Large-scale conversion of lignin to liquid through formic acid assisted solvolysis in aqueous and ethanolic reaction media: comparison of yields and product compositions
Not peer reviewed
MetadataShow full item record
Biomass has recently received much attention as an attractive renewable energy resource and a promising alternative to fossil carbon resources for production of renewable biofuels and other value-added chemicals due to being the only viable feedstock for carbon based fuels and chemicals. Within the biorefinery concept, sustainable use of biomass involves optimal exploitation of all fractions of the raw material to make products with high value. Production of 2nd generation bioethanol from the carbohydrate fraction of non-edible lignocellulosic biomass is already established and the technical feasibility of this process is well demonstrated. However, this process leaves significant amounts of lignin, a cross-linked amorphous copolymer of phenylpropane units with unique properties as by-product.
Lignin is the third most abundant biopolymer as well as the most important source of bio-based aromatics in nature, which accounts for 10–30 wt. % of the feedstock. Thus, the viability of lignocellulosic biorefinery is highly dependent on the development of efficient lignin valorisation routes. Production of value-added chemicals from lignin requires the simultaneous depolymerization of the lignin structures with subsequent hydrodeoxygenation of the lignin monomers and alkylation of aromatic rings to prevent repolymerization and char production. Thermochemical conversion of lignin through Lignin-to-Liquid (LtL) process is an innovative conversion method, which can be considered as a solvolytic process in a liquid or near-critical reaction medium at high temperature and high pressure, using an in situ hydrogen donor solvent instead of molecular hydrogen.
The Lignin-to-Liquid process and the chemical composition and bulk properties of LtL-oils produced in small laboratory scale is well developed and there is ongoing research on this approach. However, in case of development towards industrial scale production, the effect of increasing the scale must be investigated and the conversion must be optimized at larger scale. Optimizing process conditions yielding high amount of the desired products is challenging and time-consuming, especially due to the interactions between different experimental conditions. Thus, some important reaction parameters such as shorter reaction time, lower reaction temperature, and reduction of low-value side stream products, i.e. gas and solid residues, need to be improved in order to make LtL-oils competitive with petroleum-based fuels and chemicals.
The main focus in this thesis was therefore to evaluate the impact of upscaling on LtLprocess efficiency in terms of bio-oil yield and bio-oil composition. Lignin conversion conducted at small laboratory scale (0.025 L) was scaled up by a factor of 200 and reperformed using a 5 L stirred reactor to explore the effect of increased volume and stirred reaction on the product yield and product quality. Various reaction parameters were investigated and the relationship between the product yields and reaction conditions were systematically evaluated using principal component analysis (PCA). Additionally, the catalytic conversion of lignin through LtL-solvolysis was explored using two different types of catalysts, an alumina supported noble metal catalyst and an iron-based mineral catalyst.
The overall results showed similar trends relative to reaction parameters at both reaction scales, but oil yields in some cases tended to decrease from small laboratory scale to 5 L scale when using water as reaction medium. The purest lignin feedstocks resulted in highest oil yields at both scales. Comparison of the investigated solvent systems (water vs. ethanol system) showed that the highest oil yields from eucalyptus lignin-rich residue were achieved from the ethanol system at reaction temperatures below 350 °C, indicating a higher tendency for repolymerzation of lignin components to give char formation at elevated temperatures. In addition, a major increase in oil yield and a significantly decrease in char yield was observed as a function of increased stirring rate and increased level of loading in the reactor. Goethite as catalyst did not shown good conversion efficiency, while Ru/Al2O3 was found to be very efficient with oil yields above 69 wt. % on lignin intake. Overall, the highest bio-oil yield and a significant low char yield was obtained from experiment Ru/Al2O3.S1000.Max.305, indicating that combination of high stirring rate with maximum loading in the reactor in the presence of Ru/Al2O3 as catalyst at low temperature is the most optimal condition investigated in this thesis.
The bio-oil comprises a complex mixture of monomeric phenols, aromatics and more hydrogenated products, with a high H/C and a low O/C ratio. However, bio-oils from the ethanol system had higher H/C values due to the incorporation of the ethyl groups, which increased the number of alkyl units in the product. Based on results from GCMS analysis, there was no clear differences in the composition of LtL-oils from the same solvent system, while ethanol-based experiments generated bio-oils with a more complicated pattern of substitution than water-based experiments. However, concentration of the most abundant compounds identified in each solvent system showed to be mainly dependent on reaction temperature.
Furthermore, the large product volume made it possible to test fractionation ability of the produced bio-oils by means of solid phase extraction (SPE) where 65–92 wt. % of the bio-oils were separated and recovered as polarity-based fractions. The most volatile fractions were then identified using GC-MS analysis, which showed good perspectives for further development. Moreover, the lignin-enriched eucalyptus residue was investigated as feedstock in a comparative study between a direct one-step hydrodeoxygenation (HDO) and a 2-step hydrothermal liquefaction-hydrodeoxygenation (HTL-HDO) approach using Ru/C and Pd/C as catalysts in terms of product yields, quality and composition of the produced bio-oils. A general observation was that bio-oil yields decreased as a function of increased temperature, while the volatility of the bio-oils as well as total monomer yields increased with temperature. 2-step HTL-HDO significantly improved the total monomer yields while preventing char formation.
The identified compounds comprising the lignin-oils were classified as alkylphenolics, aromatics, naphthalenes, linear and cyclic alkanes, guaiacols, catechols and ketones. Alkylphenolics and aromatics were the main chemical groups identified. However, a significant increase in alkane formation was observed with increased temperature, which can be due to enhanced depolymerization and hydrogenation at more severe temperature conditions. In terms of interesting monomers, the preferred pathways were the 2-step HTL-HDO of LES using Ru/C at 410 °C, and the direct HDO of LES using Pd/C at 450 °C.
Overall, the results obtained in this project showed that increase in reaction volume was a promising option in terms of product yields and product composition, giving oil yields up to 79 wt. % on lignin intake. However, further studies to map out the optimal experimental conditions towards desired bio-oil yield and bio-oil quality as well as the development of appropriate fractionation methods to separate bio-oil components into fractions with similar chemical properties are the next steps needed to strength the biooil potential as a source for platform chemicals.