(113e) PRODUCTION of HYDROGEN and CARBON Nanotubes by CATALYTIC Decomposition of Methane USING Co-Mg-Al Catalysts

Introduction

It is considered that hydrogen will become an efficient renewable and sustainable energy carrier for the future [1]. Hydrogen is a clean fuel that emits no CO₂ when burned or used in H₂-O₂ fuel cells, can be stored as a liquid or gas, is distributed via pipelines, and has been described as a long-term replacement for natural gas [1].

Hydrogen is usually produced by steam reforming, partial oxidation or autothermal reforming of hydrocarbons and alcohols. These processes are based on mature technologies, but are quite complex involving many steps. In addition, a common problem of these processes is the formation of carbon monoxide and carbon dioxide, which are difficult to separate from hydrogen. Furthermore, for certain applications such as in polymer electrolyte membrane (PEM) fuel cells, carbon monoxide has to be removed from hydrogen in order to avoid poisoning of the electrocatalyst, increasing the costs of the traditional production routes for hydrogen.

The catalytic decomposition of methane (CDM) has been proposed in recent years as an interesting alternative route for the production of hydrogen. This process is less energy consuming than steam reforming of methane, and does not directly produce CO and CO₂ [2]. Consequently, additional steps such as water gas shift, WGS, and preferential oxidation of CO are not necessary, and this considerably simplifies the process and decreases the costs of production.

Moreover, during the CMD, nanocarbonaceous materials (NCMs) are produced such as carbon nanotubes (CNTs) and carbon nanofibres (CNFs). Since the discovery of carbon nanotubes [3], they have received considerable attention due to their excellent physical and chemical properties which allows their use in a large number of potential industrial applications. As regards of the catalysts used for the CDM process, Fe, Co, Ni and their alloys are the active metals usually used in the hydrocarbon decomposition [4]. The role of the catalyst is crucial for obtaining high activity and, in particular, selectivity towards the desired NCMs [5]. Our research group has worked on the development of Ni based catalysts, using hydrotalcite-like compounds as precursors, for the application in the CDM process [2,5,6]. We have found that the modification of catalyst composition, by insertion of MgO to obtain Ni-Mg-Al catalysts, notably increases their productivity and stability during the CDM [6]. However, the morphology of the carbon formed is mainly carbon CNFs. In this case, we study the use of the Co, in a Co-Mg-Al catalyst, as the active phase with the aim of improving the selectivity to CNTs. The influence of the activation, including calcination and reduction temperatures, and operating conditions, on hydrogen production and carbon formation has been studied. In particular, here we present the effects of the calcination temperature because strongly modifies the interactions developed between the metallic nanoparticles and the support, and, in consequence, its reducibility and reactivity.

Experimental

The catalyst, with a nominal composition (CoO)(MgO)(Al₂O₃), was prepared by coprecipitation of the metallic nitrate mixture (Co, Mg and Al) with Na₂CO₃/NaOH, at constant pH (10.2 ± 0.2) and temperature (60º C). The corresponding mixed oxide was obtained by calcination of the dried hydrated precursor in N₂ at different calcination temperatures (from 400 to 800ºC) for 11 h. The catalyst was characterized by several techniques: atomic absorption (AA), X-ray diffraction (XRD), temperature programmed reduction experiments (TPR) and specific surface area.

CMD reaction was carried out in gas phase by using a thermobalance (CI Electronics Ltd., UK, model MK2) equipped with mass flow and temperature controllers and operated as a differential reactor. This experimental system allows continuous recording of the sample weight and temperature during reaction. After reaction, the catalyst, as well as the carbon obtained, were characterised by TEM, XRD and TPO.

Results Influence of the calcination temperature

Figure 1 shows the XRD patters of the catalyst after calcination at different temperatures ranged between 400 and 800 ºC. As can be deduced from the sharpening of the peaks corresponding to the detected phases, the higher is the calcination temperature, the greater is the crystallinity of the solid. However, at the same time the specific surface area of the solid, which will ensure an adequate dispersion of the active metal, and therefore a good activity, decreases, from 264 m²/g at 400 ºC to 83 m²/g at 800 ºC, as the calcination temperature increases.

After calcination at low temperature, the solid is formed for a mixture of metallic oxides (CoO, MgO).  At higher temperatures, the XRD patterns show the presence of  peaks corresponding to a spinel-like phase (CoAl₂O₄ /MgAl₂O₄) in addition to the oxides.

In this point, it is noteworthy to remember that a that a partial change, from octahedral to tetrahedral, in the coordination of bulk M⁺³ cations can occur depending on the activation conditions [7,8]. This transformation converts the so-called ?normal spinel? to a partially ?inverted spinel?. At low calcination temperatures, M⁺² cations engage the surface octahedral hollows of the spinel, which have larger coordination index, better accessibility to the reactants, and therefore will be more active [7, 8].

Figure 2 shows the effect of calcination temperature on hydrogen formation rate, and on the amount of CNTs accumulated over the catalyst, that is a direct measurement of the H₂ produced. These experiments have been carried under dynamic conditions 5º C/min from 400 to 800 ºC in order to discriminate simultaneously the effects of the reaction temperature and of the operating time. For this. The catalyst was previously reduced ?in-situ? at 750ºC for 2 hours.

As Figure 2 indicate, the quantity of CNTs concentration, and the range of temperatures where the catalyst is active, diminishes as the calcination temperature increases. Thus, a lower amount of reduced metallic Co is obtained after the reduction, when the catalyst has been calcined at high temperatures. This is due to the high stability and low metallic dispersion of the spinel structure. Besides, at high calcination temperatures a large incorporation of the Co inside the spinel network is produced, due to the migration of the Co from octahedral to tetrahedral positions, which are much less active [7,8]. In addition, the formation of a major proportion of deactivating encapsulating carbon, is favored by large sized metallic Co particles, due to the difficulty in the diffusion through that big particles [9]. In consequence, the deactivation rate of the catalyst calcined at elevated temperatures is higher.

TEM micrographs corroborate these results. Thus, the carbonaceous deposits on the catalyst calcined at 800ºC contain mainly amorphous carbon, and a small proportion of CNFs (Figure 3-Left). Metal agglomerations also can be observed owing to the sintering caused by the high calcination temperatures. However, after calcination at 400ºC (Figure 9-Right), the NCMs are formed mainly for multiwalled CNTs (MWNT), with diameters from 8 to 10 nm and several microns of length. The proportion of deactivating amorphous carbon is low; in consequence, the yield attained along the experiment is higher (Figure 2). In accordance, metallic sintering is less severe at this calcination temperature (Figure 3-Right).

In conclusion, the activity, selectivity and resistance against deactivation of Co-Mg-Al catalysts prepared by coprecipitation strongly depend on the activation conditions, in particular on the calcination temperature. This result opens a route to optimize the catalyst performance in the simultaneous production of H₂ and CNTs by CMD.

Acknowledgements

The authors acknowledge financial support from MICINN (Spain)-FEDER, Project CTQ 2007-62545/PPQ, and the Regional Government of Aragón, Departamento de Ciencia, Tecnología y Universidad, Project CTP P02/08.

References

[1]  J.N. Armor, Appl. Catal. A: General 176 (1999) 159.

[2]  J.I. Villacampa, C. Royo, E. Romeo, J.A. Montoya, P. Del Angel and A. Monzón, Appl. Catal. A: General 252 (2003) 363.

[3] S. Ijima, Nature 354 (1991) 56.

[4] S.B. Sinnott, R. Andrews, D. Qian, A.M. Rao, Z. Mao, E.C. Dickey and F. Derbyshire, Chem. Phys. Lett. 315 (1999) 25.

[5]  M. Monthioux, L. Noé, L. Dussault, J.C. Dupin, N. Latorre, T. Ubieto, E. Romeo, C. Royo, A. Monzón and C. Guimon, J. Mater. Chem. 17 (2007) 4611.

[6] N. Latorre, J.I. Villacampa, T. Ubieto, E.Romeo, C. Royo, A. Borgna and A. Monzón, Topics Catal., 51 (2008) 158.

[7] Trifirò, F.: Vaccari, A.; Stud. Surf. Sci. and Catal., 67 (1991) 157.

[8] Jacobs, J.P.; A. Maltha, A.; Reintjes, J.G.H. ; Drimal, J.; Ponec V.; and Brongersma, H.H.,  J. Catal., 147 (1994) 294.

[9] De Chen, Kjersti O. Christensen,  E. Ochoa-Fernández,  Z. Yu, B, Tøtdal, N. Latorre, A. Monzón and A. Holmen, J. Catal.,  229 (2005) 82.