Estudio de tecnologías aplicables en biorrefinerías de materiales lignocelulósicos

  1. López Rodríguez, Mar
Dirixida por:
  1. Juan Carlos Parajó Liñares Director
  2. Valentín Santos Reyes Co-director

Universidade de defensa: Universidade de Vigo

Fecha de defensa: 08 de xuño de 2021

Tribunal:
  1. Carlos Martín Medina Presidente/a
  2. Remedios Yáñez Díaz Secretario/a
  3. Ana Soto Campos Vogal

Tipo: Tese

Resumo

There is a growing concern related to the utilization of fossil fuels, since the current economic development model, based on their massive use, is not sustainable indefinitely. There are a number of problems linked to the use of fossil fuels, for example their non-renewable character, environmental impact, price volatility and insecurity of supply. In order to solve these issues, it is necessary to develop clean and sustainable processes using renewable resources as raw materials. In this context, the vegetal biomass, which represents the major source of organic carbon on Earth, is a key resource for the sustainable manufacture of fuels, materials, chemicals and energy (Cherubini, 2010; Menon y Rao, 2012). Within the vegetal biomass, the lignocellulosic materials (LCM) stand out as potential feedstocks for the industry for a number of reasons, including abundance, sustainability and cost effectiveness (Zheng et al., 2015). LCM are mainly made up of lignin (a phenolic fraction) and polysaccharides (cellulose and hemicelluloses), and their use can be accomplished following the biorefinery concept. As indicated by Gaudino et al. (2019), a biorefinery (analogously to an oil refinery) converts all the components of a renewable source into marketable products and energy, aiming to replace the utilization of fossil resources with biomass. This concept is linked to the principles of green chemistry (Song y Han, 2015) and sustainable development. The large scale implementation of biorefineries is expect to accelerate the transition to a sustainable development, through strategies based on the circular bioeconomy (European Commission, 2012, 2015), meeting one of the objectives of the European Union. Lignocellulosic biorefineries are mainly based on the fractionation of biomass into different streams (containing hemicelluloses, cellulose and lignin, or products derived from them). The corresponding fractions can be employed as substrates for manufacturing a wide range of bio-based products (including chemicals, materials, and fuels) and heat or power (van Ree y Annevelink, 2007). In addition, biorefineries must be sustainable, as specified in the biorefinery definition given by the International Energy Agency (International Energy Agency, 2014). The main objective of this PhD Thesis is to obtain value-added products from the different fractions of wood by chemical treatments using clean technologies and sustainable reaction technologies, according to the biorefinery approach. This study makes part of the research on biorefineries developed by the group EQ-2 Biomass and Sustainable Development, which belongs to the Department of Chemical Engineering of the University of Vigo (Campus of Ourense). Specifically, the experimental work corresponds to experimental tasks defined in two Research Projects funded by the “Ministry of Economy and Competitiveness” of Spain, entitled “Advanced processing technologies for biorefineries” (reference CTQ2014-53461-R) and “Modified aqueous media for wood biorefineries” (reference CTQ2017-82962-R). Abundance and geographical distribution of forest resources According to the data presented by the FAO (Food and Agriculture Organization of the United Nations) (2020) in the “Global Forest Resources Assessment 2020 - Key findings”, the total forest area in the world is 4.06 billion ha, corresponding to the 31% of the total Earth’s area. Forests are not geographically equally distributed, and more than half of the world's forests are concentrated in five countries: the Russian Federation (20%), Brazil (12%), Canada (9%), the United States of America (8%), and China (5%). Moreover, 93% of the forest area worldwide corresponds to natural forests, while the remaining 7% is planted forest (FAO, 2020). In the European Union (EU-28) there were approximately 182 million ha of forests and other wooded land in 2015, among which 161 corresponded to forests (Bioenergy Europe, 2019b). In Spain, the total forest area (including forests and other wooded land) reported in 2018 was 28 million ha, of which 18.5 corresponded to forests (Ministerio para la Transición Ecológica, 2020). In Galicia, the total forest area (about 2 million ha) accounts for almost the 69% of the land area (Xunta de Galicia, 2018). Of the total forest area (including forests and other wooded land) more than 70% are forests (Ministerio de Agricultura, Pesca y Alimentación, 2019). Eucalyptus globulus and Pinus pinaster woods were selected as raw materials in this study due to their high availability and easy accessibility. The wood samples employed in experiments were provided by local industries. Composition of lignocellulosic materials LCM are mainly composed of polysaccharides (cellulose and hemicellulose) and an aromatic fraction (lignin). LCM also contain minor amounts of other components (such as extractives, ash and proteins), which are not considered in this study. While cellulose has a structural function within the cell wall, lignin provides cohesiveness and rigidity, and exerts a protecting role against pathogens and insects. Hemicelluloses serve as a link between the cellulose fibers and lignin, and give cohesion to the whole network (Plomion et al., 2001; Sticklen, 2008; Barakat et al., 2013; Steinbach et al., 2017). Cellulose is a linear polymer made up of D-glucose units linked by β-(1→4) glycosidic bonds. Hemicelluloses are branched polymers with backbones made up mainly of pentoses (xylose) and/or hexoses (glucose, mannose and galactose), substituted with other sugars (arabinose), uronic acids and/or deoxyhexoses (Gírio et al., 2010). Lignin is a three-dimensional, amorphous heteropolymer made up of structural units (guayacil, syringyl and p-hydroxyphenyl) derived from coniferyl alcohol, sinapyl alcohol, and p-coumaryl alcohol, respectively (Barakat et al., 2013). The selective separation of cellulose, hemicelluloses and lignin allows the manufacture of a wide range of end-products (Kamm et al., 2008), some of which are currently produced from oil. The composition of the hemicellulose fractions in hardwoods and softwoods is different. Considering the raw materials employed in this study, E. globulus hemicelluloses are dominated by acetylated glucuronoxylan; whereas the P. pinaster hemicelluloses include both glucomannan and arabinoxylan (Pérez et al., 2002; Gírio et al., 2010). The different hemicellulose composition defines the target products achievable from each of the raw materials considered. Regarding lignin, hardwoods contain mainly guayacil and syringyl units; while softwoods are richer in guayacil units and syringyl units may not appear or be present in small proportions (Barakat et al., 2013; Marriott et al., 2016). Fractionation of lignocellulosic materials Biomass fractionation refers to the selective separation of one or several of the structural components making part of the feedstock. An ideal technology would allow the separation of the biomass into its main components at high purities and yields (Peleteiro et al., 2015). The resulting fractions can be converted separately into a scope of value-added, commercial products. The integral utilization of the major feedstock components ensures a minimum waste generation. The complete fractionation of LCM usually entails a number of consecutive reaction and separation stages. A number of reported fractionation schemes begin with the separation of hemicelluloses by partial hydrolysis, followed by lignin dissolution of the hemicellulose-depleted solid, leaving a solid enriched in cellulose suitable for a number of purposes (including manufacture of glucose solutions by enzymatic hydrolysis and production of platform chemicals such as 5-hydroxymethylfurfural (HMF) and levulinic acid). These ideas are summarized in the following paragraphs. Treatments based on the solubilization of hemicelluloses The solubilization of hemicelluloses can be accomplished using different methods, including treatments employing water, alkalis or acids. This study focuses on the solubilization of hemicelluloses by hydrothermal treatments performed with hot, compressed water (also called autohydrolysis). The conversion of wood hemicelluloses into soluble fragments by hydrothermal processing proceeds through an autocatalytic hydrolysis mechanism, which results in the selective depolymerization of the hemicellulosic polysaccharides into soluble oligomers, monosaccharides and acetic acid. Additionally, limited amounts of sugar dehydration products can be obtained under harsh operational conditions. The solid phases from treatments (here denoted autohydrolyzed solids) present increased proportions of lignin and cellulose, which can be further separated (for example, by delignification treatments). Alternatively, the autohydrolyzed solids can be processed as a whole to obtain value-added products. In this PhD Thesis, Eucalyptus globulus wood samples were subjected to hydrothermal processing in order to solubilize the hemicelluloses. Individual and cross-flow coupled reaction stages were assayed for this purpose. The soluble reaction products were identified, quantified, and employed for furfural manufacture; whereas the autohydrolyzed solids were converted into HMF and/or organic acids (levulinic and formic acids). Treatments based on the dissolution of lignin: delignification Wood or autohydrolyzed wood are suitable substrates for delignification reactions allowing the breakdown of native lignin into soluble fragments. Upon delignification treatments, cellulose remains in solid phase with little alteration. Typical delignification methods include treatments in aqueous media (such as the soda, sulfite and kraft processes) or in organic media (such as the Acetocell, Acetosolv, Formacell, Milox or Lignol processes). Some methods based on the use of organic solvents (organosolv delignification) represent a sustainable, cost-efficient and environmentally friendly alternative for the integral valorization of biomass (Penín et al., 2020). In this study, methyl isobutyl ketone (MIBK, a recommended organic solvent by the CHEM21 guide) (Prat et al., 2016) was employed for E. globulus wood fractionation in biphasic media to achieve the one-pot delignification and polysaccharide conversion into platform chemicals. Treatments based on the hydrolysis of cellulose The hydrolysis of cellulose can be accomplished using acids or enzymes. Acidic treatments can be performed using dilute acid solutions at high temperatures, or concentrated acid solutions at low temperatures. Enzymatic treatments use enzyme cocktails (cellulases) containing a mixture of different activities, which act synergistically to achieve the cellulose hydrolysis (Hayes, 2009; Balat, 2011). In comparative terms, acid and enzymatic hydrolysis present a number of advantages and disadvantages respect to the other. Acidic technologies cause corrosion problems, and the catalyst has to be recovered or neutralized, resulting in additional costs. Instead, treatments with enzymes are selective and can be carried out under mild conditions, but long reaction times are needed, and the cost of enzymes is an important economic burden. Treatments using ionic liquids and deep eutectic solvents Ionic liquids (IL) are salts made up of organic cations and organic/inorganic anions. Some of them present low melting temperatures (< 100 ºC), and provide a versatile alternative to organic solvents as agents for LCM fractionation. IL can be used for diverse purposes in the scope of biorefineries, including delignification, wood dissolution, agents for physical separation or cellulose dissolution (Peleteiro et al., 2015; Penín et al., 2020). In this study, three acidic ionic liquids (AIL) (1-butyl-3-methylimidazolium hydrogen sulfate ([C4mim]HSO4), 1-(3-sulfopropyl)-3-methylimidazolium hydrogen sulfate ([C3SO3Hmim]HSO4), and 1-(3-sulfobutyl)-3-methylimidazolium hydrogen sulfate ([C4SO3Hmim]HSO4])) have been assessed as catalysts for converting wood polysaccharides into platform chemicals (furans and/or organic acids). Deep eutectic solvents (DES) are obtained by combining a hydrogen bond donor and a hydrogen bond acceptor. DES are considered as an interesting “green” option for the processing of LCM due to their low melting points, simplicity of synthesis, easy biodegradability and recyclability, among others (Kumar et al., 2020). Valorization of lignocellulosic materials: products obtained in this study from wood fractionation The aim of a LCM biorefinery is the integral valorization of the considered feedstock through the fractionation into its main components. The products obtained mainly depend on multiple factors, including the composition of the raw materials employed, the reaction and separation technologies and the operational conditions. In this study, the valorization of the cellulosic fraction of woods was achieved by hydrolysis-dehydration or hydrolysis-dehydration-rehydration reactions, in order to obtain platform chemicals (HMF, levulinic and formic acid). HMF and levulinic acid were included in the ranking of the “Top 10+4” bio-based products which can be obtained from biomass (Bozell y Petersen, 2010). This review updates a previous report from the U.S. Department of Energy (Werpy y Petersen, 2004). Both compounds are precursors of a broad range of commercial products, some of which are currently produced from oil, such as certain polymers, fuels and fine chemicals (van Putten et al., 2013; Mukherjee et al., 2015; Zheng et al., 2016). Additionally, formic acid is widely employed in the chemical industry, and is considered as a promising chemical for H2 storage (Chen et al., 2020) and for other energy applications, as the direct formic acid fuel cells (Aslam et al., 2012). The valorization of Eucalyptus globulus hemicelluloses was carried out by multi-stage hydrothermal treatments, in order to obtain concentrated solutions of soluble hemicellulose-derived products, especially xylooligosaccharides (XOS) and xylose. Furfural (produced by hydrolysis-dehydration reactions of pentosans) and acetic acid (resulting from the hydrolysis of acetyl groups linked to hemicelluloses) were also obtained in biphasic media treatments. In contrast, the major reaction products from Pinus pinaster wood hemicelluloses were HMF (from hydrolysis-dehydration reactions of hexosans), furfural (from hydrolysis-dehydration reactions of pentosans) and acetic acid (from acetyl group hydrolysis). Additionally, HMF can be rehydrated to yield levulinic and formic acids. Furfural was selected by the U.S. Department of Energy in 2004 as one the top 30 building blocks derived from biomass (Werpy y Petersen, 2004), and remained in the ranking in an update of the study (Bozell y Petersen, 2010). Furfural is a versatile platform chemical with a scope of applications, including pharmaceuticals, fragrances, cosmetics, flavors and resins (Peleteiro et al., 2016a). Acetic acid is widely employed as a solvent, and finds applications in the manufacture of polyvinyl acetate, cellulose acetate or terephthalic acid, among others (Dragone et al., 2020). XOS are short-chain oligomers made up of xylose units with prebiotic activity suitable as functional ingredients for the nutraceutical industry (Santibáñez et al., 2021). The results obtained in this PhD Thesis have been reported in five articles published in journals included in the Science Citation Index, which are included as Appendixes (from A to E) in this document. Eucalyptus globulus wood was subjected to multistage hydrothermal processing (following a crossflow scheme) with intercalated washing stages. Operational conditions leading to a maximum concentration of XOS were identified in each stage. The liquid to solid ratio (LSR) employed was 8 g/g. Water was employed in the formulation of the media employed in the first reaction stage, whereas both the liquid phases resulting from the previous stages and the corresponding washing waters (in the amount needed to keep the LSR constant) were employed for wood processing in the second and third stages. In all treatments, cellulose was almost completely recovered (94.0-94.9%) in the processed solids. Regarding the liquid phases, the target products obtained were monosaccharides (glucose, xylose, arabinose), and soluble oligo- or polysaccharides derived from glucan and xylan. Some xylan-derived saccharides contained arabinosyl substituents and acetyl groups. Furans (HMF and furfural) and organic acids (acetic acid, levulinic acid and formic acid) were obtained at low proportions. The yields of target products obtained were 20.4, 18.1, and 15.6 kg/100 kg oven-dried wood in the first, second and third stages, respectively. Regarding the concentrations of the major hemicellulose-derived saccharides, XOS achieved 15.9, 25.4 and 29.5 g/L, whereas xylose reached 3.52, 8.57, and 15.6 g/L in the liquid phases from the first, second and third autohydrolysis stages, respectively. The liquid phases from each stage were assayed for composition and structural features by HPSEC (High Pressure Size Exclusion Chromatography), HPAEC-PAD (High Performance Anion Exchange Chromatography with Pulsed Amperometric Detection) and MALDI-TOF MS (Matrix Assisted Laser Desorption/Ionization Time of Flight Mass Spectrometry). The major reaction products were monosaccharides and saccharides with degree of polymerization (DP) ≤ 7, substituted by acetyl and O-methyluronic groups. Additionally, this study included the kinetic modeling of the experimental results, and the formulation of material balances. The results dealing with the multi-stage operation of E. globulus wood are included in Appendixes A and B. The experimental information obtained on the dissolution of wood polysaccharides and their conversion into platform chemicals is included in Appendixes C, D and E, and summarized in the next paragraphs. Eucalyptus globulus raw wood was subjected to fractionation in one-pot reactions performed in biphasic media made up of acidified water and MIBK. The reactions were performed in a microwave reactor MARS 6. The main objective of this study was to achieve the one-pot conversion of polysaccharides (cellulose and hemicelluloses) into platform chemicals, and to achieve the dissolution of lignin. The target products from cellulose were levulinic and formic acids (resulting from hydrolysis-dehydration-rehydration reactions, in which glucose and HMF appeared as intermediates). Furfural was the target product from hemicelluloses (resulting from hydrolysis-dehydration of pentosans, in which xylose appeared as a reaction intermediate). Additionally, acetic acid was obtained from the hydrolysis of acetyl groups in hemicelluloses. Operating under selected conditions (180 ºC, 30 min, 1% catalyst concentration, LSR = 20, organic to aqueous phase mass ratio = 2) the total yield of target products (furfural, acetic acid, levulinic acid and formic acid in both phases) reached 45.2 g/100 g oven-dry wood. Under these conditions, lignin was recovered from the organic phase by partial evaporation and further precipitation upon water addition. The structure and molar mass distribution of the recovered lignin was studied using diverse analytical techniques. The experimental results and discussion are detailed in Appendix C. Following a related approach, Pinus pinaster raw wood was processed in biphasic media containing an aqueous solution of an AIL (acting as a catalyst) and MIBK, in order to achieve the conversion of polysaccharides into furans and organic acids. The reactions were performed in a microwave reactor MARS 6. The comparatively poor results obtained when [C4mim]HSO4 was employed as a catalyst were ascribed to the limited acidity of the AIL. When [C3SO3Hmim]HSO4 was employed as a catalyst and the reaction was carried out under selected conditions (190 ºC, 1 g catalyst/10 g oven-dry wood, LSR = 6, organic to aqueous phase mass ratio = 2, reaction time = 7.5 min), pentosans were near quantitatively converted into furfural. Longer reaction times promoted the conversion of hexosans into levulinic and formic acids, but the furfural concentrations decreased owing to the increased participation of side reactions. Favorable results were obtained after 22.5 min, a reaction time that allowed a good balance between the generation of levulinic acid and the consumption of furfural. A comparative study confirmed that the AIL [C4SO3Hmim]HSO4 was a catalyst suitable for manufacturing platform chemicals from Pinus pinaster wood, although its activity was slightly lower than the one of [C3SO3Hmim]HSO4, a fact ascribed to the longer carbon chain. The experimental results and discussion are detailed in Appendix D. Following the general idea explained in the previous paragraph, Eucalyptus globulus wood was processed in biphasic media containing an aqueous solution of the AIL [C3SO3Hmim]HSO4 and MIBK. Wood process was carried out following two different schemes. In the first approach, E. globulus raw wood was subjected to one-pot reactions to achieve the depolymerization and conversion of polysaccharides to yield furfural (from hemicelluloses) and HMF and/or levulinic and formic acid (from cellulose). Operating under selected conditions (190 ºC, 30 min, 0.10 g catalyst/g feedstock, LSR = 10, 2 g MIBK/g aqueous phase), furfural and levulinic acid were obtained at molar conversions of 81.3% and 44.8%, respectively. In the second approach, wood was subjected to a hydrothermal treatment to yield a liquid phase containing hemicellulose-derived soluble saccharides, and a solid phase enriched in cellulose and lignin. The liquid phase was supplemented with [C3SO3Hmim]HSO4 and MIBK to obtain furfural from its precursors at near 78% molar conversion. The solids leaving the hydrothermal stage were treated in the same type of media (aqueous solution of [C3SO3Hmim]HSO4 and MIBK) to produce levulinic acid at a maximum molar conversions of 49.5%. The experimental results and discussion are detailed in Appendix E.