Motivated by the population growth, climate change and limited fossil fuel resources, renewable alternatives for fuels and chemicals production are becoming more and more important. Biomass, especially residual lignocellulosic biomass shows a significant potential as feedstock for bioenergy, due to its high carbon content and short-term availability. Among the thermochemical conversion technologies, fast pyrolysis for biomass liquefaction can be considered already well stablished, as several commercial plants are spread worldwide. However, fast pyrolysis bio-oil, the main product of fast pyrolysis, currently shows limited bioenergy application as boiler fuel for heat production. It can be explained by its chemical composition and properties, as fast pyrolysis bio-oil is an acidic multi-component product, with low energetic density due to its high content of water and oxygenated compounds. Moreover, wood is the only feedstock currently used commercially.
In order to expand the feedstock range and application viability, an additional upgrading treatment may be required in order to improve the fast pyrolysis properties, meeting existing fuel standards. ... mehrIn order to do so, catalytic hydrotreatment is considered a promising upgrading treatment, as it is a well-known technology currently applied in petroleum refineries for heteroatoms removal from crude oil. However, due to the differences in chemical composition, the hydrotreatment conditions applied to crude oil cannot be simply applied to fast pyrolysis bio oil. Although research in this field has been carried out for a few decades, there are still open questions to enable hydrotreatment to produce fuel oils from residual biomass in stable processes. By developing a robust fast pyrolysis bio-oil hydrotreatment process, small biorefineries units could be installed near to feedstock sourcing or even be installed in biorefinery units already stablished, such as a sugarcane biorefinery, in which high volumes of residual biomass are generated. Also, co-processing of crude oil and fast pyrolysis bio-oil in petroleum refineries may be a feasible option.
In view of the importance of the hydrotreatment for expansion of the range of chemicals obtained by thermochemical conversion of residual biomass, the presented work investigated the hydrotreatment of fast pyrolysis bio-oil applying nickel-based catalysts. In a systematic evaluation nickel-based catalysts with different metal loading, supports and promoters have been studied. Overall, six nickel-based catalyst were screened and compared to ruthenium supported in activated carbon. The hydrotreatment conditions in terms of reaction time, temperature and pressure were optimized and fast pyrolysis bio-oils derived from beech wood and residual biomass (sugarcane bagasse) were hydrotreated. Additionally, the heavy phase separated from beech wood bio-oil, characterized by its high content of lignin-derived compounds, was hydrotreated. The effect of deactivation by sulphur on the hydrotreatment was investigated by use of model substances in a continuously operated trickle bed reactor, since with this reactor the deactivation can be observed depending on time (in contrast to batch experiments). Finally, a 2 step upgrading approach of a previously upgraded fast pyrolysis bio-oil was proposed and verified.
Initially two high loaded nickel-based catalysts (monometallic nickel and nickel chromium) were evaluated in comparison to Ru/C by batch hydrotreatment of beech wood bio-oil at 80 bar, 4 h, 175 °C and 225 °C. Both nickel-based catalysts revealed similar hydrodeoxygenation activities for the conditions applied and the nickel catalysts showed the higher hydrogenation activity compared to Ru/C. The nickel-chromium catalyst demonstrated the highest activity for conversion of organic acids, ketones and sugars, attributed to the strength of the acid sites promoted by chromium oxide. When applied in a second hydrotreatment step of a previously upgraded oil, the oxygen content of the oil was reduced by 64.8 % in comparison to the original feedstock while the water concentration was reduced by 90 %. Nearly 96 % of the organic acids were converted and the higher heating value was increased by 90.1 %. Despite nickel-chromium demonstrated the best activity in the one step hydrotreatment reactions and contributed significantly in the 2-step upgrading, the oxygen content of 25.3 wt.% dry basis in the upgraded oil was still considered high. Thus, the upgrading conditions were further optimized, aiming to achieve higher hydrodeoxygenation performance.
The conditions of batch hydrotreatment were optimized with nickel-chromium catalyst considering two pressures (80 and 100 bar), four temperatures (175 °C, 225 °C, 275 °C and 325 °C), for both the complete beech wood fast pyrolysis bio-oil, as well as for the heavy phase after spontaneous separation induced by intentional ageing of the bio-oil. At higher temperatures, increased hydrodeoxygenation levels were reached, while at higher pressure larger hydrogen consumption was observed with no significant influence on hydrodeoxygenation. The best conditions among all tested was obtained by hydrotreating the beech wood bio-oil at 325 °C and 80 bar; in this case, 43 % of hydrodeoxygenation was reached. Although improved hydrodeoxygenation activity observed with nickel-chromium at optimized conditions, the results motivated the synthesis and evaluation of new nickel-based catalysts, targeting higher deoxygenation levels.
In the next part of this study, four nickel-based catalyst were synthesized by wet impregnation and evaluated for the hydrotreatment of beech wood fast pyrolysis bio-oil. The catalysts were supported in silica and zirconia and the influence of copper as promoter was studied. Among them, nickel-silica was the most active for hydrodeoxygenation, reducing the oxygen content of the upgraded beech wood fast pyrolysis bio-oil by more than 50 %. The highest degree of water removal as well as low gas and char production were also considered good properties attributed to this catalyst. The investigation on repeated cycles of hydrotreatment with the same catalyst showed a remaining activity even after the fourth reuse, in which 43 % of oxygen was removed. Thus, based on the results obtained with Ni/SiO2, this catalyst was selected together with nickel-chromium catalyst to be used for hydrotreatment of fast pyrolysis bio-oil from residual biomass, as until this point the study had considered only wood-based fast pyrolysis bio oil.
Based on the studies so far, the integration of hydrotreatment into a thermochemical conversion route of residues in a sugarcane refinery was proposed. For that, the study encompassed sugarcane bagasse characterization, fast pyrolysis and hydrotreatment of the so derived bio-oils with nickel-chromium and nickel-silica catalyst. The detailed investigation of the bagasse and the fast pyrolysis bio-oil compositions allowed the correlation of the biomass building blocks with the monomers obtained. The hydrotreatment showed that nickel-chromium showed highest activity for organic acids conversion, as previously observed with beech wood bio-oil, whereas nickel-silica revealed more active for conversion of aromatics. Hydrodeoxygenation of 43.3 % was obtained with nickel-silica. Although both catalysts demonstrated to be active at the conditions evaluated, the high viscosities of the upgraded oils in comparison to those obtained from fast pyrolysis showed that polymerization took place and must be further investigated in detail, as it is one of the limiting factors for further application of fast pyrolysis bio-oil hydrotreatment. Overall, this studied showed to be very promising and future studies are planned.
In the final part of the thesis, both high loaded nickel-based catalysts studied in the first chapters were selected for a detailed investigation in a continuous operated tricked bed hydrotreatment reactor, due to the similar nickel concentration, nickel particle size and support. The selection of both catalysts aimed to investigate the influence of sulfur on long term catalyst deactivation and the role of chromium in catalyst deactivation. Both catalysts were active for conversion of model substances over more than 48 h of reaction time. By the presence of sulfur, the selectivity of both catalysts changed, mainly towards alkene formation, while the activity remained in the same range. Formation of Ni3S2 was observed for both catalysts, but the highest intensity in the diffraction peak of metallic nickel in the nickel-chromium catalyst might be an indication of higher resistance to sulfur poisoning in comparison to Ni catalyst. In general, the catalysts were active for the conditions tested, although the hydrogenation activity was compromised by sulfur poisoning.
Overall, all the catalysts tested in this study were active for hydrotreatment of fast pyrolysis bio-oils. If only stabilization of reactive compounds such as aldehydes and furfurals is required, all of them could be considered suitable candidates. In terms of hydrodeoxygenation activity, Ni/SiO2 showed the highest performance, while nickel-chromium showed to be the most active for conversion of organic acids and superior hydrogenation capacity than Ni/SiO2.
