Dissertations / Theses on the topic 'Feedstock pretreatment'

Forest residues represent an abundant and potentially sustainable source of biomass which could be used as a feedstock for a biomass-to-chemicals-and-fuels process (biorefinery). However, due to the heterogeneity of forest residues, one of the expected challenges will be to obtain an accurate material balance of both the starting and pretreated material. As current compositional analysis methods have been developed to quantify more homogenous feedstocks such as whitewood and agricultural crops, it is likely that they will have difficulty in providing a complete material balance for these more diverse substrates. The research work initially assessed the robustness of established methods to quantify a variety of forest residues (bark, hog fuel, forest thinnings, logging residue, disturbance wood) before and after steam pretreatment. It was anticipated that the diverse chemistry and heterogeneity of forest residues would make it difficult to obtain an accurate material balance. Although the NREL recommended methods provided a reasonable estimate of carbohydrate components of the various feedstocks, method revision was necessary to accurately quantify the non-carbohydrate components and thus obtain an acceptable summative mass closure. This was particularly evident for high-extractive containing residues such as bark. After steam pretreatment, the incomplete removal of extractives from the pretreated material proved to be more problematic. The refined material balance methods were subsequently used to evaluate the potential of using pretreated forest residues as a biorefinery feedstock. Acid catalysed steam pretreatment was not as effective on forest residues and poor sugar yields were obtained despite using high enzyme loadings. It was likely that, in the acidic medium resulting from SO₂ catalysed steam pretreatment, the extractives reacted with the lignin and consequently restricted enzyme accessibility to the cellulose. In contrast, an alkaline pretreatment effectively removed most of the extractives and lignin from cellulosic components of the bark. The resulting cellulose-rich, water insoluble component could be almost completely hydrolyzed. It was apparent that established analytical methods will have to be modified to obtain a representative material balance of both the starting and pretreated material and that, even with “tailoring” pretreatment/fractionation strategies, forest residues will prove to be challenging feedstocks for any potential bioconversion process.

Secondary generation biofuels such as cellulosic biofuels rely on large portions of cellulosic bioresources, which may include forests, perennial grasses, wood and agricultural residues. Switchgrass is one promising feedstock for biofuel production. In the present study, thesis work focused on the chemical and structural profiles and hydrothermal pretreatment of switchgrass. Four populations of switchgrass were investigated for their chemical properties among populations and morphological portions, including the compositions of lignin and carbohydrates, extractives content, higher heating value (HHV), and syringyl:guaiacyl (S:G) ratio. The results demonstrate similar chemical profiles and lignin structure among the four populations of switchgrass. Morphological fractions of switchgrass including leaves, internodes, and nodes differ significantly in chemical profiles and S:G ratios of lignin. The structure of isolated cellulose from switchgrass SW9 is similar between leaves and internodes. The structure of isolated lignin from leaves and internodes of switchgrass SW9 differs in S:G ratio and molecular weight. Hydrothermal pretreatment of leaves and internodes indicates that a similar chemical composition and chemical structure for pretreated leaves and internodes. The degree of polymerization (DP) for cellulose of the pretreated internodes is 23.4% greater than that of the pretreated leaves. The accessibility of pretreated leaves measured by Simons' Staining technique is greater than that of pretreated internodes. Pretreated leaves have a 32.5-33.8% greater cellulose-to-glucose conversion yield than do pretreated internodes.

Lignocellulose, mostly from agricultural and forestry resources, is a potential renewable material for sustainable development of biorefineries. From previous studies, reducing sugar production through biological pretreatment involves two steps: solid-state fermentation (SSF) for delignification, followed by enzymatic hydrolysis by adding celluloytic enzymes (cellulase and xylanase etc.). In the process described in this thesis, the necessary enzymes are produced in-situ and the hydrolysis proceeds directly after the solid-state fermentation. Enzyme hydrolysis releases free amino nitrogen (FAN), reducing sugar and many other potential nutrients from the fermented materials. This method additionally avoids the need for removal of inhibitors compared with conventional chemical pretreatment processes. A range of solid-state fermentations were carried out to investigate the effect of washing procedure, particle size and nitrogen supplement on Trichoderma longibrachiatum growth. From these preliminary studies it was concluded that nitrogen supplementation is a crucial factor to improve significantly the fungi growth and production of feedstock using sugarcane bagasse as raw material. In order to evaluate the influence of environmental humidity on petri dish experiments, moist environments were investigated, with over 75% relative humidity to limit water evaporation from solid-state fermentation. The results showed that moist environments gave approximately 1.85 times the reducing sugar yield than dry environments. The process of simultaneous enzymatic hydrolysis of substrates and fungal autolysis were also studied. The degree of hydrolysis was affected by initial fermented solid to liquid ratio, temperature and pH range. The optimal conditions for subsequent hydrolysis of fermented solids were determined. The optimal solid to liquid ratio, 4% (w/w), temperature 50°C and pH 7 were established. The highest final reducing sugar, 8.9 g/L and FAN, 560 mg/L, were measured after 48 h. The fungal autolysis was identified by image analysis as well as by the consumption of nutrient and the release of free amino nitrogen and phosphorous. Solid state fermentation in a multi-layer tray bioreactor and a packed-bed bioreactor were also developed, with moist air supply for oxygen provision and heat removal. Fermented solids in the multi-layer bioreactor led to the highest subsequent hydrolysis yield on reducing sugar, FAN and Inorganic Phosphorous (IP), 222.85 mg/g, 11.56 mg/g and 19.9 mg/g, respectively. These series of fermentation experiments illustrate the feasibility for the application of consolidated bioprocessing, through simultaneous pretreatment and enzyme production as a more economic and environment-friendly process compared with those reported for chemical pretreatment followed by commercial enzyme process. A growth kinetic model regarding both growth and respiration is also proposed. Ethanol production was studied using the generic feedstock produced from sugarcane bagasse and soybean hulls. Total ethanol yield reached 0.31 mg g-1 (61.4% of theoretical yield) after 30 h of submerged fermentation. The result of subsequent fermentation has already shown the potential of the generic microbial feedstock to be used to produce varied products depending on the microorganism utilised.

Doctor of Philosophy
Grain Science and Industry
Praveen V. Vadlani
Lignocellulosic biomass feedstocks are a sustainable resource required for rapid growth of bio-based industries. An integrated approach, including plant breeding, harvesting, handling, and conversion to fuels, chemicals and power, is required for the commercial viability of the lignocellulosic-based biorefineries. Optimization of conversion processes, including biomass pretreatment and hydrolysis, is a challenging task because of the distinct variations in composition and structure of biopolymers among biomass types. Efficient fermentation of biomass hydrolyzates comprising of different types of sugars is challenging. The purpose of this doctoral research was to evaluate and optimize the various processing steps in the entire the biomass value chain for efficient production of advanced biofuels and chemicals from diverse biomass feedstocks. Our results showed that densification of bulky biomass by pelleting to better streamline the handling and logistic issues improved pretreatment and hydrolysis efficiencies. Alkali pretreatment was significantly more effective than acid pretreatment at same processing conditions for grass and hardwood. The ethanol-isopropanol mixture, and glycerol with 0.4% (w/v) sodium hydroxide were the promising organic solvent systems for the pretreatment of corn stover (grass), and poplar (hardwood), respectively. None of the pretreatment methods used in this study worked well for Douglas fir (softwood), which indicates a need to further optimize appropriate processing conditions, better solvent and catalyst for effective pretreatment of this biomass. The brown midrib (bmr) mutations improved the biomass quality as a feedstock for biochemicals production in some sorghum cultivars and bmr types, while adverse effects were observed in others. These results indicated that each potential sorghum cultivar should be separately evaluated for each type of bmr mutation to develop the best sorghum line as an energy crop. Development of an appropriate biomass processing technology to generate separate cellulose and hemicellulose hydrolyzates is required for efficient 2,3-butanediol (BD) fermentation using a non-pathogenic bacterial strain, Bacillus licheniformis DSM 8785. This culture is significantly more efficient for BD fermentation in single sugar media than Klebsiella oxytoca ATCC 8724. Though K. oxytoca is a better culture reported so far for BD fermentation from diverse sugars media, but it is a biosafety level 2 organism, which limits its commercial potential.

Processing of lignocellulosic biomass for transportation fuels and other biocommodities in integrated biorefineries has been proposed as the future for emerging sustainable economies. Currently bioprocessing strategies are all multi-step processes involving extensive physicochemical pretreatments and costly amounts of exogenous enzyme addition. Consolidated bioprocessing (CBP), or direct microbial conversion, is a strategy that combines all the stages of production into one step, thus avoiding the use of expensive pretreatments and exogenous enzymes that reduce the economic viability of the products produced. With a growing trend towards increased consolidation, most of the reported work on CBP has been conducted with soluble sugars or commercial reagent grade cellulose. For CBP to become practical fermentative guidelines with native feedstocks and purified cellulose need to be delineated through specific substrate characterization as it relates to possible industrial fermentation. By carefully reviewing the fundamentals of biomass pretreatments for CBP, a comparative assessment of the fermentability of non-food agricultural residue and processed biomass was conducted with Clostridium thermocellum DSMZ 1237. Cell growth, and both gaseous and liquid fermentation end-product profiles of C. thermocellum as a CBP processing candidate was characterised. Batch fermentation experiments to investigate the effect of cellulose content, pretreatment, and substrate concentration, revealed that higher yields were correlated with higher cellulose content. Pretreatment of native substrates that increased access of the bacterial cells and enzymes to cellulose chains in the biomass substrate were key parameters that determined the overall bioconversion of a given feedstock to end-products. The contribution of amorphous cellulose (CAC) in different biomass substrates subjected to the same pretreatment conditions was identified as a novel factor that contributed to differences in bioconversion and end-product synthesis patterns. Although the overall yield of end products was low following bioaugmentation with exogenous glycosyl hydrolases from free-enzyme systems and cellulosome extracts. Treatment of biomass substrates with glycosyl hydrolase enzymes was observed to increase the rate of bioconversion of native feedstocks in biphasic manner during fermentation with C. thermocellum. A “quotient of accessibility” was identified as a feedstock agnostic guideline for biomass digestibility.
October 2015

Development of biomass feedstocks with desirable traits for cost-effective conversion is one of the main focus areas in biofuels research. As suggested by techno-economic analyses, the success of a lignocellulose-based biorefinery largely relies on the utilization of lignin to generate value-added products, i.e. fuels and chemicals. The fate of lignin and its structural/compositional changes during pretreatment have received increasing attention; however, the effect of genetic modification on the fractionation, depolymerization and catalytic upgrading of lignin from genetically engineered plants is not well understood. This study aims to fractionate and characterize the lignin streams from a wild-type and two genetically engineered switchgrass (Panicum virgatum) species (low lignin content with high S/G ratio and high lignin content) using three different pretreatment methods, i.e. dilute sulfuric acid, ammonia hydroxide, and aqueous ionic liquid (cholinium lysinate). The structural and compositional features and impact of lignin modification on lignin-carbohydrate complex characteristics and the deconstruction of cell-wall compounds were investigated. Moreover, a potential way to upgrade low molecular weight lignin to lipids by Rhodococcus opacus was evaluated. Results from this study provide a better understanding of how lignin engineering of switchgrass influences lignin fractionation and upgrading during conversion processes based on different pretreatment technologies.

碩士
國立臺灣大學
森林環境暨資源學研究所
106
The objective of this experiment is to obtain the eco-friendly and well-aligned cellulose film. The experiment divide into two parts. First, we use three materials (BEK, Avicel and bacterial cellulose) treated with pretreatment and use N-methylmorpholine-N-oxide (NMMO) to produce regenerated cellulose films. Second, bacterial cellulose is cultivated on different condition. We use statistic method, POM, SEM, profilometer, and AFM to determine the morphology of cellulose films and use GPC and DP to determine the properties of cellulose. The pretreatment used in this research is phosphorylation and polyelectrolytes. Phosphorylation and polyelectrolytes help cellulose dissolve in NMMO which decrease the roughness of thin films.Bacterial cellulose is cultivated in different condition, such as substrate, tilt height and condition mode. The result shows that tilt height,substrate and condition mode can''t influence the roughness of bacterial cellulose. However, the alignment of bacterial cellulose can be influenced by fluid direction. Pretreatment can obtain well-aligned cellulose film which can be applied in many different field.

The depletion of fossil fuel reserves and global warming are the two main factors contributing to the current demand in clean and renewable energy resources. Biofuels are renewable energy resources and have an advantage over other renewable resources due to biofuels having a zero carbon footprint and most feedstock is abundant. The use of biofuels brought about major concerns and these include food, water and land security. The use of lignocellulose as bioethanol feedstock can provide a solution to the food, water and security concerns. Biofuels such as bioethanol can be produced from lignocellulose by breaking down the structure of lignocellulose liberating fermentable sugars. Amaranth lignocellulose has a potential to be used as a feedstock for bioethanol production because amaranth plants has a high yield of biomass per hectare, require very little to no irrigation and have the ability to withstand harsh environmental conditions. The aim of this study was to investigate the viability of amaranth as a feedstock for bioethanol production by using alkaline assisted microwave pretreatment. Alkaline pretreatment of amaranth using Ca(OH)2, NaOH and KOH at various concentrations (10-50 g kg-1 of alkaline solution in water) was carried out at different energy input (6-54 kJ/g). The pretreated broth was enzymatically hydrolysed using Celluclast 1.5L, Novozyme 188 and Tween 80 at pH 4.8 and 50oC for 48 hours. The hydrolysate was further fermented to ethanol using Saccharomyces cerevisiae at a pH of 4.8 and 30oC for 48 hours. The effect of microwave pretreatment on amaranth lignocellulose was evaluated using scanning electron microscopy (SEM) and Fourier transform infrared spectroscopy (FTIR). The monomeric sugars and ethanol were quantified using high performance liquid chromatography (HPLC). A maximum sugar yield of 0.36 g/g of biomass was obtained for pretreatment with 30 g kg-1 Ca(OH)2 solution in water, 0.24 g/g of biomass was obtained for pretreatment with 50 g kg-1 NaOH solution in water and 0.21g/g of biomass was obtained for pretreatment with 50 g kg-1 KOH solution in water at 32 kJ/g of energy input. After enzymatic hydrolysis the yields increased to 0.43 g/g, 0.63 g/g and 0.52 g g-1 of biomass for Ca(OH)2 , KOH and NaOH pretreated biomass respectively. The highest ethanol yield obtained was found to be 0.18 g/g of biomass from fermentation of KOH pretreated broth. The ethanol yield obtained from fermentation of Ca(OH)2 and NaOH pretreated broth was 0.13 g/g of biomass and 0.15 g/g of biomass respectively. The results showed that an increase in concentration of alkaline solution and an increase in energy input liberate more sugars. A decrease in biomass loading was found to increase the total sugar yield. Pretreatment with KOH was found to liberate more pentose sugars than the other alkaline solutions. The morphological changes shown by the SEM images showed that microwave irradiation is effective in breaking the structure of amaranth lignocellulose. The structural changes shown by the FTIR also validated that alkaline bases were effective in breaking the lignin, cellulose and hemicellulose linkages and liberating more sugars in the process. This work has demonstrated the enormous potential that amaranth lignocellulose has on being a feedstock for bioethanol production.
MSc (Engineering Sciences in Chemical Engineering), North-West University, Potchefstroom Campus, 2014

The research represents a first approach to measure the utilization potential of ultrasonic pretreatment on six different substrates: fat, oil and grease (FOG), paper sludge, ground switch grass, ground hay, ground wheat straw, and cut wheat straw. Several laboratories techniques were applied to determine the influence of ultrasonication on biogas production and yield, biogas quality, and digestibility ratio. With the data, mathematical definitions of Net Energy Balance and Net Economy Balance were computed to draw a first justification or rejection of the use of this pretreatment technology for the specific substrates. Ultrasonic pretreatment has a significant effect on biogas production and yield as well as digestibility ratio (p-value < 0.0001) from the early stages of digestion until as far as 50 days of digestion. Ultrasonication and macro particle size management did not influence significantly the methane (CH4) content in the biogas (p-value = 0.1793). Also, the impact of ultrasonication on the substrate varies between all studied feedstock. Most of the ultrasonicated digestion cases studied provided a negative Net Energy and Economic Balance except for FOG where a certain window of utilization was found. In the context of an ultrasonication process retrofit upgrade, the technology looks to be more useful for substrates that are hard to digest when the retention time is, unfortunately, longer than common retention time. In the context of a new facility, a design that includes an understood ultrasonication technology has yet a small potential success depending on several variables. The ultrasonication technology for anaerobic digestion is hard to recommend due to its energy consumption that, in many cases, overshadows the energy surplus derived from its use.
MITACS

The suitability of water hyacinth (Eichornia crassipes) as a viable feedstock for renewable energy production was investigated in this project. Water hyacinth used in this study was harvested from the Vaal River near Parys in the northwest region of the Free State province, South Africa (26°54′S 27°27′E). The wet plants were processed in the laboratory at the North-West University by separating the roots from the leaves and the stems, thus obtaining two separate water hyacinth feedstock. Characterisation of the feedstock showed that the stems and leaves are more suitable for bio-energy production than roots, due to the higher cellulose and hemicellulose content and very low lignin content of the stems and leaves. Water hyacinth was evaluated as feedstock for the production of bio-ethanol gel, bio-ethanol, bio-oil and bio-char. The recovery of water from the wet plants for use in bio-refining or for use as drip-irrigation in agriculture was also investigated. Cellulose was extracted from water hyacinth feedstock to be used as a gelling agent for the production of ethanol-gel fuel. A yield of 200 g cellulose/kg dry feedstock was obtained. The extracted cellulose was used to produce ethanol-gel with varying water content. The gel with properties closest to the SANS 448 standard contained 90 vol% ethanol and 10 vol% water, with 38 wt% cellulose. This gel was found to ignite readily and burn steadily, without flaring, sudden deflagrations, sparking, splitting, popping, dripping or exploding from ignition until it had burned to extinction, as required by SANS 448. The only specifications that could not be met were the viscosity (23,548 cP) and the high waste residue (32 wt%) left after burning. The other major concern is the extremely high costs involved with the manufacturing of ethanol-gel from water hyacinth cellulose. It can be concluded that ethanol-gel cannot be economically produced using water hyacinth as feedstock. Chemical and enzymatic extraction of water from the feedstock, which is stems and leaves or roots, showed that the highest yield of water was obtained using a combination of Celluclast 1.5 L, Pectinex Ultra SP-L and additional de-ionised water. A yield of 0.89 ± 0.01 gwater/gwater in biomass was realised. This is, however, only 0.86 wt% higher than the highest yield obtained (0.87 ± 0.01 gwater/gwater in biomass) using only Pectinex Ultra SP-L and de-ionised water. It is recommended to use only Pectinex Ultra SP-L and de-ionised water at a pH of 3.5 and a temperature of 40°C. Using one enzyme instead of two reduces operating costs and simplifies the chemical extraction process. The extracted water, both filtered and unfiltered, was not found to be suitable for domestic use without further purification to reduce the total dissolved solids (TDS), potassium and manganese levels. Both the unfiltered and filtered water were, however, found to be suitable for industrial and agricultural purposes, except for the high TDS levels. If the TDS and suspended particle level can be reduced, the extracted water would be suitable for domestic, industrial and agricultural use. The potential fermentation of the sugars derived from the water hyacinth, using ultrasonic pretreatment, was investigated. Indirect ultrasonic treatment (ultrasonic bath) proved to be a better pretreatment method than direct sonication (ultrasonic probe). The optimum sugar yield for the ultrasonic bath pretreatment with 5% NaOH was found to be 0.15 g sugar/g biomass (0.47 g sugar/g available sugar) using an indirect sonication energy input of 27 kJ/g biomass. The optimum sugar yield is lower than those reported in other studies using different pretreatment methods. Theoretically a maximum of 0.24 g ethanol can be obtained per g available sugar. This relates to an ethanol yield of 0.08 g ethanol/kg wet biomass. The low yield implies that ethanol production from water hyacinth is not economically feasible. The production of bio-oil and bio-char from water hyacinth through thermochemical liquefaction of wet hyacinth feedstock was investigated. An optimum bio-char yield of 0.55 g bio-char/g biomass was achieved using an inert atmosphere (nitrogen) at 260°C and the stems and leaves as feedstock. With the roots as feedstock a slightly lower optimum yield of 0.45 g bio-char/g biomass was found using a non-reducing atmosphere (carbon monoxide) at 280°C. The bio-oil yield was too low to accurately quantify. As water is required during thermochemical liquefaction, it was found unnecessary to dry the biomass to the same extent as was the case with the pretreatment and fermentation of the water hyacinth, making this a more feasible route for biofuel production. Bio-char produced through liquefaction of roots as the feedstock and leaves and stems as the other feedstock had a higher heating value (HHV) of 10.89 ± 0.45 MJ/kg and 23.31 ± 0.45 MJ/kg respectively. Liquefaction of water hyacinth biomass increased the HHV of the feedstock to a value comparable to that of low grade coal. This implies a possible use of water hyacinth for co-gasification. The most effective route for bio-energy production in the case of water hyacinth was found to be thermochemical liquefaction (12.8 MJ/kg wet biomass). Due to the high production costs involved, it is recommended to only use water hyacinth as a feedstock for biofuel production if no alternative feedstock are available.
MIng (Chemical Engineering), North-West University, Potchefstroom Campus, 2014

Lignocellulosic biomass is composed of cellulose, hemicellulose, lignin and extraneous compounds (waxes, fats, gums, starches, alkaloids, resins, tannins, essential oils, silica, carbonates, oxalates, etc). The sugars within the complex carbohydrates (cellulose and hemicellulose) can be accessed for cellulosic bioethanol production through ethanologenic microorganisms. However, the composite nature of lignocellulosic biomass, particularly the lignin portion, presents resistance and recalcitrance to biological and chemical degradation during enzymatic hydrolysis/saccharification and the subsequent fermentation process. This leads to a very low conversion rate, which makes the process uneconomically feasible. Thus, biomass structure requires initial breakdown of the lignocellulosic matrix. In this study, two types of biomass pretreatment were applied on barley straw grind: radio-frequency (RF)-based dielectric heating technique using alkaline (NaOH) solution as a catalyst and steam explosion pretreatment at low severity factor. The pretreatment was applied on barley straw which was ground in hammer mill with a screen size of 1.6 mm, so as to enhance its accessibility and digestibility by enzymatic reaction during hydrolysis. Three levels of temperature (70, 80, and 90oC), five levels of ratio of biomass to 1% NaOH solution (1:4, 1:5, 1:6, 1:7, & 1:8), 1 h soaking time, and 20 min residence time were used for the radio frequency pretreatment. The following process and material variables were used for the steam explosion pretreatment: temperature (140-180oC), retention time (5-10 min), and 8-50% moisture content (w.b). The effect of both pretreatments was assessed through chemical composition analysis and densification of the pretreated and non-pretreated biomass samples. Results of this investigation show that lignocellulosic biomass absorbed more NaOH than water, because of the hydrophobic nature of lignin, which acts as an external crosslink binder on the biomass matrix and shields the hydrophilic structural carbohydrates (cellulose and hemicellulose). It was observed in the RF pretreatment that the use of NaOH solution and the ratio of biomass to NaOH solution played a major role, while temperature played a lesser role in the breakdown of the lignified matrix, as well as in the production of pellets with good physical quality. The heat provided by the RF is required to assist the alkaline solution in the deconstruction and disaggregation of lignocellulosic biomass matrix. The disruption and deconstruction of the lignified matrix is also associated with the dipole interaction, flip flop rotation, and friction generated between the electromagnetic charges from the RF and the ions and molecules from the NaOH solution and the biomass. The preserved cellulose from the raw sample (non-treated) was higher than that from the RF alkaline pretreated samples because of the initial degradation of the sugars during the pretreatment process. The same observation applies to hemicellulose. This implies that there is a trade-off between the breakdown of the biomass matrix/creating pores in the lignin and enhancing the accessibility and digestibility of the cellulose and hemicellulose. The use of dilute NaOH solution in biomass pretreatment showed that the higher the NaOH concentration, the lower was the acid insoluble lignin and the higher was the solubilized lignin moieties. The ratio of 1:6 at the four temperatures studied was determined to be the optimal. Based on the obtained data, it is predicted that this pretreatment will decrease the required amount and cost of enzymes by up to 64% compared to using non-treated biomass. However, the use of NaOH led to an increase in the ash content of biomass. The ash content increased with the decreasing ratio of biomass to NaOH solution. This problem of increased ash content can be addressed by washing the pretreated samples. RF assisted-alkaline pretreatment technique represents an easy to set-up and potentially affordable route for the bio-fuel industry, but this requires further energy analysis and economic validation, so as to investigate the significant high energy consumption during the RF-assisted alkaline pretreatment heating process. Data showed that in the steam explosion (SE) pretreatment, considerable thermal degradation of the energy potentials (cellulose and hemicellulose) with increasing acid soluble and insoluble lignin content occurred. The high degradation of the hemicellulose can be accounted for by its amorphous nature which is easily disrupted by external influences unlike the well-arranged crystalline cellulose. It is predicted that this pretreatment will decrease the required amount and cost of enzymes by up to 33% compared to using non-treated biomass.The carbon content of the solid SE product increased at higher temperature and longer residence time, while the hydrogen and oxygen content decreased. The RF alkaline and SE treatment combinations that resulted to optimum yield of cellulose and hemicellulose were selected and then enzymatically digested with a combined mixture of cellulase and β-glucosidase enzymes at 50oC for 96 h on a shaking incubator at 250 rev/min. The glucose in the hydrolyzed samples was subsequently quantified. The results obtained confirmed the effectiveness of the pretreatment processes. The average available percentage glucose yield that was released during the enzymatic hydrolysis for bioethanol production ranged from 78-96% for RF-alkaline pretreated and 30-50% for the SE pretreated barley straw depending on the treatment combination. While the non-treated sample has available average percentage glucose yield of just below 12%. The effects of both pretreatment methods (RF and SE) were further evaluated by pelletizing the pretreated and non-pretreated barley straw samples in a single pelleting unit. The physical characteristics (pellet density, tensile strength, durability rating, and dimensional stability) of the pellets were determined. The lower was the biomass:NaOH solution ratio, the better was the quality of the produced pellets. Washing of the RF-alkaline pretreated samples resulted in pellets with low quality. A biomass:NaOH solution ratio of 1:8 at the three levels of temperature (70, 80, and 90oC) studied are the RF optimum pretreatment conditions. The higher heating value (HHV) and the physical characteristics of the produced pellets increased with increasing temperature and residence time. The steam exploded samples pretreated at higher temperatures (180ºC) and retention time of 10 min resulted into pellets with good physical qualities. Fourier transform infrared-photoacoustic spectroscopy (FTIR-PAS) was further applied on the RF alkaline and SE samples in light of the need for rapid and easy quantification of biomass chemical components (cellulose, hemicellulose, and lignin). The results obtained show that the FTIR-PAS spectra can be rapidly used for the analysis and identification of the chemical composition of biofuel feedstock. Predictive models were developed for each of the biomass components in estimating their respective percentage chemical compositions.

As part of ongoing research to evaluate the bioconversion of softwood feedstocks to ethanol, the feasibility of using Douglas-fir residues was investigated. Whitewood feedstocks were pretreated using SC₂-catalyzed steam explosion under three pretreatment severity conditions (low, medium and high) with a goal of demonstrating efficient fermentation of the hemicellulose-rich water-soluble (WS) fraction. The chosen severity had a pronounced effect on the recovery of the hemicellulose sugars from the feedstock, as well as the proportion of monomeric sugars. Low-severity pretreatment resulted in the greatest recovery of hemicellulose sugars, with yields decreasing significantly for an increase in severity. Reduced-severity pretreatments also resulted in the recovery of fewer sugar decomposition products in the WS fraction, which translated into improved fermentation using an SSL-adapted strain of Saccharomyces cerevisiae. In contrast, the WS fraction obtained under high-severity conditions could not be fermented. Douglas-fir feedstocks containing bark (10, 20, 30 and 100% w/w) were pretreated under medium-severity to evaluate bark's impact on bioconversion. Bark had only a minor impact on the yield of hemicellulose sugars, with no negative effect on the monomeric sugar recovery. However, bark caused a significant decrease in the soluble sugar concentration. Process-derived fermentation inhibitors (e.g., furfural, HMF) also decreased, while naturally occurring inhibitors (e.g., lipophilic compounds) increased with greater bark loading. Despite this increase, bark had no detrimental impact on the rate of fermentation, and all hydrolysates were fermentable to high ethanol yield. Efforts to increase the low sugar concentration in the WS fractions were evaluated, in an attempt to increase the ethanol concentration recovered following fermentation. Increasing the sugar concentration in the WS fraction by physical means resulted in decreased rates of fermentation and reduced ethanol yields, even at low concentration factors (2- to 3-fold). As an alternative strategy, the sugar concentration in the WS fraction was augmented with carbohydrate derived from the water-insoluble cellulose component. Enzymatic hydrolysis of the cellulose component directly in the WS fraction proved unsatisfactory, due to inhibition by both carbohydrate and non-carbohydrate components. However, supplementation (1:1) of the WS fraction with the cellulose hydrolysate obtained separately in buffer provided improved sugar concentration, and significantly faster fermentation due to the effective dilution of inhibitors in the WS fraction. Using this approach, the initial hexose sugar concentration in the whitewood WS fraction was increased by 56%, and a final ethanol concentration of 23.4 g L⁻¹ was obtained.

Tese no âmbito do doutoramento em Biociências, especialização em Biotecnologia orientada pelo Professor Doutor José António Couto Teixeira, pelo Professor Doutor António Manuel Veríssimo Pires, e pelo Doutor João Miguel dos Santos Almeida Nunes e apresentada ao Departamento de Ciências da Vida da Faculdade de Ciências e Tecnologia da Universidade de Coimbra.
Lignocellulosic biomass, such as forest and agriculture residues or dedicated energy crops, is a promising renewable feedstock for the production of advanced biofuels and chemical building blocks. Lactic acid (LA) has been identified as one with high potential, playing an essential role in industrial applications ranging from the food industry to life-sciences. Moreover, LA is widely used for producing green, biodegradable and biocompatible polylactic acid polymers (PLA). In order to develop an efficient process for the production of LA from lignocellulosic biomass, complementary to the selection of the biomass, process optimization must be carried out. For this, three main operations have to be considered - (1) biomass pretreatment, (2) enzymatic saccharification to obtain fermentable sugar by cellulases and (3) the fermentation of sugars by suitable microorganisms to lactic acid. The selection of the raw material as well as the development of the main process operations are the focus of this work. The selection of the raw material was focused on evaluating two mixtures of lignocellulosic biomass (M1-4 and M2-3), forest ecosystems and biological resources from marginal land, in order to co-produce oligosaccharides, solid fuel and glucose under a biorefinery concept. The selection of mixtures was based on different criteria, namely, territorial distribution, fire risk during summer months and total sugar content. The two mixtures were submitted to autohydrolysis pretreatment under non-isothermal conditions (in the range of 190 ºC - 240 ºC corresponding to severity of 3.71 to 4.82). Both mixtures were compared in terms of fractionation (cellulose and lignin recoveries and hemicellulose solubilization) and for enzymatic susceptibility of cellulose. The highest xylan recoveries (62 and 69 %), as xylose and xylooligosaccharides, were achieved for both mixtures in the liquid phase at 206 ºC. Moreover, enzymatic susceptibility of these pretreated mixtures was also improved from 45 to 90 % of glucose yield by increasing pretreatment severity and at 206 ºC glucose yield from enzymatic hydrolysis resulted in 60.1 % and 73.7 % for M1-4 and M2-3, respectively, these results led to the selection of the mixture M2-3 for further processing. The solid fraction of M2-3 resulted from autohydrolysis (AM2-3) at 206 ºC was subsequently delignified by uncatalyzed ethanol-organosolv process to recover hemicellulose, cellulose and lignin in separate streams. Three factors were evaluated in the experimental design of organosolv process: ethanol concentration (30–80%), temperature (160–200 ºC) and time (20–60 min). Organosolv process showed that the best compromise between lignin removal and cellulose preservation was obtained at highest temperature and ethanol concentration (p-value of 0.05). Maximal delignification (49.40%) was obtained at the highest severity condition (200ºC, 60 min, 80 % EtOH). Moreover, 35.32 g/L glucose, corresponding to a glucose yield of 49.65 %, was produced from enzymatic hydrolysis of delignified biomass. FTIR analysis of the isolated lignins (OL1–OL10) showed that the main lignin structure was not changed, while thermal analysis revealed Tg values from 73 to 85 ºC. All OL presented radical scavenging activity as high as the commercial antioxidant BHT. Considering the glucose yield of solid fraction from AM2-3 and from organosolv, the last one did not increased enzymatic susceptibility and for this reason the following processes did not include this step. Whereas enzymatic susceptibility improved by increasing pretreatment severity, M2-3 was presented to autohydrolysis pretreatment at 226 ºC .The solid fraction (AM2-3) obtained was submitted to separated hydrolysis and fermentation (SHF) and simultaneous saccharification and fermentation (SSF) for LA production. LA yield on glucose obtained for both assays was 1 g/g, although the volumetric productivity of SSF (2.5 g/Lh) was higher than SHF (0.8 g/Lh). Therefore, the SSF process was optimized through a factorial design to evaluate the effect of independent variables, solids load and enzyme-substrate ratio (ESR), on LA production. The maximum concentration of LA was obtained using the highest solids load (16 %) and with the highest ESR (54 FPU/g). Finally, scale up of LA was performed in a bioreactor under the optimized conditions in Erlenmeyer flasks, being obtained 61.74 g/L of LA at 44 h which corresponds to LA yield of 0.97 g/g. In order to present a quantitative evaluation of the environmental loads associated with LA production from M2-3, it was compared with the lactic acid production from non-renewable resources and modeled using the Life Cycle Assessment method through SimaPro®. The life cycle approach took into account the raw material, transport, pretreatment, saccharification and fermentation and LA recovery considering 1 tonne of LA as the functional unit. The major environmental savings obtained by replacing one tonne of fossil-based LA by biobased LA are : 4056.60 kg CO2 eq. of global warming potential; 193.03 kBq U235 eq. of ionizing radiation potential; 3.78 kg C2H4 eq of photochemical oxidation potential; 0.73 kg PO4 3- eq freshwater eutrophication potential; 9569.40 kg 1,4-DB eq. of terrestrial ecotoxicity potential; 99.32 kg 1,4-DB eq. of fresh water aquatic ecotoxicity potential; 137.69 kg 1,4-DB eq. of marine aquatic ecotoxicity potential; 94.89 human toxicity potential and 126.63 m2 of land use. Auxiliary chemicals, electricity and enzyme used in the biobased LA production are most relevant to the total environmental impacts. Biobased LA production significantly reduces the impact on the environment, giving 60 % environmental savings compared to fossil-derived LA. The results obtained in this work demonstrate the potential of lignocellulosic biomass as an unexploited raw material for an economical and environmental viable solution to produce lactic acid by fermentation.
A biomassa lenhocelulósica, como resíduos florestais e agrícolas ou culturas energéticas dedicadas, é uma matéria-prima renovável promissora para a produção de biocombustíveis avançados e químicos de plataforma. O ácido láctico (LA) foi identificado como um de elevado potencial, desempenhando um papel essencial em aplicações industriais, que vão desde a indústria alimentar até às ciências da vida. Além disso, o ácido láctico é amplamente utilizado na produção de polímeros de ácido poliláctico (PLA) verdes, biodegradáveis e biocompatíveis. A fim de desenvolver um processo eficiente de produção de LA a partir de biomassa lenhocelulósica, complementar à seleção da biomassa, a otimização do processo deve ser realizada. Assim, três operações principais devem ser consideradas: (1) pré-tratamento da biomassa, (2) sacarificação enzimática para obter açúcares fermentáveis através de celulases e (3) fermentação de açúcares por microrganismos adequados ao ácido láctico. A seleção da matéria-prima, bem como o desenvolvimento das principais operações do processo são o foco deste trabalho. A seleção da matéria-prima centrou-se na avaliação de duas misturas de biomassa lenhocelulósica (M1-4 e M2-3), ecossistemas florestais e vegetação natural, com o objetivo de coproduzir oligossacarídeos, combustível sólido e glucose sob um conceito de biorrefinaria. A seleção das misturas foi baseada em diferentes critérios, nomeadamente a distribuição territorial, risco de incêndio durante os meses de Verão e teor total de açúcar. As duas misturas foram submetidas a um pré-tratamento de autohidrólise em condições não isotérmicas (na gama de 190 ºC - 240 ºC correspondente a uma severidade de 3.71 a 4.82). Ambas as misturas foram comparadas em termos de fracionamento (recuperações de celulose e lenhina e solubilização de hemicelulose) e de suscetibilidade enzimática da celulose. As maiores recuperações de xilanos (62 e 69 %), como xilose e xilooligossacarídeos, foram obtidas para ambas as misturas na fase líquida a 206 ºC. Além disso, a suscetibilidade enzimática destas misturas pré-tratadas foi também melhorada de 45 a 90 % em rendimento da glucose, com o aumento da severidade do pré-tratamento e, a 206 ºC, o rendimento de glucose da hidrólise enzimática resultou em 60.1 % e 73.7 % para M1-4 e M2-3, respectivamente, esses resultados levaram à seleção da mistura M2-3 para os processos posteriores. A fração sólida resultante da auto-hidrólise (AM2-3) a 206 ºC foi subsequentemente delignificada pelo processo de etanol-organosolv não catalisado para recuperar hemicelulose, celulose e lenhina em fluxos separados. Foram avaliados três fatores no desenho experimental do processo organosolv: concentração de etanol (30-80 %), temperatura (160-200 ºC) e tempo (20-60 min). O processo organosolv mostrou que o melhor compromisso entre a remoção da lignina e a preservação da celulose foi obtido nas condições extremas de temperatura e concentração de etanol (p-value de 0.05). A delignificação máxima (49.40%) foi obtida na condição de maior severidade (200 ºC, 60 min, 80 % EtOH). Além disso, 35.32 g/L de glucose, correspondendo a um rendimento de glucose de 49.65 %, foi produzida a partir da hidrólise enzimática da biomassa delignificada. A análise FTIR das lenhinas isoladas (OL1-OL10) mostrou que a estrutura principal da lenhina não foi alterada, enquanto que a análise térmica revelou valores de Tg de 73 a 85 ºC. Todas as lenhinas (OL1-OL10) apresentavam atividade antioxidante tão elevada quanto o antioxidante comercial BHT. Considerando o rendimento de glucose da fração sólida do AM2-3 e do organosolv, este último não aumentou a suscetibilidade enzimática e por esse motivo os processos a seguir não incluíram esta etapa. Considerando que a suscetibilidade enzimática melhorou com o aumento da severidade do pré-tratamento, M2-3 foi submetido ao pré-tratamento de auto-hidrólise a 226 ºC. A fração sólida (AM2-3) obtida foi submetida a hidrólise e fermentação em separado (SHF) e sacarificação e fermentação em simultâneo (SSF) para produção de LA. O rendimento de LA em glucose obtido para ambos os ensaios foi de 1 g/g, embora a produtividade volumétrica de SSF (2.5 g/Lh) tenha sido superior a SHF (0.8 g/Lh). Portanto, o processo SSF foi otimizado através de um desenho fatorial para avaliar o efeito das variáveis independentes, carga de sólidos e relação enzima-substrato (ESR), na produção de LA. A concentração máxima de LA foi obtida com a maior carga de sólidos (16 %) e com a maior ESR (54 FPU/g). Por fim, o aumento de escala do LA foi realizado em biorreator nas condições otimizadas nos frascos Erlenmeyer, sendo obtido 61.74 g/L de LA às 44 h que corresponde a rendimento de LA de 0.97 g/g. Para apresentar uma avaliação quantitativa das cargas ambientais associadas à produção de LA a partir de M2-3, esta foi comparada com a produção de ácido láctico a partir de recursos não renováveis e foi modelada utilizando o método de Avaliação do Ciclo de Vida através do SimaPro®. A abordagem do ciclo de vida teve em conta a matéria-prima, transporte, pré-tratamento, sacarificação e fermentação e recuperação de LA, considerando 1 tonelada de LA como a unidade funcional. As maiores poupanças ambientais obtidas através da substituição de uma tonelada de LA de base fóssil por LA de base biológica são : 4056.60 kg CO2 eq. de potencial de aquecimento global; 193.03 kBq U235 eq. de potencial de radiação ionizante; 3.78 kg C2H4 eq de potencial de oxidação fotoquímica; 0.73 kg PO43- eq de potencial de eutrofização de água doce; 9569.40 kg 1,4-DB eq. de potencial de ecotoxicidade terrestre; 99.32 kg 1,4-DB eq. de potencial de ecotoxicidade aquática de água doce; 137.69 kg 1,4-DB eq. de potencial de ecotoxicidade aquática marinha; 94.89 potencial de toxicidade humana e 126.63 m2 de uso da terra. Os produtos químicos auxiliares, eletricidade e enzimas utilizados na produção de LA de base biológica são os mais relevantes para os impactos ambientais totais. A produção de LA de base biológica reduz significativamente o impacto sobre o ambiente, proporcionando 60 % de poupança ambiental em comparação com o LA de origem fóssil. Os resultados obtidos neste trabalho demonstram o potencial da biomassa lignocelulósica como matéria-prima inexplorada para uma solução económica e ambientalmente viável para a produção de ácido lático por fermentação.
BLC3 Evolution Lda