Lippens, B.C. ; de Boer, J.H., J. Catal. 1965, 4 , 319        [ Links ]

[17]     Harkins, W.D. ; Jura, G., J. Am. Chem. Soc. 1944 , 66 , 1366        [ Links ]

[18]     Gurvitsch, L. J. Phys. Chem. Soc. Russ. 1915, 47, 805        [ Links ]

[19]     Mortier, W.J.; Pearce, J.R., Amer. Mineral. 1981, 66, 309        [ Links ]

[20]     Colella, C., Mineral Deposita, 1996, 31, 554        [ Links ]

[21]     Godelitsas, Ath.; Armbruster, Th,  Microporous and Mesoporous Mat., 2003, 61, 3         [ Links ]

[22]     Strumm, W,  Chemistry if the Solid State Interface, Processes at the Mineral- Water and Particle-Water Interface in Natural Systems, J Wiley and Sons, NY, 1992         [ Links ]

[23]     Brown Jr, G. E.; Parks, G. A.; O’Day, P. A. in: D.J.Vaughan, R.A.D. Pattrik (Eds) Mineral Surfaces, Mineral Soc. Series, Vol 5, Chapman & Hall, London , 1995, 129         [ Links ]

[24]     Kim, Y. G.; Kim, Y. Ch.; Hong, S. B.; Kim, M. H.; Kim, Y. P.; Uh, Y. S., Catal. Lett, 1999, 57, 179        [ Links ]

[25]     Brunauer, S.; Deming, L.S.; Deming, W.S.;Teller, E. , J. Am. Chem. Soc.1940,  62 , 1723        [ Links ]

[26]     Geobaldo, F.; Lamberti, C.; Ricchiardi, G.; Bordiga, S.; Cecchina, S; Turnes Palomino, A.; Otero Arean, C, J Phys. Chem., 1995, 99, 11167        [ Links ]

[27]     Mumpton, F. A., Amer. Mineral., 1960, 45, 351        [ Links ]

[28]     Lo Jacono, M.; Cimino, A.; Schuit, G.C.A., Gazzetta Chimica Italiana, 1973, 103, 1281        [ Links ]

[29]     Kaim, W.;  Schwederski,  B., Bioinorganic Chemistry: Inorganic elements in the Chemistry of Life, J. Wiley and Sons, NY, 1994         [ Links ]

[30]     Jones, A.; McNicol, B., Temperature Programmed Reduction for Solid Materials Characterization , Marcel Dekker, Inc , NY,  1986        [ Links ]

[31]     Gutierrez, L.; Boix, A.; Petunchi, J. O., J. Catal, 1998, 178, 179        [ Links ]

[32]     Carrascull, A., Grzona, C.; Lick, D; Ponzi, M; Ponzi, E, React. Kinet. Catal. Lett., 2002, 75, 63        [ Links ]

[33]     I. Lick, E. ; Ponzi, E., React. Kinet. Catal.Lett., 2003, 79, 197        [ Links ]

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License