Catalytic activity of nanometric pure and rare earth-doped SnO2 samples

Catalytic activity of nanometric pure and rare earth-doped SnO2 samples

Available online at www.sciencedirect.com Materials Letters 62 (2008) 1677 – 1680 www.elsevier.com/locate/matlet Catalytic activity of nanometric pu...

608KB Sizes 1 Downloads 16 Views

Available online at www.sciencedirect.com

Materials Letters 62 (2008) 1677 – 1680 www.elsevier.com/locate/matlet

Catalytic activity of nanometric pure and rare earth-doped SnO2 samples Ingrid T. Weber a,⁎, Antoninho Valentini b , Luiz Fernando D. Probst b , Elson Longo c , Edson R. Leite c a

Fundamental Chemistry Department, DQF/CCEN — Av. Prof. Luiz Freire, s/n, Cidade Universitária CEP 50740-540 UFPE, Brazil b Labcath, Department of Chemistry — UFSC CP 476, CEP 88040 900, Florianópolis, SC, Brazil c CMDMC-LIEC, Department of Chemistry — UFSCar, Rod. Washington Luis, km 235 CEP: 13565-905, São Carlos, SP, Brazil Received 4 July 2007; accepted 22 September 2007 Available online 29 September 2007

Abstract This paper describes aspects of the catalytic properties of nanometric pure and rare earth (RE = La or Ce)-doped SnO2 powders, synthesized by a chemical route known as the modified polymeric precursor route (MPPR). Methanol oxidation was chosen as the probe reaction. It was verified that using up to 1 mol% of dopant no detectable influence was found, however changes in the reaction profile were observed with higher concentrations (5 mol%). The nature of dopant was also found to play an important role in the route of the reaction. Ce-doped samples exhibited high methanol conversion even at low temperatures. On the other hand, La-doped samples seem to have a delayed reaction, exhibiting significant conversion only at temperatures higher than 350 °C. © 2007 Elsevier B.V. All rights reserved. Keywords: Rare earth-doped SnO2; Nanomaterials; Methanol oxidation, Catalysts

1. Introduction The catalytic conversion of many types of alcohol is of considerable interest from both the industrial and environmental standpoints. Many kinds of catalysts have been studied over the last three decades. One of these materials is SnO2. Pure SnO2 is not usually employed for catalytic purposes but, in association with other oxides such as Mo, Nb, Sb and so forth, very interesting results can arise. Little information is available in the literature regarding the SnO2-RE (RE = rare earth) system. However, some reports can be found on the sensor properties, morphological characteristics and catalytic activity of this type of system [1–5]. An important fact pointed out in these studies is that the catalytic properties of some semiconductor oxides are strongly related to their acid/base characteristics [2,5–7]. While SnO2 is a predominantly acidic oxide (isoelectronic point, iep= 4–7), the rare earth oxides are mostly basic (ex. iepLa2O3 = 10–12 and iepY2O3 = 11 [8]). Some authors [2,6] believe that either the catalytic activity or sensor properties may be related to the electronegativity of the cations present in the system (and thereby automatically related to the acid/base characteristics). According to these authors, cations ⁎ Corresponding author. Tel.: +55 81 2126 8440; fax: +55 81 2126 8442. E-mail address: [email protected] (I.T. Weber). 0167-577X/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2007.09.058

can be divided into three groups. Low electronegative cations (group i), such as Ca2+ and Cs2+, are considered to enhance sensitivity, while moderately electronegative cations (group ii), especially d block transition metals, are considered to reduce sensitivity. The highly electronegative group (group iii) does not significantly affect the material sensitivity. The two cations investigated in this study have low values of electronegativity (Ce4+ = 1.23 and La3+ = 1.10 [9]) and therefore belong to group i. It is known that alcohols can undergo two kinds of oxidation, i.e., dehydration or dehydrogenation, depending on the physicochemical characteristics of the catalyst [2,10]. If the oxide is principally a Lewis acid, dehydration should occur. Conversely, if the catalyst oxide is principally a Lewis base, the dehydrogenation route is likely to be favored. C2 H5 OH→C2 H4 þ H2 O

ðacidic oxideÞð1Þ

C2 H5 OH→CH3 OH þ H2

ðbasic oxideÞð2Þ

The resulting products can be successively oxidized into CO2, CO and H2O. The H2 can diffuse through the SnO2 lattice, forming surface hydroxyl groups or even water molecules when it reacts with another hydroxyl group [10]. According to the literature [10], dehydrogenation occurs on basic sites and requires higher temperatures than dehydration.

1678

I.T. Weber et al. / Materials Letters 62 (2008) 1677–1680

In the light of this, this paper deals with the catalytic properties of RE-doped SnO2 nanopowders involved in methanol oxidation. As mentioned above, some studies have already investigated REdoped SnO2 systems. However they have usually treated RE dopants as a family; i.e., they focus on the similarities (and not the differences) brought about by RE ions. Here we compare not only the similarities but also the differences in terms of the morphological aspects and catalytic activity of La- or Ce-doped SnO2 powders. The main objective is to better understand the factors that affect the catalytic performance of SnO2. 2. Experimental details 2.1. Preparation of samples A modified polymeric precursor route was used to prepare the samples. This method consists of the preparation of a polymeric resin containing randomly dispersed cations, obtained by the reaction of a chelate and a polyalcohol [11]. The polymeric resin is then pyrolized and the resulting material heat-treated. In this study two series (of Ce- and La-doped SnO2 powders respectively) containing 0.1, 1 and 5 mol% of dopant were prepared.

First, SnCl2 was added to an aqueous solution of citric acid ([citric acid] = 2.5 M) to form the tin citrate. The ratio between tin and citric acid was set at 3:1, in mol. The citrate was precipitated by the slow addition of NH4OH (pH ≤ 3) and filtered. Ethylene glycol (citric acid/ethylene glycol, ratio = 60/40 wt.%) and the dopant citrates (obtained by adding a soluble salt, such as a nitrate or a carbonate, to the aqueous citric acid solution) were added to the tin citrate. The solution was homogenized by magnetic stirring at 70 °C and polymerization occurred at around 120 °C. The heattreatment was carried out in two stages: (i) pyrolisis at 350 °C for 4 h and then (ii) crystallization at different temperatures for 2 h under oxygen flow. Material characterization was performed using X-ray diffraction (XRD), BET curves and Transmission Electronic Microscopy (TEM). The XRD analysis (using a Siemens D-5000 Diffractometer) was performed in a θ–2θ configuration (from 20 to 60°), using 0.030 °C as the interval for each step. Crystallite sizes were estimated on the (110) peak by Scherrer's equation. 2.2. Chromatographic assays Chromatographic assays were conducted with the effluents of a reactional line to evaluate the catalytic properties of pure, La-

Fig. 1. XRD patterns of (a) 5 mol% Ce-doped and (b) 5 mol% La-doped samples, treated at different temperatures for 2 h.

I.T. Weber et al. / Materials Letters 62 (2008) 1677–1680

Fig. 2. Specific surface area obtained using BET curves — a) Ce- and b) Ladoped samples.

and Ce-doped samples. Samples that had been heat-treated at 550 °C were selected to perform these assays. Methanol vapor was obtained by saturating the carrying gas (synthetic air) at low temperatures. Porapack Q and molecular sieve separation columns were used for the chromatograph, the former to analyze carbon dioxide and water and the latter, used with a trap to retain water, and methanol to analyze hydrogen (H2), methane and carbon monoxide. The injection conditions were set at: initial column temperature= 80 °C, final column temperature= 110 °C, and heating rate = 5 °C/min. The chromatograph was operated using a 60 mA current and a 0.67 L/s gas flow. Before each assay, a blank curve was obtained to identify the effects of the thermal oxidation of methanol. Commercial grade O2, methanol, hydrogen, carbon monoxide and methane were used to correct the experimental areas of the chromatograms.

1679

dopant can be observed. The addition of a small amount of dopant caused the doped system to show larger crystallite sizes and lower surface area compared to the pure system. Though, the crystallite size remains almost constant throughout the temperature range. On the other hand, the addition of 5 mol% of dopant produced the opposite effect, particularly at moderately high temperatures. This dependence is more pronounced in the samples containing La. One hypothesis to be considered to explain this effect is that the dopant affects the mobility of grain boundaries in different ways, depending on its concentration [12,13]. Small amounts of dopant may provide conditions favoring boundary mobility and, hence, grain growth. On the other hand, higher dopant concentrations may lead to supersaturation [12,14], which is responsible for hindering grain growth. Fig. 3 shows several TEM images, from which it can be observed that agglomerates are formed by a superposition and junction of particles. The TEM images are consistent with the XRD crystallite size data: the pure sample heat-treated at 550 °C showed particles of about 14.9 ± 4.6 nm, while Ce- and La-doped samples had values of 12.0 ± 3.9 and 13.7 ± 3.7 nm, respectively. The catalytic assays showed that samples doped with up to 1 mol% of Ce or La displayed a curve profile quite similar to the pure system, indicating that, up to this concentration, the effect of the dopant was negligible despite all differences in morphological characteristics. Higher dopant concentrations are therefore required to enhance catalytic performance. Fig. 4 shows the pattern of methanol consumption in the presence of 5 mol% doped samples. It was observed that Ce-doped sample displayed high catalytic activity throughout the temperature range. In fact, the Ce-doped samples seemed to exhibit high catalytic activity even at temperatures lower than 200 °C. In contrast, pure and La-doped samples required higher temperatures to achieve effective methanol consumption. The pure sample displayed an almost linear pattern, rising continuously and becoming more pronounced at temperatures above 300 °C. The La-doped sample behaved in a specific manner, i.e., insignificant catalytic activity up to 300 °C and a change at 350 °C, when it showed powerful catalytic activity. Similar results were observed by Jinkawa et al. [2], who found that SnO2 loaded with La2O3 produced a stepped ethanol conversion curve. These profiles indicate that the influence of the dopant goes far beyond a simple morphological modification. If only morphological factors were taken into account, the La-doped sample, which possessed the largest surface area, should display the most powerful catalytic activity. Instead, the Ce-doped sample was observed to have the most powerful catalytic activity, probably due to the contribution of the Ce ions to the catalytic process. This kind of catalytic superposition effect has already been observed in SnO2:Ce [15] and SnO2·Al2O3 [10] systems. The La was not found to have any catalyzing effect by itself. Moreover, it is believed that alcohol molecules adsorpt onto an electron-deficient center (such as Sn+ 4) via OH dipoles. Hence, alcohol adsorption should be disfavored when Sn+ 4 ions are replaced by La3+ ions. The presence of Ce4+ ions does not alter the distribution of electrons, and consequently, the alcohol adsorption.

3. Results and discussion Three types of samples were studied here: pure and La- and Ce-doped SnO2 nanopowders. According to the XRD data, up to 950 °C only the casseterite phase was observed. Figs. 1 and 2 and Table 1 show the crystallite size and specific surface area for samples as a function of crystallization temperature. Ce- and La-doped samples present a quite similar BET curve profile, suggesting that the nature of the ion does not alter the characteristics of particles. However, for both curves, a dependence on the concentration of

Table 1 Mean crystallite size calculated using Scherrer's equation Temperature (°C)

Crystallite size (nm) Pure SnO2

0.1% Ce

1% Ce

5% Ce

0.1% La

1% La

5% La

350 550 750

3.9 8.0 13.7

8.9 9.0 9.1

7.3 7.3 7.4

4.7 7.2 8.3

13.3 14.0 14.6

6.6 8.2 10.4

2.4 6.6 7.8

1680

I.T. Weber et al. / Materials Letters 62 (2008) 1677–1680

Fig. 3. TEM images of pure (left), 5 mol% Ce-doped (center) and 5 mol% La-doped (right) samples heat-treated at 550 °C.

Another point to be taken in account is the possibility of a second phase segregation. It is known that Ce-doped samples produce a separate second phase (CeO2) more easily than La-doped ones (Sn2La2O7) [14,16]. Although this second phase cannot be detected, it is possible that small amounts of ceranite phase begin to segregate on heat-treatment. These spots of ceranite can act as adsorption centers or centers of high performance catalysis (due to the catalytic effect of Ce), which can assist reactions occurring at low temperatures [17]. So far as oxidation products are concerned, carbon dioxide and water are the main products, indicating that, under these conditions, full oxidation of the methanol occurs. These results concur with the proposal of Jinkawa et al. [2], who predicted that alcohol decomposition catalyzed by semiconductors gives rise to a chain reaction, producing carbon dioxide and water, irrespective of the route taken (dehydrogenation or dehydration).

4. Conclusions The catalytic tests carried out using methanol vapor revealed that at concentrations of up to 1 mol%, the dopants studied here did not affect the catalytic process, showing curve profiles quite similar to those of the pure system. On the other hand, increasing the dopant concentration to 5 mol% led to quite marked effects. These effects resulted not only from morphological changes brought about by the dopants, but also have been associated with

Fig. 4. Methanol consumption in presence of 5 mol% doped catalysts.

the onset of a second phase (still undetected) or with the combination of the catalytic properties of the cations. A complex combination of all of these factors is also a likely possibility. Acknowledgements We are grateful to the following Brazilian funding agencies: FAPESP, CNPq and FACEPE. References [1] D. Wang, J. Jin, D. Xia, Q. Ye, J. Long, Sensors and Actuators B 66 (2000) 260. [2] T. Jinkawa, G. Sakai, J. Tamaki, N. Miura, N. Yamazoe, Journal of Molecular Catalysis A 155 (2000) 193. [3] H. Teterycz, H. Licznerski, H. Nitsch, K.K. Wisniewski, J.L. Golonka, Sensors and Actuators B 47 (1998) 153. [4] K. Fukui, S. Nishida, Sensors and Actuators B 45 (1997) 101. [5] N.L.V. Carreño, H.V. Fajardo, A.P. Maciel, A. Valentini, F.M. Pontes, L.F.D. Probst, E.R. Leite, E. Longo, Journal of Molecular Catalysis A 207 (2004) 89. [6] N. Yamazoe, Sensors and Actuators B 5 (1991) 7. [7] K. Fukui, Sensors and Actuators B 24-25 (1995) 486. [8] J.S. Reed, Principles of Ceramics Processing, John Willey & Sons, New York, 1995. [9] J.E. Huheey, E.A. Keiter, R.L. Keiter, Inorganic Chemistry, Harper Collins College Publishers, New York, 1993. [10] M. Ivanovskaya, P. Bogdanov, G. Faglia, P. Nelli, G. Sberveglieri, A. Taroni, Sensors and Actuators B 77 (2001) 268. [11] Pechini, M.; U.S. Pat. 1967, No. 3330697. [12] E.R. Leite, A.P. Maciel, I.T. Weber, P.N. Lisboa-Filho, E. Longo, C.O. PaivaSantos, C.A. Paskoscimas, Y. Maniette, H. Schreiner, Advanced Materials 14 (12) (2002) 905. [13] R.J. Brook, in: F.F.Y. Wang (Ed.), Ceramic Fabrication Process, vol. 9, Academic Press, New York, 1976, Ch. 17. [14] A.P. Maciel, P.N. Lisboa-Filho, E.R. Leite, C.O. Paiva-Santos, W.H. Schreiner, Y. Maniette, E. Longo, Journal of the European Ceramic Society 23 (2003) 707. [15] F.M. Gonçalvez, P.R.S. Medeiros, L.G. Appel, Applied Catalysis A 208 (2001) 265. [16] I.T. Weber, A.P. Maciel, P.N. Lisboa-Filho, C.O. Paiva-Santos, W.H. Schreider, Y. Maniette, E.R. Leite, E. Longo, Nanoletters 2 (2002) 969. [17] A. Vazquez, T. Lopez, R. Gomez, X. Bokhimi, Journal of Molecular Catalysis A 167 (2001) 91.