Hydrogen production through steam reforming of model bio-oil aqueous fraction using metal-supported mesoporous catalysts

Resumen

The work presented in this Doctoral Thesis is part of the hydrogen production and storage line and has been developed in the Rey Juan Carlos University, specifically, in the Chemical and Environmental Engineering Group laboratories. In particular, it has been focused on the production of renewable hydrogen from model compounds present in the aqueous fractions obtained from thermochemical treatments of biomass and residual waste. For this purpose, cobalt supported catalysts have been developed in order to maximize the production of hydrogen, minimizing the formation of secondary products such as coke, which decrease the catalyst activity. World demand for primary energy rises by 1.3 % each year as a consequence of global economic growth, the increase in population and advances in technology. Currently, more than 90 % of the world's energy demand is supplied by fossil fuels, emitting approximately 32 billion tons of CO2 into the atmosphere each year. In addition to the strong environmental impacts, fossil fuels are ever-dwindling supplies, and oil prices wildly fluctuate, affecting the industry and the ability of consumers to buy goods and services. Therefore, the number of countries with energy policies based on hydrogen technologies is increasing. The use of hydrogen for energy production does not generate polluting emissions, since only heat and water vapor are produced. Likewise, the application of hydrogen in fuel cells allows the production of sustainable electricity with high yields. Therefore, it has the potential to significantly reduce CO2 emissions, contributing to limiting global climate change. Despite all these advantages, hydrogen is not available in free form in nature, reason why, currently hydrogen is produced mainly from the steam reforming of natural gas. Consequently, the application of hydrogen as the energy vector of the future, requires the development of new technologies that allow its production from renewable sources. Among the different processes for the production of renewable hydrogen, the steam reforming of the bio-oil aqueous fraction obtained from thermochemical treatments of biomass and wastes becomes interesting. In general, the steam reforming of the oxygenated compounds contained in the bio-oil aqueous fractions is a complex process that involves a large number of reactions. Hence, the hydrogen yield is affected by the development of secondary reactions such as methanation and coke formation reactions, in addition to the thermodynamic limitation of the reforming reaction and the water-gas shift reaction. Thus, the viability of this process is related to the development of stable and active catalysts, with high selectivity towards hydrogen as the product of interest. Therefore, the main objective of this research work is the development of cobalt-based catalysts supported on the mesoporous SBA-15 material, which allows a better dispersion of the active phase. To date, Co catalysts have been less studied than Ni catalysts despite they provide high activity at moderate temperatures and increase the hydrogen yield. In addition, in this Thesis, Co/SBA-15 has been modified using several promoters, to improve their catalytic performance in the steam reforming of both: acetic acid, as a model compound of the aqueous phase obtained from thermal treatments of biomass (chapters 1-3), and, a mixture of model compounds (acetic acid, hydroxyacetone and phenol), that represent an approximation to a real aqueous fraction in the last chapter. This Doctoral Thesis has been divided into four different chapters, the most relevant results of which are summarized below: I. Effect of the incorporation of reducibility promoters on Co/CaSBA 15. In the first part of this chapter, the effect of SBA-15 modification with Ca before the Co impregnation is studied since according to the literature, calcium increases the metal-support interaction and provides basicity to the catalyst. The characterization results showed how Ca incorporation improved the metal dispersion at the expense of the reducibility as demonstrated by H2-TPR and XPS. The lower reducibility of Co/CaSBA-15 led to lower conversion and hydrogen yield compared to the unpromoted sample. To address this detrimental effect, in the second part of this chapter, the addition of reducibility promoters such as Cu, Ag and Ce was perform leading to an enhancement of the catalytic performance increasing the acetic acid conversion. However, the presence of Cu did not improve the hydrogen yield, attributed to its predominant role in the decarboxylation reaction of acetic acid rather than in the reforming reaction. As a consequence of the high dispersion and good reducibility (comparable to that obtained for Co/SBA-15), Co-Ce/CaSBA-15 sample reached the best catalytic behavior in the acetic acid steam reforming increasing the conversion and the hydrogen yield up to 99 % and 71.8 % respectively. That means, a 20 % increase for the conversion and 62 % for the hydrogen yield, compared to Co/CaSBA-15 catalyst. II. Acetic acid steam reforming using bimetallic Co-M/SBA-15 (M: Cu, Ag, Cr, Ce). Since the incorporation of calcium in the SBA-15 support implies an additional stage in the catalyst preparation and based on the results obtained in the previous chapter where Ca addition did not improve the results obtained with Co/SBA-15, we decided to continue using bare SBA-15 as support. In this chapter, the influence of promoters (Cu, Ag, Ce and Cr) addition on the physicochemical properties of Co/SBA-15 catalyst and, on the catalytic activity of acetic acid steam reforming is studied. The results showed how Cu and Ag addition led to a decrease in the Co reduction temperature. In the case of the Cu-doped catalyst, the Co crystal size was maintained when compared to the Co/SBA-15 catalyst, while, in the case of the Ag-containing catalyst, the Co crystal size was increased. The addition of Ce does not affect significantly neither Co oxides reducibility nor Co crystallites size. For its part, the addition of Cr considerably reduced the Co crystallite size, giving rise to higher dispersion of the active phase over the support. Catalytic results showed how Co-Cu/SBA-15, Co Ag/SBA-15 and Co-Ce/SBA-15 reduced the hydrogen selectivity. However, the addition of Cr gave rise to an improvement in the catalytic behavior, leading to the highest hydrogen selectivity of the prepared catalysts with values around 72 % that are very close to the thermodynamic equilibrium. In addition, a lower coke deposition was observed in this catalyst with a greater presence of defective carbon nanofilaments. III. Preparation of agglomerated Co-Cr/SBA-15 catalysts for hydrogen production through acetic acid steam reforming. This chapter is devoted to the optimization of the agglomeration process of Co Cr/SBA-15. A series of Co-Cr/SBA-15 extrudates were prepared by varying the bentonite content, used as binder, and particle size in order to get catalyst particles suitable to be used in a steam reformer at industrial scale. The characterization results did not show remarkable changes in the physicochemical properties after the extruding process of the powdered sample, while the mechanical strength increases with the binder content. The external and internal diffusion tests allowed the selection of the suitable flow rate and catalysts particle size to avoid any concentration gradients. Under these experimental conditions, Co-Cr/SBA-15 extrudates with a 30 wt.% of bentonite and a particle size with an effective diameter of 1.5 mm, showed a good catalytic performance in the acetic acid steam reforming, reaching a hydrogen yield around 60 % and similar results in terms of conversion when compared to the powder form. IV. Study of coke formation in the steam reforming of bio-oil aqueous fractions. With the aim of studying the causes of deactivation in the reforming reactions using Co/SBA-15, this chapter studies the evolution of the used catalyst at different times in the steam reforming of a mixture of model compounds that mostly represent the aqueous fraction obtained from biomass. Deactivation of Co/SBA-15 is a consequence of coke formation, its composition and Co sintering. The oxidation of Co was discarded as a cause of deactivation since no cobalt oxides were found by XRD in the used catalysts. Results evidenced that Co/SBA-15 is deactivated following a process with two consecutive stages. The first one, at short reaction times, ascribed to the formation of defective carbon nanofibers over the catalyst surface as well as inside the SBA-15 pores, leading to isolated metallic Co particles. During this step, some Co particles were also encapsulated within coke nanofibers, hindering the catalytic role of these active centers. During the second stage, coke nanofilaments became more ordered growing out of the support without covering new pores, as inferred from almost constant BET surface values. In both stages, Co crystallites size indicated a slight sintering with minor influence on Co/SBA-15 deactivation. These results, along with a t-Student analysis, have led to the conclusion that the progressive growth of carbon nanofibers and its nature (C/H ratio) have more influence on the conversion drop than the other studied variables.