Biochemical Conversion of Biomass:

Hydrothermal Pretreatment, By-Product Formation, Conditioning, Enzymatic

Saccharification, and Fermentability

Dimitrios Ilanidis

Department of Chemistry Umeå University, Sweden 2021

This work is protected by the Swedish Copyright Legislation (Act 1960:729) Dissertation for PhD ISBN: 978-91-7855-519-2 Electronic version available at: http://umu.diva-portal.org/ Printed by: KBC Service Center Umeå Umeå, Sweden 2021

Dedication / Quote

To my family

You will never do anything in this world without courage. It is the greatest quality of the mind next to honor.

- Aristotle

i Table of contents Abstract ...... iv List of abbreviations ...... v List of publications ...... vii Enkel sammanfattning på svenska ...... viii 1. Biorefining of lignocellulosic biomass ...... 1 2. Composition of lignocellulosic biomass ...... 2 2.1 Lignocellulosic feedstocks ...... 2 2.1.1 Agricultural residues ...... 2 2.1.2 Residues from forestry and forest industry ...... 2 2.2 Constituents of lignocellulose ...... 3 2.2.1 Cellulose ...... 4 2.2.2 Hemicelluloses ...... 4 2.2.3 Lignin ...... 5 2.2.4 Other constituents ...... 5 3. Biochemical conversion of lignocellulosic biomass ...... 6 3.1 Pretreatment ...... 7 3.1.1 Pretreatment methods ...... 8 3.1.2 Hydrothermal pretreatment ...... 9 3.1.2.1 Severity factor ...... 12 3.2 Enzymatic hydrolysis of cellulose ...... 12 3.2.1 Classic cellulolytic enzymes ...... 13 3.2.2 Auxiliary enzymes and LPMO ...... 13 3.3 Fermentation ...... 13 3.3.1 Fermentation with yeast and filamentous fungi ...... 15 3.3.2 Bacterial fermentation ...... 15 4. Bioconversion inhibitors ...... 16 4.1 Enzyme inhibitors ...... 16 4.2 Microbial inhibitors ...... 16 4.2.1 Aliphatic acids ...... 17 4.2.2 Aliphatic aldehydes ...... 18 4.2.3 Benzoquinones ...... 18 4.2.4 Furan aldehydes ...... 18 4.2.5 Phenylic compounds ...... 18 5. Conditioning ...... 19 5.1 Treatment with alkali ...... 19 5.2 Treatment with reducing agents ...... 19 5.3 Other treatment methods ...... 20

ii 6. Aims of the study ...... 21 7. Methods ...... 22 7.1 Analysis of biomass ...... 22 7.1.1 Determination of water insoluble solids content in pretreated biomass ...... 22 7.1.2 Compositional analysis of pretreated solids using sulfuric acid ... 22 7.1.3 Pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS) . 24 7.1.4 Nuclear magnetic resonance (NMR) ...... 24 7.1.5 Fourier-transform infra-red spectroscopy (FTIR) ...... 24 7.1.6 Scanning electron microscopy (SEM) ...... 25 7.1.7 X-ray diffraction (XRD) ...... 26 7.1.8 Analytical enzymatic saccharification ...... 26 7.2 Analysis of pretreatment liquids ...... 26 7.2.1 Post-hydrolysis using sulfuric acid ...... 26 7.2.2 High-performance liquid chromatography (HPLC) ...... 27 7.2.3 High-performance anion-exchange chromatography (HPAEC) .. 27 7.2.4 Gas chromatography - mass spectrometry (GC-MS) ...... 27 7.2.5 Liquid chromatography - mass spectrometry (LC-MS) ...... 28 7.2.6 Determination of total phenolics and Total Aromatic Content (TAC) ...... 28 7.2.7 Total Carboxylic Acid Content (TCAC) ...... 29 8. Results and discussion ...... 30 8.1 Pretreatment conditions, by-product formation, enzymatic digestibility, and fermentability (Papers I, II & III) ...... 30 8.1.1 Paper I ...... 30 8.1.2 Paper II ...... 31 8.1.3 Paper III ...... 33 8.2 By-product formation, conditioning and fermentability (Papers IV & V) ...... 34 9. Conclusions and further perspectives ...... 37 10. Acknowledgments ...... 38 11. References ...... 39

iii Abstract Lignocellulosic residues have great potential as feedstocks for production of bio-based chemicals and fuels. One of the main routes is biochemical conversion, which typically includes pretreatment, enzymatic saccharification, microbial fermentation of sugars, and valorization of hydrolysis lignin. Utilization of a broad variety of lignocellulosic feedstocks and development of more efficient conversion processes are advantageous for making bio-based commodities affordable. Biochemical conversion of lignocellulosic biomass was investigated with the overall aim to understand how variations in pretreatment conditions affected yields, formation of bioconversion inhibitors, enzymatic digestibility of pretreated materials, and fermentability. Experiments included estimation of pseudo-lignin content and quantitation of recently discovered microbial inhibitors, such as formaldehyde and p-benzoquinone. Conditioning of pretreated material to improve the efficiency of reactions with biocatalysts was further investigated. Hydrothermal pretreatment of sugarcane bagasse was investigated by using both autocatalyzed and sulfuric-acid-catalyzed pretreatment and by varying temperature and time in such a way that the severity factor was maintained at one of three predetermined values. For autocatalyzed pretreatments, the enzymatic digestibility of the pretreated solids was directly proportional to the severity. Pretreatment conditions that were just harsh enough to almost quantitatively solubilize hemicelluloses gave the best results. Potential effects of the redox environment during hydrothermal pretreatment were investigated by addition of oxygen gas or nitrogen gas in experiments with sugarcane bagasse and Norway spruce. The investigation demonstrated that gas addition, and especially addition of oxygen gas, can be used to modulate the severity of acidic hydrothermal pretreatment. Hydrothermal pretreatment of wheat straw was investigated to evaluate the impact of pretreatment conditions on newly discovered inhibitors, enzymatic digestibility, and fermentation. An increase of the temperature up to 190 °C in autocatalyzed pretreatment led to high combined glucose and xylose yields; up to ~480 kg/ton (dry weight) raw wheat straw. A correlation between enzymatic digestibility and removal of hemicelluloses was observed. A techno-economical evaluation of several conditioning methods for slurries of steam-pretreated spruce indicated that treatment with sodium sulfite was the most favorable option. Treatments with sulfite and dithionite successfully decreased the concentration of formaldehyde. Results also indicate that increased temperature in conditioning of hydrolysate could to some extent compensate for using lower dosages of sodium dithionite.

iv List of abbreviations

ASL Acid-soluble lignin

BQ p-Benzoquinone

CBH Cellobiohydrolase

CBP Consolidated bioprocessing

(CP/MAS) 13C NMR Cross-polarization magic angle spinning carbon-13 nuclear magnetic resonance

DAD Diode array detector

DMBQ 2,6-Dimethoxybenzoquinone

EG Endoglucanase

EU European Union

FAOSTAT Food and Agriculture Organization Corporate Statistical Database

FTIR Fourier transform infra-red

GC Gas chromatography

HCl Hydrochloric acid

HHF Hybrid hydrolysis and fermentation

HMF 5-Hydroxymethylfurfural

HPAEC High-performance anion-exchange chromatography

HPLC High-performance liquid chromatography

IEA International Energy Agency

v LPMO Lytic polysaccharide monooxygenase

MeOH Methanol

PAD Pulsed amperometric detector

Py-GC/MS Pyrolysis - gas chromatography/mass spectrometry

RI Refractive index

RID Refractory index detector

SEM Scanning electron microscopy

SHF Separate hydrolysis and fermentation

SSF Simultaneous saccharification and fermentation

TAC Total aromatic content

TCAC Total carboxylic acid content

TMS Trimethylsilyl

WIS Water-insoluble solids

XRD X-ray diffraction

vi List of publications

Paper I Ilanidis, D.; Stagge, S.; Jönsson, L.J.; Martín, C. (2021). Effects of operational conditions on auto-catalyzed and sulfuric-acid-catalyzed hydrothermal pretreatment of sugarcane bagasse at different severity factor. Ind. Crops Prod. 159, 113077.

Paper II Ilanidis, D.; Wu, G.; Stagge, S.; Martín, C.; Jönsson L.J. (2021). Effects of redox environment on hydrothermal pretreatment of lignocellulosic biomass under acidic conditions. Bioresour. Technol. 319, 24211.

Paper III Ilanidis, D.; Stagge, S.; Jönsson, L.J.; Martín, C. (2021). Hydrothermal pretreatment of wheat straw: Effects of temperature and acidity on by-product formation and inhibition of enzymatic hydrolysis and ethanolic fermentation. Agronomy 11, 487.

Paper IV Ilanidis, D.; Stagge, S.; Alriksson, B.; Cavka, A.; Jönsson, L.J. Comparison of efficiency and cost of methods for conditioning of slurries of steam- pretreated spruce. Manuscript.

Paper V Ilanidis, D.; Stagge, S.; Alriksson, B.; Jönsson, L.J. Factors affecting detoxification of softwood enzymatic hydrolysates using sodium dithionite. Manuscript.

Additional work by the author

Jönsson, L.; Ilanidis, D.; Alriksson, B. Method and system for reducing inhibitory substances of a lignocellulosic biomass-based material. EP20217671.5.

vii Enkel sammanfattning på svenska

Rester från jordbruk och skogsbruk, till exempel halm, bagass från sockerrör och lågvärda fraktioner bestående av lövved och barrved, kan användas för att producera biobaserade kemikalier och drivmedel. Sådana rester består av lignocellulosa, där cellulosa, hemicellulosa och lignin utgör de organiska huvudbeståndsdelarna. Ett tillvägagångssätt är biokemisk konvertering, som bl.a. omfattar förbehandling, enzymatisk försockring av cellulosa, fermentering av socker och vidareförädling av lignin.

En förbehandlingsteknik som har kommit långt med avseende på industriell implementering är s.k. hydrotermisk förbehandling, där lignocellulosan under relativt kort tid (t.ex. 10 minuter) upphettas till höga temperaturer (t.ex. 200 °C). Förbehandlingen syftar huvudsakligen till att bryta ned hemicellulosan, vilket gör råvaran mer mottaglig för enzymatisk försockring. Den förbehandlade lignocellulosan får ett surt pH eftersom det bildas syror när hemicellulosa bryts ned. Ibland tillsätter man extra syra, t.ex. svavelsyra, för att hemicellulosan lättare ska brytas ned till socker. Det bildas även biprodukter, vilket minskar utbytet och kan leda till att enzymatisk försockring och mikrobiell fermentering blir hämmade. Det är därför viktigt att förstå hur förbehandlingsbetingelser påverkar biproduktbildning, försockring och förjäsning. Om det bildas hämmande biprodukter kan det vara fördelaktigt att konditionera det förbehandlade materialet, dvs. förändra dess egenskaper så att det får god kompatibilitet med enzymer och mikroorganismer och det blir högre utbyte i enzymatiska och mikrobiella processer.

Denna avhandling presenterar forskning om hydrotermisk förbehandling av bagass från sockerrör, barrved och halm, samt hur olika parametrar som temperatur, tid och närvaro av gaser under förbehandlingsprocessen påverkar utbyte och kemisk sammansättning, biproduktbildning, försockring och förjäsning. Tillsats av syrgas befanns ha stor påverkan på förbehandlingen. Biprodukter som tidigare har varit dåligt kända har påvisats och kvantifierats, t.ex. s.k. pseudo-lignin (ett falskt lignin som består av partiellt nedbrytna kolhydrater och som uppvisar vissa likheter med riktigt lignin), alifatiska aldehyder och bensokinoner.

Metoder som kan användas för konditionering har jämförts och studerats i högre detaljgrad än tidigare. Speciellt har olika metoder för behandling med reduktionsmedel studerats och jämförts med mer konventionella metoder, som behandling med kalciumhydroxid och aktivt kol. Forskningen har ökat

viii förståelsen av hur olika konditioneringsmetoder fungerar och kan komma att få tillämpning i bioraffinaderier, speciellt då barrved används som råvara.

ix

1. Biorefining of lignocellulosic biomass

Petrol and other fossil-based fuels are currently the most affordable energy sources, but their availability will rapidly decrease with growing global demand in the future. Almost 80% of the used energy, globally, is based on fossil fuels (IEA, 2018). Nevertheless, many countries have targets that include a vehicle fleet that is independent of fossil fuels as an important step towards the vision of sustainable energy supply without any net emissions of greenhouse gases (GHG). The production and supply of energy increases over the years. In order to address this problem through the production of renewable energy, technologies such as biomass refining need to be further developed (Nanda et al., 2014). Biorefining of biomass refers to processes in which several products, such as biofuels, green chemicals, and materials, are produced from different constituents of lignocellulosic biomass. Biorefining processes include, for instance, chemical pulping using the Kraft or the sulfite process, thermochemical conversion, and biochemical conversion. For instance, the Kraft process is used by the Äänekoski bioproducts mill in Finland and by Södra Cell Mönsterås in Sweden to make products such as chemical pulp, renewable energy, tall oil, turpentine, and biomethanol. The sulfite process is used by several existing biorefineries, for example Domsjö Fabriker in Sweden and Borregaard in Norway, to make specialty cellulose, lignosulfonates, bioethanol, and biogas. Thermochemical processes include pyrolysis, combustion, gasification, and hydrothermal liquefaction. Biochemical processes involve the use of enzymes and/or microorganisms. Lantmännen Agroetanol in Norrköping, Sweden, produces bioethanol, protein feed, and carbon dioxide used as carbonic acid. The main raw material is grain, but advanced biofuels are also produced using waste and residues from food industry. Ethanol is the most commonly used biofuel. USA and Brazil produce the largest amounts, mainly from corn grain and sugarcane as feedstocks (Manochio et al., 2017; AFDC). In 2015, the share of energy from renewable sources expressed as gross final consumption reached 16.7%. For the EU, this was an increase from 8.5% in 2004. Increased production of biofuels would benefit from extending the feedstock base. Lignocellulosic biomass, particularly residues from agriculture and forestry, is an abundant resource that is not used for production of food. Therefore, biochemical processes aiming at producing bioethanol and other biofuels from lignocellullosic biomass could play a significant role in the energy supply of the future.

1

2. Composition of lignocellulosic biomass

2.1 Lignocellulosic feedstocks Trees, shrubs, wood products, grasses, and agricultural waste (such as sugarcane bagasse, wheat straw, corn stover, rice straw, and cotton stalks) are made up of three main organic constituents; cellulose, hemicellulose, and lignin (McKendry, 2002). Lignocellulosic biomass, such as agricultural and agro-industrial residues and forest residues are abundant resources, and they are considered as very interesting and significant resources for biochemical conversion of biomass into bio-based products (Sánchez, 2009).

2.1.1 Agricultural residues Four main crops providing agricultural residues are corn, wheat, rice, and sugarcane (Zabed et al., 2016). Their total global production is estimated to more than 5300 million tons of dry weight biomass per year (Jiang et al., 2017; Dias et al., 2018). Corn residues include stover and husks, with a yield of 4.0 tons/acre (Kim and Dale, 2004). The global production of sugarcane bagasse amounts to around 540 million tons per year (Bezerra and Ragauskas, 2016). Moreover, wheat straw and rice straw are two of the most abundant agricultural residues and have great potential as feedstocks (Saini et al., 2015). The yearly production of wheat straw worldwide is estimated to 980 million tons (FAOSTAT). Moreover, the co-utlization of these argicultural residues can have a large effect on bioethanol production (Susmozas et al., 2020). Residues derived from olive trees are also considered as potential lignocellulosic material. Cultivation of olive trees has been increased from around nine million ha in 2014 to almost eleven million ha in 2020 (Romero- García et al., 2014; Contretas et al., 2020). The residues of these cultivations (olive tree pruning, olive stones, and olive pomace) do currently not have any application. Biorefinery concepts based on olive tree residues have been investigated (Contretas et al., 2020). Recent studies focus also on conversion of avocado biomass into bio-based products (García-Vargas et al., 2020).

2.1.2 Residues from forestry and forest industry With regard to forest trees, the feedstocks that are suitable for biorefinery processes include low-quality wood, small trees, branches and tops from harvested trees, and forest-industrial residues (Limayem and Ricke, 2012). Woody feedstocks can be divided into two groups, viz. hardwood and

2

softwood. Hardwood comes from broad-leaved trees, whereas softwood comes from conifers. Hardwoods include, for example, aspen, poplar, birch, willow, and cottonwood, whereas pine and spruce are two of the most abundant softwoods. Due to damage caused by rot and insects, all wood logs are not suitable for production of sawn goods or pulp. Low-quality wood is sold as fuel wood and to much lower price than pulp wood and timber. Branches and tops are other low-value fractions from forestry. Bark and sawdust are examples of residues from the forest industry. Wood waste is also a potential feedstock. Advantages with woody feedstocks include availability throughout the year, well developed logistics and markets, and widespread certification schemes such as Forest Stewardship Council (FSC) and Programme for the Endorsement of Forest Certification (PEFC).

2.2 Constituents of lignocellulose Plant-derived biomass consists mainly of plant cell walls. The main constituent of fresh biomass is typically water, but the composition is typically discussed on dry-weight basis. With regard to the primary cell wall, the main organic constituents are cellulose, hemicellulose, and pectin, whereas the main organic constituents of secondary cell walls are cellulose, hemicellulose, and lignin (Vanholme et al., 2013) (Fig. 1). Lignocellulosic biomass has a complex structure. Depending on the species, the fractions of the constituents can vary (Hu and Ragauskas, 2012). For instance, in hardwood species and on dry-weight basis, the fraction of cellulose varies between 40 and 55%, the fraction of hemicellulose between 24 and 40%, and the fraction of lignin between 18 and 25%. In dried softwood, the fraction of cellulose fluctuates between 45 and 50%, the fraction of hemicellulose between 25 and 35%, and the fraction of lignin between 25 and 35% (Reshamwala et al., 1995).

3

O

P

Cellulose Lignin Hemicellulose

PLANT PLANT CELLS SECONDARY CELL WALL

Pentoses

CH2OH CH2OH CH2OH HC HC HC

HC HC HC

OCH3 OCH3 OH3C OH OH Hexoses OH

coniferyl alcohol sinapyl alcohol p-coumaryl alcohol CELLULOSE HEMICELLULOSE LIGNIN

Figure 1: A model of the molecular structure of the main constituents of lignocellulosic material (secondary plant cell wall). Components are arranged so that the cellulose and hemicellulosic chains are embedded in lignin. The structural formulas for lignin are the monomeric precursors, the monolignols.

2.2.1 Cellulose Cellulose, which is the most abundant organic polymer on earth, is a linear polymer chain formed by anhydroglucose units linked into glucan chains (Klemm et al., 2005; Sjöström, 1993). Cellulose is the β-1,4-polyacetal of cellobiose (4-O-β-D-glucopyranosyl-D-glucose) (Sjöström, 1993). The chemical formula of cellulose is (C6H10O5) (Ragauskas, 2013). Cellulose has both crystalline and non-crystalline structures, and the coalescence of several polymer chains leads to the formation of microfibrils, which are in turn united to form fibres (Yang et al., 2011). Cellulose contains both crystalline and amorphous regions.

2.2.2 Hemicelluloses Hemicelluloses are branched heteropolysaccharides (Sjöström, 1993). Hemicellulose is a mixture of hexose and pentose units, including units of glucose, mannose, galactose, xylose, arabinose, and sugar acids (Nanda et al., 2014). The chemical composition of hemicellulose is related to the biological

4

origin. Hardwood hemicelluloses are composed mainly of xylan, and make up 20–35% of the dry-matter content. In softwood, hemicelluloses mainly consist of galacto-glucomannans and make up 15–27% of the dry-matter content (Hon and Shiraishi, 1991; Pu et al., 2007).

2.2.3 Lignin

Lignin is a complex aromatic polymer ((C9H10O3(OCH3)0.9-1.7)X) consisting of phenylpropane units (Demirbas and Balat, 2007). Lignin is synthesized from oxidative coupling of p-hydroxycinnamyl alcohol monomers (Boerjan et al., 2003). The major monoaromatic building blocks of lignin, referred to as monolignols, are coniferyl alcohol, sinapyl alcohol, and p-coumaryl alcohol (Fig. 1). When incorporated into lignin, these blocks are called guaiacyl (G), syringyl (S), and p-hydroxyphenyl (H) units (Vanholme et al., 2012). Common intermonomeric linkages include β-O-4, β-1, β-β, and β-5 linkages (Li et al., 2015). The composition of lignin is dependent on the biological origin. Softwood lignin is almost exclusively composed of guaiacyl units, while hardwood lignin is composed of both syringyl and guaiacyl units (S-G lignin) (Ragauskas, 2013). Lignin from grasses and herbs typically contains a substantial fraction of H units.

2.2.4 Other constituents

Besides cellulose, hemicelluloses, and lignin, lignocellulosic materials consist of small fractions of low-molecular-weight constituents, such as extractives, and ash (Sjöström, 1993). Extractives include terpenoids, fats, waxes, steroids, and phenolics (Sjöström, 1993; Marques et al., 2010) (Fig. 2).

Figure 2: Soxhlet extraction determination of extractives in pretreated wheat straw using different pretreatment conditions.

More specifically, phenolic constituents are found both in the bark and in the heartwood, whereas terpenoids and steroids are located in the resin pathways and participate in the defense system of the plant (Rowell et al., 2005).

5

In addition to extractives, wood contains inorganic constituents, typically referred to as ash (Sjöström, 1993) (Fig. 3). The content of ash in dry biomass varies from less than 1% to almost 15% (Eom et al., 2011). The ash content of woody biomass is typically in the lower part of that range, whereas herbaceous biomass might contain high fractions of ash.

Figure 3: Ash determination of raw and hydrothermally pretreated wheat straw.

3. Biochemical conversion of lignocellulosic biomass

Biochemical conversion of lignocellulosic biomass typically includes diminution, pretreatment, enzymatic hydrolysis, and microbial fermentation (Fig. 4). Due to the recalcitrance of lignocellullosic biomass, it is necessary to include a pretreatment step before enzymatic hydrolysis in order to make the cellulose more accessible (Galbe and Zacchi, 2012).

6

Enzymes Microbes

Sugars Pretreatment Enzymatic Fermentation Bio-based chemicals hydrolysis and materials

Catalyst Heat Hydrolysis Bio-based chemicals lignin and materials

Figure 4: Schematic overview of biochemical conversion of lignocellulosic biomass into bio-based chemicals and materials through a separate hydrolysis and fermentation (SHF) process.

3.1 Pretreatment Pretreatment is the alteration of structural and chemical properties of lignocellulosic biomass, leading to more efficient hydrolysis of carbohydrates to fermentable sugars (Chang and Holtzapple, 2000) (Fig. 5). There are many different pretreatment techniques and they can be divided into groups in different ways. Pretreatment processes include hydrothermal pretreatment with or without added acid, mild alkaline pretreatment, oxidative pretreatment, chemical pulping processes, pretreatment using alternative solvents, and biological processes (Jönsson and Martín, 2016). Hydrothermal pretreatment is sometimes performed by adding acid and/or by steam explosion (Gandla et al., 2018). The term hydrothermal pretreatment is, however, sometimes used exclusively for pretreatment processes without any added chemicals except water. Dilute-acid pretreatment is sometimes used to refer to pretreatment with low fractions of acid added to the mixture. Pretreatment of biomass contributes to the cost of biorefining of lignocellulosic biomass due to the use of pressurized systems operated at high temperatures and use of chemicals (Sassner at al., 2008). The feasibility and efficiency of a pretreatment method is closely related to the raw material, further processing, and the end products. Determination of yields and recovery of sugars are not enough by themselves to characterize a pretreatment, and other parameters need be examined as well (Galbe and Wallberg, 2019).

7

Figure 5: Schematic representation of the impact of pretreatment on lignocellulosic biomass.

3.1.1 Pretreatment methods

Mild alkaline pretreatment targets lignin and some hemicelluloses (Jönsson and Martín, 2016). Under alkaline pretreatment, bonds between lignin and hemicellulose are broken, which makes cellulose more accessible to enzymes. The most commonly used types of alkali are sodium hydroxide and potassium hydroxide (Galbe and Zacchi, 2012). One drawback of that pretreatment method is the cost of the alkali, and the cost of the acid that is required to adjust the pH to slightly acidic conditions suitable for enzymatic and microbial reactions (typically in the pH range 5-6). An advantage is formation of oligomeric hemicellulose fragments that can potentially be extracted and separated for other uses. Mild alkaline pretreatment has been found to be most useful for grasses and similar types of biomass. Fractionation methods used in chemical pulping mainly target lignin, but also affect hemicelluloses. Chemical pulping processes include the Kraft process, the sulfite process, the soda process, and the organosolv process (Jönsson and Martín, 2016; Mboowa, 2021). The soda pulping process can be used for non- woody plants, whereas Kraft and sulfite pulping are typically used for woody materials (Jönsson and Martín, 2016). Borregaard, a Norwegian biorefinery company, has developed BALI, a sulfite process designed for pretreatment of lignocellulosic biomass (Rødsrud et al., 2012; Costa et al., 2020). Much effort has been devoted to adapting the organosolv process to a biomass pretreatment method. Lignol Innovations in Vancouver, Canada, has investigated organosolv pretreatment in demonstration scale (Arato et al., 2005). Removal of most of the lignin and parts of hemicelluloses by chemical

8

pulping processes results in a material that has very good susceptibility to cellulolytic enzymes, but a typical challenge with chemical pulping or lignin- first approaches include the extensive use of chemicals and, hence, a potential need for expensive and energy-consuming chemical recovery systems. The oxidative pretreatment methods include the use of oxidants, such as hydrogen peroxide, and processes such as ozonolysis and wet oxidation. Wet oxidation experiments are performed by applying oxygen (air) and water at high temperatures and alkaline conditions. The aim is to remove the lignin and parts of the hemicelluloses (Jönsson and Martín, 2016). Pretreatment with simultaneous wet oxidation and hydrogen peroxide seems to be more efficient leading to less formation of by-products. Oxidative pretreatment has been found to be useful for grasses and similar types of biomass. Pretreatment with ionic liquids aims at dissolution of lignocellulose disorganizing the non-covalent interactions of lignocellulosic substrates, but would typically also include a partial degradation of the biomass (Karatzos et al., 2012; Jönsson and Martín, 2016). The most common are the imidazolium- based ionic liquids (Brandt et al., 2013; Gräsvik et al., 2014; Gremos et al., 2011). Some studies of ionic-liquid pretreatments have included investigations of the mass balance (Karatzos et al., 2012; Wang et al., 2017). Further investigations are needed in order to make that type of pretreatment energy efficient by developing strategies for recycling of solvents, achieve recovery of degradation products and improved mass balances, and achieve compatibility with water in moisture-containing biomass.

3.1.2 Hydrothermal pretreatment Hydrothermal pretreatment is the most established and widely used pretreatment method (Hu and Ragauskas, 2012; Gandla et al., 2018). Hydrothermal pretreatment can be performed both with (acid-catalyzed) and without externally added acid catalysts (autocatalyzed), even though in both cases the pretreatment environment will be acidic since this type of pretreatment degrades hemicelluloses leading to formation of carboxylic acids (Garrote et al., 2008). The acidity can be increased by adding acids, such as in dilute-acid pretreatment, but the acidity can also be increased simply by using higher pretreatment temperature or longer residence time, as more and more carboxylic acids will form the more severe the pretreatment conditions are. Therefore, the effects of hydrothermal pretreatment with only water and treatments with externally added acid can be very similar. Hydrothermal pretreatment aims at removal of hemicelluloses to make cellulose more accessible to enzymes. There are typically some effects on cellulose and lignin, but the strategy is to first degrade hemicelluloses to a

9

hemicellulose hydrolysate (sometimes referred to as pretreatment liquid), to save the cellulose for the enzymatic saccharification step, and to save the lignin as a solid residue referred to as hydrolysis lignin. The sugars in hemicellulose hydrolysates and cellulose hydrolysates are typically refined through microbial fermentation processes. The hydrolysis lignin is typically used for energy production through combustion or is further valorized to lignin-derived chemicals and materials. The approach is sometimes referred to as a sugar-platform process, as sugars are important intermediates.

The temperature range of hydrothermal pretreatment is typically from 120 °C to 210 °C (Hu and Ragauskas, 2012), but sometimes it is carried out slightly above that range. If steam explosion is used, the lignocellulosic biomass is first loaded into the reactor, which is heated with steam and pressurized. Then, the pressure is suddenly released. This leads to a mechanical disruption of the fibrous material. The high efficiency of hydrothermal pretreatment and the simplicity of the process makes it very promising for industrial applications (Ruiz et al., 2013). Most biochemical conversion processes that have reached demonstration scale or industrial implementation are based on some form of hydrothermal pretreatment. For instance, the Biorefinery Demo Plant (BDP) in Örnsköldsvik, Sweden, has pretreatment with continuous steam explosion and a capacity of up to two tons of feedstock per day. For some years, Inbicon operated a facility in Kalundborg, Denmark, in which hydrothermal pretreatment was used for production of bioethanol and animal feed from wheat straw. The capacity was up to around one ton of feedstock per hour and around 5.4 million L of bioethanol per year. Other companies and commercial plants, such as POET in Emmetsburg (USA), Abengoa in Hugoton (USA), and Beta Renewables in Crescentino (Italy), have also been using some type of hydrothermal pretreatment to demonstrate biochemical conversion of lignocellulosic biomass into biofuels and other products. Moreover, Clariant in Straubing (Germany) has investigated biochemical conversion of Miscanthus biomass into lignocellulosic sugars and bioethanol using hydrothermal pretreatment. Clariant is planning a new plant with an annual production capacity of 50,000 tons of cellulosic ethanol in Romania. St1 has a demonstration plant in Kajaani (Finland) in which sawdust and wood residues are used to produce 10 million L of cellulosic ethanol per year. Future plans for biochemical conversion using hydrothermal pretreatment include NordFuel in Haapavesi (Finland), and SEKAB's and Praj's joint venture in Örnsköldsvik.

10

Figure 6: Pressurized Parr reactor made of T-316 stainless steel and designed for operating at temperatures up to 225°C for pretreatment of lignocellulosic biomass.

Hydrothermal pretreatment with acid catalyst (acid-catalyzed pretreatment) is a common procedure that has also been implemented in an industrial context. The temperature and the acidic conditions can vary depending on the lignocellulosic biomass and other aspects (Galbe and Wallberg, 2019). As hydrothermal pretreatment without acid catalyst, this type of pretreatment primarily targets the hemicellulose (Jönsson and Martín, 2016). Acids that are used are sulfuric acid, sulfur dioxide, phosphoric acid, hydrochloric acid, and nitric acid. This method results in efficient hemicellulose solubilization with high recovery of hemicellulosic sugars in the pretreatment liquids, and it is suitable for softwood, hardwood, and agricultural residues making it suitable for commercial-scale biorefineries (Silveira et al., 2015). Acid catalysis leads to an increased surface area of lignocellulosic biomass. Degradation by- products may reach concentrations that are inhibitory to enzymes and microorganisms (Jönsson and Martín, 2016; Martín et al., 2018).

11

Figure 7: Solids after autocatalyzed and sulfuric acid-catalyzed hydrothermal pretreatment of sugarcane bagasse using different pretreatment conditions.

3.1.2.1 Severity factor The severity factor is a function of residence time and temperature. It is widely used to compare pretreatment conditions. Together with residence time and temperature, the pretreatment pH is taken into consideration in the combined severity factor (Overend and Chornet, 1987; Chum et al., 1990). SF = Log Ro

R0 = t × exp[(Tr−Tb)/14.75]

Combined severity factor (CSF) = Log (Ro) – pH

…where t is the holding time of the pretreatment in minutes, Tr is the pretreatment temperature in °C, and Tb is the reference temperature (100 °C).

3.2 Enzymatic hydrolysis of cellulose Enzymatic hydrolysis of cellulose is the step after the pretreatment of lignocellulosic biomass in biochemical conversion of biomass to bio-based materials and biofuels. During that process, a complex mixture of several enzymes (cellulolytic and hemicellulolytic) hydrolyze the cellulose and residual hemicelluloses to monosaccharides (Ragauskas, 2013). The crystallinity of cellulose could play an important role in the enzymatic digestibility according to previous studies (Xu et al., 2012; Li et al., 2017) although other studies indicated that there is no correlation between crystallinity and efficiency of enzymatic saccharification (Yu et al., 2011;

12

Nelson et al., 2017). The effects of hemicellulose and lignin on enzymatic digestibility are probably more important than the crystallinity. Advantages of enzymatic saccharification include mild conditions and low energy cost. Enzymes that are included in that process are endoglucanases (EGs), cellobiohydrolases (CBHs), β-glucosidases, xylanases, mannanases, and lytic polysaccharide monooxygenases (LPMOs) (Bommarius et al., 2014; Yang et al., 2011).

3.2.1 Classic cellulolytic enzymes There are three main hydrolytic enzymes involved in degradation of cellulose to glucose. These are endoglucanases (EGs), cellobiohydrolases (CBHs), and β-glucosidases (Van Dyk and Pletschke, 2012). The main source is the filamentous fungus Trichoderma reesei (Hypocrea jecorina) (Kubicek, 2013). Firstly, endoglucanases catalyze the cleavage of amorphous regions of cellulose producing oligosaccharides which will be attacked by cellobiohydrolases, which act at the ends of chains and produce cellobiose. Finally, β-glucosidase hydrolyzes cellobiose to glucose (Lynd et al., 2002).

3.2.2 Auxiliary enzymes and LPMO Hemicellulases, such as endo-xylanase, endo-mannanase, β-xylosidase, α- glucuronidase, α-arabinosidase, α-galactosidase, and others, are enzymes that target hemicelluloses and degrade bonds in the backbone and side-chains of hemicellulose (Meyer et al., 2009; Varnai et al., 2011). Lytic polysaccharide monooxygenase (LPMO) is an oxidoreductase that targets the crystalline regions of cellulose leading to cellulose degradation and creating shorter cellulose chains with oxidized ends (Johansen, 2016). An oxidant, such as moleclar oxygen, and an electron donor are needed in the reaction (Horn et al., 2012).

3.3 Fermentation Fermentation is the process of converting the sugars to biofuels using microorganisms, such as yeast, bacteria, and filamentous fungi. There are different strategies for converting biomass to biofuels using separate hydrolysis and fermentation (SHF), simultaneous saccharification and fermentation (SSF) (Galbe and Zacchi, 2002), hybrid hydrolysis and fermentation (HHF) (Zhong et al., 2019), and consolidated bioprocessing (CBP) (Lynd et al., 2017) (Fig. 8). One of the main benefits of using SHF is that the separation between enzymatic hydrolysis and fermentation allows to perform both approaches under optimal conditions. An SHF approach also

13

facilitates recovery of enzymes and yeast. Nevertheless, a major drawback of SHF is enzyme inhibition caused by increased sugar concentrations. Additionally, as it has been reported by Öhgren et al. (2007) that bioconversion inhibitors that inhibit the fermentation process have stronger influence on an SHF process than on an SSF process. That two separate steps are required also affects the cost of the process. Inhibition of enzymes in an SSF approach is lower, which results in higher yields. A major weakness of the SSF approach is that enzymatic hydrolysis and fermentation have different optimal conditions, enzymatic hydrolysis typically around 50 °C and fermentation typically around 30 °C, and it is necessary to use non-optimal conditions for at least one of the processes. In another approach, sometimes referred to as hybrid hydrolysis and fermentation (HHF), the lignocellulosic biomass is first hydrolyzed at approximately 50°C using enzymes, and, after some time, the temperature is lowered to around 30°C and yeast is added (Zhong et al., 2019). The major benefit of the HHF approach is that both enzymatic saccharification and microbial fermentation can be performed under optimal conditions at least during a part of the process. In another approach, referred to as CBP (consolidated bioprocessing), the fermenting microorganism contributes to the enzymes that are required for the saccharification process (Lynd et al., 2017). The goal of that approach is to combine three steps (production of enzymes, enzymatic hydrolysis, and fermentation) into one.

Pretreatment

SHF SSF HHF CBP

Saccharification Saccharification Saccharification Enzyme production Fermentation Saccharification

Fermentation Saccharification Fermentation Fermentation

Figure 8: Schematic process for SHF, SSF, HHF, and CBP.

14

3.3.1 Fermentation with yeast and filamentous fungi

The most commonly used microoganism in ethanol production is baker’s yeast, Saccharomyces cerevisiae (Favaro et al., 2019). Fermentation processes in which S. cerevisiae yeast is used are typically performed at around 30°C, but some industrial strains allow temperatures of up to around 40°C. Conventional ethanol processes result in approximately 8% (v/v) ethanol from fermentation medium containing 15-16% sugars and generate large volumes of effluent (Puligundla et al., 2019). By performing high-gravity fermentations, effluent volumes can be reduced and ethanol recovery becomes more energy efficient. S. cerevisiae is used because it has the ability to grow very fast and produce ethanol in a large scale (Sedlak and Ho, 2004). S. cerevisiae mainly converts hexose sugars to ethanol, and therefore many studies have been performed on developing engineered S. cerevisiae yeast that would ferment the pentose sugars xylose and arabinose to ethanol (Kricka et al., 2014). However, under aerobic conditions there are non-engineered S. cerevisiae strains that utilize xylose (Kollaras et al., 2011). Other microorganisms often studied in this context include the natural xylose- fermenting yeast Scheffersomyces (Pichia) stipitis. Alternative organisms sometimes used in studies of ethanol production also include Mucor indicus and Trametes versicolor. M. indicus is a saprophytic zygomycete that is able to assimilate several sugars (Sues et al., 2005). Many studies have been conducted on M. indicus in terms of its dimorphism, the production of chitosan and chitin, and its ability to be used for the removal and recovery of waste materials (Sues et al., 2005). Another alternative microorganism is the white-rot fungus T. versicolor, which is capable of growing under hypoxic and anoxic conditions utilizing both hexoses and pentoses concomitantly (Kudahettige et al., 2012).

3.3.2 Bacterial fermentation Several types of bacteria have the ability to convert carbohydrates to bioethanol. Among them, Escherichia coli, Zymomonas mobilis, and Clostridium acetobutylicum are some of the most investigated (Geddes et al., 2011; Yamashita et al., 2008; Ragauskas, 2013). For C. acetobutylicum, butanol and acetone are, however, typically the desired products (ABE fermentation) (Ragauskas 2013; Panesar et al., 2006). Z. mobilis has the ability to metabolize glucose, sucrose, and fructose and produce ethanol. Thermophilic anaerobic bacteria can provide high ethanol yields and ferment various biomass-derived sugars (Lynd et al., 2002). However, bacterial fermentation is not currently the most common method for industrial ethanol production.

15

4. Bioconversion inhibitors

Formation of bioconversion by-products/inhibitors, which are derived by degradation of lignocellulosic biomass during pretreatment, is important in biochemical conversion of biomass, since they inhibit the enzymes and the microorganisms used for enzymatic saccharification and fermentation (Jönsson and Martín, 2016; Ko et al., 2015; Pienkos and Zhang, 2009).

4.1 Enzyme inhibitors Cellulolytic enzymes are subject to feedback inhibition by sugars. Monosaccharides and disaccharides, for instance glucose and cellobiose, respectively, have been investigated in this context (Berlin et al., 2007; Teugjas and Väljamäe, 2013). Studies by Kumar and Wyman (2014) focused on the inhibitory effects of mannan oligosaccharides on cellulases. However, Zhai et al. (2016) reported that oligomeric sugars had little influence on cellulases, in contrast to monosaccharides and phenols. Regarding phenolic compounds in the pretreatment liquids, high inhibitory impact was found for the enzymatic saccharification process (Ximenes et al., 2011; Kim et al., 2013; Zhai et al., 2016). The effects of phenols can potentially be due to enzyme denaturation. A water-insoluble by-product, pseudo-lignin, can have a negative impact of enzymatic saccharification (Normark et al., 2016; Shinde et al., 2018). Pseudo-lignin refers to thermal degradation products derived from carbohydrates that remain insoluble in acid in determination of Klason lignin.

4.2 Microbial inhibitors

Microbial inhibitors can be divided into five main groups, viz. aliphatic acids, aliphatic aldehydes, benzoquinones, furan aldehydes, and phenylic compounds (Fig. 9).

16

Microbial inhibitors

Benzoquinones O Furan aldehydes

O O O O O Aliphatic aldehydes OH

O O HO

O O Phenylic compounds Phenolic O Non-phenolic O Aliphatic acids O Extractives OH O HO O 2% Ash OH O OH <1% O HO O HO Phenolic O

OH Lignin 25-30% Cellulose 40-45% Hemicellulose 25-30%

Figure 9: Formation of microbial inhibitors during acid pretreatment of lignocellulosic biomass.

4.2.1 Aliphatic acids Acetic acid, formic acid, and levulinic acid belong to the group aliphatic acids. Acetic acid results from hydrolysis of acetyl groups in hemicelluloses, whereas formic acid and levulinic acid result from the degradation of sugars and furan aldehydes into smaller compounds (Jönsson and Martín, 2016). For softwood, the concentration of acetic acid in the pretreatment liquid will be low due to the low acetyl content of softwood. The formation of acetic acid is high when hardwood or agricultural residues are used (Kristensen et al., 2009). The concentrations of formic acid and levulinic acid are strongly related to harsh pretreatment conditions causing degradation of sugars. Thus, pretreatment conditions such as high temperature, long reaction time, and strong acidity are causes of formation of levulinic acid and formic acid. Although acetic acid, formic acid, and levulinic acid are typically predominant among aliphatic carboxylic acids, Du et al. (2010) reported 16 aliphatic acids, 15 aromatic acids, and one heteroaromatic acid (2-furoic acid) in lignocellulosic hydrolysates. Whereas acetic acid and formic acid were always common, levulinic acid was present in high concentrations only under the most acidic condition studied (Du et al., 2010).

17

4.2.2 Aliphatic aldehydes

The group aliphatic aldehydes includes formaldehyde and acetaldehyde, which were discovered in pretreated biomass relatively recently (Cavka et al., 2015a). The toxic effect of formaldehyde on yeast is strong, even if the concentration in pretreatment liquids is rather low (Cavka et al., 2015a; Martín et al., 2018). Formation of formaldehyde can occur through partial degradation of lignin, more specifically by cleavage of hydroxymethyl groups in lignin (Gierer, 1970).

4.2.3 Benzoquinones Benzoquinones such as p-benzoquinone (BQ) and 2,6-dimethoxybenzo- quinone (DMBQ), were discovered in lignocellulose hydrolysates more recently (Stagge et al., 2015). However, the strong toxic effect on yeast has been known for a long time. Larsson et al. (2000) reported that a concentration of BQ as low as 20 mg/L (190 µM) was sufficient to completely block a fermentation process with S. cerevisiae yeast.

4.2.4 Furan aldehydes

5-Hydroxymethylfurfural (HMF) and furfural belong to the group furan aldehydes. Furan aldehydes are carbohydrate degradation products. More specifically, furfural and HMF are derived from dehydration of pentoses and hexoses, respectively (Jönsson and Martín, 2016). Compared to benzoquinones, phenylic substances, and aliphatic aldehydes, the molar toxicity of furan aldehydes is low. However, as aliphatic acids, they might be present in high concentrations in pretreatment liquids.

4.2.5 Phenylic compounds Phenylic compounds include phenolic aromatics, such as for example p- hydroxycinnamic acid (p-coumaric acid), and non-phenolic aromatics, such as for example cinnamic acid. Several individual phenols, e.g. vanillin, coniferyl aldehyde, syringaldehyde, p-hydroxybenzaldehyde, acetovanillone, and ferulic acid have been identified in the pretreatment liquids (Mitchell et al., 2014). Vanillin and coniferyl aldehyde are two of the most well known phenolic microbial inhibitors. They are derived through partial degradation of guaiacyl lignin. Most phenolic inhibitors are formed from the degradation of lignin, but there are also some phenolic extractives (Jönsson and Martín, 2016).

18

5. Conditioning

Conditioning is a general term typically used for maneuvers performed after the pretreatment of lignocellulosic biomass in order to make pretreated material suitable for processing using biocatalysts, i.e. enzymes and microorganisms. Conditioning would include pH adjustment, nutrient addition, and actions targeting bioconversion inhibitors. When the target is specifically microbial inhibitors, the term detoxification is often used. As some treatments affect inhibitors of both enzymes and microbes, the more general term conditioning is sometimes preferred. Conditioning might or might not be a separate process step, as some methods can be employed directly in bioreactors just before, at the same time, or even after biocatalysts are added to the mixture (Alriksson et al., 2011).

5.1 Treatment with alkali

Alkaline treatment using calcium hydroxide, sodium hydroxide, or ammonium hydroxide is one of the most investigated detoxification methods (Alriksson et al., 2006). The conventional approach, treatment with calcium hydroxide, is typically referred to as overliming (Martinez et al., 2000). The pH of the pretreated biomass or hydrolysate is raised to values around 9 or 10 and sometimes even higher. Afterwards the pH is adjusted to approximately 5.5 to make the medium suitable for the fermentation process. The mechanism behind alkaline treatment is not fully elucidated, but chemical conversion of inhibitors is more important than precipitation phenomena (Jönsson et al., 2013). The efficiency of the detoxification is closely dependent on the conditions (temperature, time and pH). Nilvebrant et al. (2003) found that alkali treatments should be performed at low temperature (below 30 oC) to avoid extensive sugar degradation. Alriksson et al. (2006) optimized alkali detoxification testing different conditions and reported promising results.

5.2 Treatment with reducing agents A more recent method for conditioning is the use of reducing agents, such as sodium dithionite, sodium sulfite, and sodium borohydride. Other reducing agents are also useful (Alriksson et al., 2011), but relatively inexpensive industrial chemicals, such as the sodium salts of the sulfur oxyanions sulfite and dithionite, are preferred for industrial applications to minimize costs. Furthermore, they can be combined with enzymatic saccharification and microbial fermentation, and do not require a separate process step. Alriksson et al. (2011) reported on improved fermentability using S. cerevisiae and relatively low concentrations of sodium dithionite or sodium sulfite. Cavka

19

and Jönsson (2013) reported on the effects of adding sodium borohydride to pretreatment liquids of both Norway spruce and sugarcane bagasse. Whereas both sodium sulfite and sodium dithionite had a positive effect also on enzymatic saccharification, sodium borohydride gave no such effect (Cavka and Jönsson, 2013). The results suggested that hydrophilization of aromatic inhibitors was important to avoid enzyme inhibition.

5.3 Other treatment methods

Other detoxification methods include the treatment with enzymes such as laccase and peroxidase, treatment using heating and vaporization, and biological treatment with microbes such as Coniochaeta ligniaria, Trichoderma reesei, and Ureibacillus thermosphaericus (reviewed by Jönsson et al., 2013). Biological treatments target phenolics, furan aldehydes, and weak acids and successfully remove these inhibitors or inactivate them (Ko et al., 2015). Liquid-liquid extraction and solid-liquid extraction are techniques that have been used previously as detoxification methods. For instance, treatment with activated charcoal (a solid-liquid extraction technique) has been found to result in improved fermentability (Parajó et al., 1997; Guo et al., 2013). Detoxification of hydrolysates generated from pretreatment of olive pomaces with activated charcoal was recently investigated to minimize the concentrations of inhibitors (López-Linares et al., 2020).

20

6. Aims of the study

The aims of this study were to investigate how different hydrothermal pretreatment conditions for lignocellulosic biomass affected formation of inhibitors, formation of pseudo-lignin, recovery, enzymatic digestibility of cellulose, and fermentability of pretreatment liquid. Additionally, evaluation of different conditioning methods that might address the inhibition problem was investigated. The specific objectives were to: I. Understand how variations in temperature and time affected hydrothermal pretreatment of sugarcane bagasse and wheat straw, and particularly formation of by-products. II. Investigate how the pretreatment liquids affected the catalytic performance of cellulolytic enzymes and the fermentability using yeast. III. Evaluate how compositional changes of pretreated solids affected enzymatic digestibility. IV. Explore how the presence of nitrogen gas and oxygen gas during hydrothermal pretreatment affected Norway spruce and sugarcane bagasse. V. Investigate potential differences between conditioning of hydrolysate and slurry, including variation in WIS content, using sodium sulfite and sodium dithionite. VI. Compare conditioning with industrial reducing agents (sodium sulfite, sodium dithionite, sodium borohydride, and hydrogen) and standard methods such as overliming and treatment with activated charcoal. VII. Investigate how pH, temperature, and dosage of dithionite used for conditioning affect the fermentability of hydrolysates using yeast.

21

7. Methods

7.1 Analysis of biomass

7.1.1 Determination of the content of water-insoluble solids in pretreated biomass

The slurry obtained after pretreatment of biomass is a mixture of solid and liquid material, which is separated for further analyses. That procedure is very important for the calculation of mass balances and for making accurate conclusions about the process. The separation of the liquid phase from the solid phase can be carried out by centrifugation or filtration. The solid phase should be washed with water in order to remove residual liquor. This procedure is applicable for determining enzymatic digestibility (Sluiter et al., 2008). Homogenized slurry is added to a Büchner funnel with filter paper and is filtered by vacuum filtration. Then, water is added for the washing step. The weight of the filter paper before the process and the weight of the filter paper and biomass after the process are used to calculate the content of water- insoluble solids (WIS) (Sluiter et al., 2008).

7.1.2 Compositional analysis of pretreated solids using sulfuric acid The chemical composition of lignocellulosic biomass can be determined using TAPPI standard T249 with sulfuric acid (TAPPI, Peachtree Corners, GA, USA) and using a protocol developed by the National Renewable Energy Laboratory (NREL) (Sluiter et al., 2012). However, raw material analysis using sulfuric acid goes back at least to the work of Peter Klason in the beginning of the 20th century (Sluiter et al., 2010). An extraction step is required to get accurate results. The total content of extractives is typically determined gravimetrically after evaporation of the solvent (which could be for example ethanol, acetone, or a mixture of petroleum ether and acetone) that was used for the extraction. For more detailed analysis of the extractives, analysis with gas chromatography (GC) after derivatization by silylation is a commonly used method. After extraction, the solid residue is dried and incubated with sulfuric acid. The sulfuric acid is used for fractionation of the lignocellulosic biomass by hydrolysis of the polysaccharides (cellulose, hemicelluloses) to monosaccharides. The

22

treatment with sulfuric acid is typically performed in two steps. In the first step, powdered wood is treated with 72% (w/w) sulfuric acid at 30°C for one hour (Sluiter et al., 2012). In the second step, the mixture is diluted with water and the wood is treated with 4% (w/w) sulfuric acid, and the suspension is autoclaved at 121°C for one hour. The main fraction of the lignin, referred to as Klason lignin, is not solubilized by sulfuric acid, and it is measured gravimetrically. A small portion of the lignin, referred to as ASL (acid-soluble lignin) is solubilized together with the sugars and is determined by using UV/Vis spectrophotometry at λ 240 nm. Conventional determination of carbohydrates (typically arabinan, galactan, glucan, mannan, and xylan) is done by determining the corresponding monosaccharides using HPLC (high-performance liquid chromatography) with RI (refractive index) detector. A more advanced approach that offers better resolution and higher sensitivity is HPAEC (high-performance anion- exchange chromatography) with pulsed amperometric detection (PAD). Different lignocellulosic materials exhibit different recalcitrance. If, on the one hand, conditions during two-step hydrolysis with sulfuric acid are too weak, carbohydrates might not be quantitatively degraded to monosaccharides and there may be an underestimation of the carbohydrate content. If, on the other hand, conditions are too severe, monosaccharides might be degraded to furans and carboxylic acids and there would still be an underestimation of the carbohydrate content. Formation of furans can also contribute to too high ASL values, as heteroaromatics, such as furans, absorb in UV in a similar way as lignin-derived aromatics. It is, nevertheless, common to use standard conditions regardless of the properties of the material being analyzed. The solvent used for extraction and the wavelength used for determining ASL are sometimes varied depending on the lignocellulosic material being analyzed. Determination of the hemicellulose composition can also be achieved by trimethylsilyl (TMS) derivatisation and GC. First, inositol is added to lignocellulosic biomass that was dried overnight. Afterwards, under the methanolysis step, 2 M HCl/MeOH is added to the samples, which are then flushed with nitrogen gas and incubated. Then, methanol is added and the samples are evaporated under nitrogen. The procedure is repeated two times. Finally, in the silylation step, Tri-Sil reagent is added to the samples, which are heated at 80 °C. Then, hexane is added and the samples are centrifuged. The analysis of the samples is then done by using GC.

23

7.1.3 Pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS) Pyrolysis-gas chromatography coupled with mass spectrometry (Py-GC/MS) is a rapid and sensitive method for analyzing lignocellulosic biomass, especially for characterization of lignin (Meier et al., 1992). A small amount of freeze-dried sample is heated to around 500 °C in a gas stream consisting of helium, nitrogen, or argon. Under these conditions, the sample is degraded and the fragments are separated by using gas chromatography and identification of the fragments is achieved by using mass spectrometry. Through Py-GC/MS, the ratio of carbohydrate and lignin (C:L ratio) in the lignocellulosic biomass can be determined, as well as the ratio of lignin units, i.e. syringyl (S), guaiacyl (G), and p-hydroxyphenyl (H) units. Although Py- GC/MS provides information about the C:L ratio, the method does not provide information about each individual carbohydrate. However, it is a very helpful tool for the determination of pseudo-lignin formation, since in the Py-GC/MS method partially degraded carbohydrates are detected as carbohydrates and not as lignin (Normark et al., 2016). Therefore, a combination of Py-GC/MS and sulfuric acid compositional analysis can be used for the estimation of pseudo-lignin by subtracting the Py-GC/MS values of lignin from the values recorded by sulfuric acid compositional analysis.

7.1.4 Nuclear magnetic resonance (NMR) Nuclear magnetic resonance (NMR) spectroscopy is based on the physical characteristics of the nuclear spin of atoms in a strong magnetic field (Stenius and Vuorinen, 1999). Solution-state NMR is the most commonly used technique. However, nowadays solid-state NMR has been developed and is also commonly used. Solid-state NMR spectra are broad, as the full effects of anisotropic or orientation-dependent interactions are observed in the spectra, whereas spectra in solution-state NMR are sharper. Solid-state cross- polarization magic-angle spinning (CP/MAS) 13C NMR is used with the advantage that it can detect fine structures that would not be detected with conventional technique. CP/MAS 13C NMR spectroscopy is also an important tool for determination of cellulose crystallinity (Sannigrahi et al., 2008).

7.1.5 Fourier-transform infra-red spectroscopy (FTIR)

Fourier-transform infra-red (FTIR) spectroscopy is an analytical technique used in lignocellulose chemistry to identify structural features, such as chemical substituents, that correspond to peaks in the spectra (Fig. 10). In FTIR, infra-red light is used to scan samples and observe chemical properties.

24

Firstly, the biomass is ground with potassium bromide and then a spectrometer with diffuse reflectance mode and spectral resolution is used. The preparation of samples is easy and analysis is fast, which make the method attractive for analysis of lignocellulosic biomass. Regarding analysis of lignocellulosic biomass, vibrations at wavenumbers between 800 and 1800 cm-1 are related to hemicelluloses, cellulose, and lignin. More specifically, the aromatic vibrations at the wavenumbers 1510 and 1595 cm-1 are related to guaiacyl and syringyl units of lignin, respectively (Faix, 1991; Faix and Böttcher, 1992). Moreover, the bands close to 1429 cm-1 and 897 cm-1 are related to crystalline and amorphous structures in cellulose (Nelson and O’Connor, 1964). A.U

500 700 900 1100 1300 1500 1700 Wavenumber (cm-1)

Figure 10: FTIR spectra of sugarcane bagasse including untreated bagasse and bagasse pretreated (with autocatalysis or sulfuric acid-catalysis) in the presence of nitrogen gas, oxygen gas, and without.

7.1.6 Scanning electron microscopy (SEM) Scanning electron microscopy (SEM) is a type of electron microscopy that produces images of a samples by scanning surfaces with a focused electron beam (Singh et al., 2020). The electrons in the beam interact with the sample, producing various signals that can be used to obtain information about the surface topography and composition. Structural changes of lignocellulosic biomass after pretreatment have been studied by using SEM analysis.

25

7.1.7 X-ray diffraction (XRD) X-ray diffraction (XRD) is used for analysis of the crystallinity of cellulose and of crystallite size providing more information compared to other methods, for instance NMR and FTIR, which only determine values for crystalline and non- crystalline cellulose (Park et al., 2010). The crystallinity of cellulose is calculated from the ratio of the area of peaks representing crystalline cellulose and total peak area. The crystallite size is calculated using the Scherrer equation.

7.1.8 Analytical enzymatic saccharification Analytical enzymatic saccharification is a method that is used to study the digestibility of pretreated solids (Gandla et al., 2018). It is performed in small scale, typically by using mg amounts of lignocellulosic biomass in an acetate buffer. Prior to analytical enzymatic saccharification, the pretreated slurry is filtered through vacuum filtration for separation of the liquid and the solid phases. The solid phase is washed and is then used in the analytical enzymatic saccharification. The main difference between analytical enzymatic saccharification and preparative enzymatic saccharification is that the aim of analytical enzymatic saccharification is to assess digestibility whereas the aim of preparative enzymatic saccharification is to convert as much as possible of the carbohydrate polysaccharides to monosaccharide sugars using appropriate enzyme loadings. Therefore, analytical enzymatic saccharification should not be exhaustive, i.e. the aim is not to degrade all polysaccharides but only a part of them so that differences in digestibility become clear. If saccharification is driven to exhaustion so that all polysaccharides are depleted, it becomes difficult to discern differences in digestibility. An enzyme mixture including at least cellulases (cellobiohydrolase and endoglucanase) and β-glucosidase is used. The samples are typically incubated at 45 °C for 72 h. Samples are taken for analysis of the monosaccharide composition. The inhibitory effect of a pretreatment liquid on the cellulolytic enzymes can be examined by comparing reactions in pretreatment liquid and in acetate buffer.

7.2 Analysis of pretreatment liquids 7.2.1 Post-hydrolysis using sulfuric acid Before analysis of pretreatment liquids using HPLC, HPAEC, and/or GC-MS, which will be described later, the pretreatment liquids may be post-

26

hydrolyzed using 4% (w/w) sulfuric acid and autoclaving at 121 oC for 60 min. This procedure ensures complete degradation of oligosaccharides to monosaccharides, and therefore complete quantification of sugars. In hydrothermal pretreatment with low severity, the hemicelluloses are mainly solubilized into xylo- or mannooligosaccharides. For that reason, an addition of a post-hydrolysis step might be required to break down the oligosaccharides to monosaccharides (Nakasu et al., 2016).

7.2.2 High-performance liquid chromatography (HPLC) In high-performance liquid chromatography (HPLC), a sample mixture or analyte in a solvent, the mobile phase, is pumped at high pressure through a column with chromatographic packing material, which is called stationary phase. HPLC instruments are typically equipped with refractory index detectors (RID) and UV/Vis detectors or diode array detectors (DAD). Analysis of sugars, ethanol and some bioconversion inhibitors can be achieved by HPLC analysis (Sluiter et al., 2012).

7.2.3 High-performance anion-exchange chromatography (HPAEC) High-performance anion-exchange chromatography (HPAEC) is a method that can be used to separate monosaccharides, which can then be detected using a pulsed amperometric detector (PAD) (Sullivan and Douek, 1994). The advantage of the HPAEC method is highly selective separation of monomeric sugars at high pH using a strong anion-exchange stationary phase, and the high sensitivity of the PAD.

7.2.4 Gas chromatography - mass spectrometry (GC- MS) GC is the separation technique of choice for small and volatile molecules, such as extractives. In the begining, the sample is volatilized and separated into various components. After that, the components are eluted from the column at different times. As they leave the GC column, they are ionized by the mass spectrometer using electron or chemical ionization sources. The peak height corresponds to the quantity of the individual compounds. Mass spectrometry (MS) instruments are used and provide analytical results and information of mass, elemental composition, and molecular structure.

27

In order to make a detailed analysis of, for instance, extractives, a derivatization step is required prior to GC analysis, which helps in the separation and detection processes. One of the most common derivatization methods is silylation, in which the samples are treated with BSTFA (bis- (trimethylsilyl)-trifluoroacetamide) and TMCS (trimethyl-chlorosilane) to create TMS (trimethylsilyl) derivatives. Treatment with MSTFA (N-methyl-N- (trimethylsilyl) trifluoroacetamide) addresses the problem of by-products in the derivatization reaction, which are very volatile highly polar substances. 7.2.5 Liquid chromatography - mass spectrometry (LC- MS)

Liquid chromatography–mass spectrometry (LC–MS) is a technique that is commonly used in analytical chemistry and combines separation by liquid chromatography and detection by analysis of mass using mass spectrometry. LS-MS has become an important tool in biotechnology, pharmaceutical industry, and food industry. LC-MS without derivatization has been used for determination of bioconversion inhibitors in several studies. Kieber and Mopper (1990) used LC-MS for determination of carbonyl compounds in seawater samples after derivatization with DNPH (2,4- dinitrophenylhydrazine). Stagge et al. (2015) used LC-MS for determination of p-benzoquinone (BQ) and 2,6-dimethoxy-1,4-benzoquinone (DMBQ) in pretreated lignocellulose using DNPH derivatization (Stagge et al., 2015). The samples are first incubated overnight with DNPH reagent solution, and the next day they are diluted with acetonitrile before analysis using UHPLC-ESI- QqQ-MS (ultra-high performance liquid chromatography-electrospray ionization-triple quadrupole-mass spectrometry).

7.2.6 Determination of total phenolics and Total Aromatic Content (TAC) Folin-Ciocalteu's reagent is often used for determination of total phenolics in the pretreatment liquids. Persson et al. (2002) compared three different methods for analysis of total phenolic content in lignocellulosic hydrolysates, the Prussian Blue method, the Folin Ciocalteu method, and a method based on a horseradish peroxidase electrode. They found that the Folin-Ciocalteu method was less sensitive to interference than the Prussian Blue method. For lignocellulosic hydrolysates, vanillin is suitable as a calibration standard (Persson et al., 2002), as it is a common phenolic substance in hydrolysates of different types of biomass. The color that is generated after 2 h incubation is read at 760 nm in a spectrophotometer.

28

Determination of Total Aromatic Content (TAC) was developed as a fast and simple analytical approach to cover all aromatic and heteraromatic substances in lignocellulose hydrolysates, i.e. both phenylic substances (phenolic and non-phenolic aromatics) and furans (Wang et al., 2018). It is carried out by UV-spectrophotometric determination of the absorbance at 280 nm, a wavelength with high molar absorptivity for substances such as furfural, HMF, and phenols.

7.2.7 Total Carboxylic Acid Content (TCAC)

TCAC determination was developed by Wang et al. (2018) as a titration method to analyze the total carboxylic acid content in lignocellulosic hydrolysates. Although aliphatic carboxylic acids such as acetic acid, formic acid, and levulinic acid, typically make up a large share of the carboxylic acid content of lignocellulosic hydrolysates, Du et al. (2010) demonstrated that hydrolysates contain a multitude of both aliphatic and aromatic carboxylic acids. Although many common carboxylic acids have pKa values of around 4.5, formic acid has a pKa value of only 3.75 (Haynes, 2011). Therefore, it is important that the pH of the hydrolysate sample has a value well below 3.75 before the titration starts. Thus, the pH is first adjusted to 2.8 prior to titration (Wang et al., 2018). Titration with an aqueous solution of sodium hydroxide is then carried out until neutral pH is reached, which is well above the normal pKa value of carboxylic acids.

29

8. Results and discussion

8.1 Pretreatment conditions, by-product formation, enzymatic digestibility, and fermentability (Papers I, II & III)

8.1.1 Paper I Agricultural residues, e.g. straw and sugarcane bagasse, have great potential as feedstocks for production of advanced biofuels, green chemicals, and bio- based materials. This is because agricultural residues are abundant, often not used to any larger extent, and are relatively easy to deconstruct to sugars. The study reported in Paper I was focused on biochemical conversion of sugarcane bagasse, which, seen from a global perspective, is one of the most abundant agricultural residues. The main goal of pretreatment is to increase the digestibility of cellulose to enzymatic saccharification through alteration of the structure and the chemical composition of the lignocellulosic biomass, leading to more efficient hydrolysis of polysaccharides to fermentable sugars. Pretreatment also affects the yields of carbohydrates and lignin, and by- products formed during pretreatment will inhibit cellulases and fermenting microorganisms. There is a large number of studies on hydrothermal pretreatment of sugarcane bagasse, and many of them have been focused on the effects of pretreatment conditions on the composition and enzymatic saccharification of the pretreated solids, and on analysis of some inhibitors (Cruz et al., 2012; Rocha et al., 2012; Brienzo et al., 2017; Santo et al., 2018; Fockink et al., 2018). However, previous studies did not cover newly discovered inhibitors, such as aliphatic aldehydes and benzoquinones. Important by-products, such as pseudo-lignin, which represents a loss in yield, were not quantitated. Furthermore, the impact of pretreatment temperature, time, and acidity on inhibitor formation and enzymatic digestibility is poorly understood. In addition, few previous studies have addressed both the level of enzyme inhibition and the level of microbial inhibition. Thus, in this study we investigated the effects of temperature, time and addition of sulfuric acid on formation of inhibitors and pseudo-lignin, recoveries and yields, enzymatic digestibility of pretreated materials, enzyme inhibition, and inhibition of ethanolic fermentation caused by the formed inhibitors. The severity factor was kept constant at a set of pre-determined values to study individual effects of temperature and time.

30

Compositional analysis of pretreated solids showed higher glucan content in pretreated solids than in raw sugarcane bagasse, and it decreased proportionally with the temperature when sulfuric acid was added, as the acid promoted hydrolysis of glucan (Table 2, Paper I). These results are in agreement with previous reports on hydrothermal pretreatment in the presence of acid catalyst (Hu and Ragauskas, 2012). The xylan content decreased with the increase of temperature, both in autocatalyzed and acid- catalyzed modes. With sulfuric acid, there was almost no xylan left in the pretreated solids due to very high hemicellulose solubilization under these conditions (Table 2). Autocatalyzed pretreatment at intermediate conditions (190 °C for 14 min) resulted in the largest difference between glucan and xylan recovery, and in the highest yield of total sugar (Fig. 1, Paper I). In this study it was revealed that pseudo-lignin formation increased with the most severe pretreatment conditions, and that the increase was higher for sulfuric-acid- catalyzed pretreatments than for autocatalyzed ones (Table 3, Paper I). Only under the most severe conditions of autocatalyzed pretreatments a comparable formation of pseudo-lignin was observed, which is in line with studies by Batista et al. (2019). The CP/MAS 13C NMR spectra of the solids showed clear differences between the pretreatments with or without sulfuric acid (Fig. 3, Paper I). The acid-treated samples had much higher lignin/carbohydrate ratio. The formation of inhibitors showed different trends, but was usually higher after acid-catalyzed than after autocatalyzed treatments, and increased with the temperature at a given severity factor (Table 5, Paper I). For autocatalyzed pretreatments, the enzymatic digestibility of pretreated solids was directly proportional to the severity factor (Fig. 5, Paper I). For weak pretreatment conditions, remaining hemicellulose caused poor enzymatic digestibility and, consequently, low sugar yields. Moreover, the degree of inhibition of enzymes increased with the temperature at a given severity factor (Fig. 5, Paper I). That phenomenon might be correlated to the increase in concentration of total phenolics, as phenolics are well-known inhibitors of cellulolytic enzymes (Ximenes et al., 2011; Zhai et al., 2016). In general, analysis of toxicity showed low ethanol yield, volumetric productivity, and glucose consumption rate, which suggests that detoxification or other countermeasures against inhibitors would be required to reach reasonable fermentability (Fig. 6, Paper I).

8.1.2 Paper II Sugarcane bagasse, an agricultural residue, and Norway spruce, a common softwood in Nordic forestry, can play significant roles as feedstocks in

31

biochemical conversion of biomass to biofuels and advanced materials. Although quite different with regard to chemical composition, both sugarcane bagasse and softwood are rather recalcitrant materials, and harsh pretreatment conditions are needed to degrade the hemicelluloses and make enzymatic saccharification efficient. During harsh pretreatment conditions, inhibitors will be formed (Jönsson and Martín, 2016). The toxicity of the inhibitors might depend on the presence of molecular oxygen during pretreatment, but this is poorly understood and no systematic studies have previous been conducted in which the effects on formation of a wide range of inhibitors, enzymatic digestibility, and fermentability have been monitored. Sugarcane bagasse and wood chips of Norway spruce were hydrothermally pretreated with and without sulfuric acid. The pretreatments were performed either with 10 bars of nitrogen gas or oxygen gas, or without any gas supply to the pretreatment reactor. The aim of the study was to investigate how gas addition and redox environment during hydrothermal pretreatment affected the formation of inhibitors, which impact it had on the toxicity of pretreated material for S. cerevisiae, and its impact on the enzymatic digestibility of the pretreated solids. Reduced yields of pretreated solids were obtained with oxygen gas and nitrogen gas compared to reactions without any gas supply. This was mostly due to solubilization of hemicelluloses and partial degradation of cellulose (Table 1, Paper II). Compositional analysis using a standard protocol (Sluiter et al., 2012) showed higher lignin content in samples that were pretreated in the presence of oxygen gas (Table 1, Paper II). Py-GC/MS analysis showed lower lignin content than analytical acid hydrolysis, which was attributed to formation of pseudo-lignin (Table 2, Paper II). Pseudo-lignin is an aromatic substance formed from carbohydrates during thermal reactions (Brownell and Saddler, 1984; Shinde et al., 2018). It is analyzed as Klason lignin. FTIR spectroscopy showed clear differences among samples pretreated with and without gas supply for both sugarcane bagasse and Norway spruce. Except for glucose, analysis of the chemical composition of pretreatment liquids showed reduced sugar concentrations after pretreatments with gas supply. This was due to higher degree of degradation of hemicellulosic sugars. Glucan recoveries were considerably reduced (approximately 31%) by adding oxygen gas during pretreatment (Fig. 1A, Paper II). The presence of both nitrogen gas and oxygen gas led to increased concentrations of most inhibitory compounds (Tables 3 and 4, Paper II), indicating more severe pretreatment conditions. The pretreatment liquids obtained after gas addition also showed higher toxicity in experiments with S. cerevisiae. Furthermore, samples pretreated with oxygen gas showed significantly improved enzymatic digestibility due to higher degree of hemicellulose solubilization (Fig. 1, Paper

32

II). The presence of the pretreatment liquids had a negative inhibitory impact on enzymatic digestibility, showing lower values compared to those obtained for reactions with acetate buffer as medium (Fig. 1, Paper II). The gas addition had larger impact than anticipated, and studies of gas addition under less severe pretreatment conditions would be of interest to investigate in the future.

8.1.3 Paper III

The transition from energy from fossil-based fuels to energy from renewable resources for creating a greener society is one of the biggest challenges (Ragauskas et al., 2006). Lignocellulosic biomass is an abundant feedstock that can be used for biochemical conversion to bio-based products. Wheat straw, the annual production of which is slightly less than 1.0 billion tons, could be an important raw material. Pretreatment and efficient biocatalytic action are essential features of conversion of biomass to biofuels in high yields. Therefore, a comparison of the effects on wheat straw of different hydrothermal pretreatment conditions was made with regard to yield and chemical composition of pretreated solids, enzymatic digestibility of pretreated solids, and formation of inhibitors and their inhibitory impact on both enzymatic digestibility and fermentability of yeast.

Compositional analysis of untreated and pretreated wheat straw showed a reduction in the yields of pretreated solids mostly due to solubilization of hemicelluloses. The reduction was proportional to the increase in pretreatment temperature. That phenomenon was more remarkable among the autocatalyzed hydrothermal pretreatments (Table 1 and Fig. 1, Paper III). The decrease of the yields of pretreated solids with increasing severity of pretreatment is in line with previous studies, such as Min et al. (2018) and Chen et al. (2018). The xylan content decreased from 19.5% at 160 ºC to 2.5% at 205 ºC (Table 1, Paper III). The reduction of yields and xylan content indicates increasing hemicellulose solubilization with increasing temperature, as expected for hydrothermal pretreatment. The recovery of glucan showed values above 92% for all pretreatment conditions, whereas xylan recovery was below 9% for the most severe one (205 ºC) (Fig. 2, Paper III). By using Py- GC/MS analysis, it was possible to show pseudo-lignin formation at 205 ºC (Table 2, Paper III). Determination of ash content showed a gradual increase from 160 ºC to 205 ºC (Table 1, Paper III).

Analysis of the pretreatment liquids showed an increase of the concentrations of glucose, xylose, and arabinose with increasing temperature (Table 3, Paper

33

III). It is noteworthy that the concentration of xylooligomers was twice as high at 190 ºC as at 175 ºC (Table 3, Paper III). Similar trends have been observed by Chen et al. (2018). Generally, both increasing temperature and using sulfuric acid resulted in increased formation of inhibitors (Table 4, Paper III). The concentrations of aliphatic acids increased with increasing temperature, and that phenomenon was more remarkable above 190 ºC (Table 4, Paper III). The same trend was observed with regard to furfural and HMF, and for most phenolic compounds (Table 4, Paper III). The concentration of total phenolics reached its peak at 205 ºC (7.2 g/L), a large increase from 3.0 g/L at 190 ºC (Table 4, Paper III).

Increasing temperature resulted in increased enzymatic digestibility of pretreated solids up to 98% at 205 ºC (Fig. 3, Paper III). After pretreatment at the weakest pretreatment conditions (160 ºC), the enzymatic digestibility, 45%, was almost half of that observed after pretreatment at 205 ºC (Fig. 3, Paper III). Using pretreatment liquid as reaction medium negatively affected the enzymatic digestibility showing a degree of inhibition of up to 7.3% at 205 ºC. That would be a large effect in an industrial process, so the finding about enzyme inhibition is an important contribution to the knowledge on hydrothermal pretreatment. A strong correlation between enzymatic digestibility and xylan content in pretreated solids was observed (Fig. 4, Paper III), which highlights the significance of almost quantitative removal of xylan from pretreated solids for achieving efficient enzymatic saccharification. This finding is in agreement with a previous study by Leu at al. (2013), who pointed out the importance of removal of hemicelluloses in pretreated lodgepole pine. Moreover, strong inhibitory effects on yeast were shown for samples pretreated at 205 ºC leading to a relative growth rate below 0.4 (Fig. 5, Paper III).

8.2 By-product formation, conditioning and fermentability (Papers IV & V)

Production of biofuels by biochemical conversion of lignocellulosic biomass of softwood is gaining more attention. Recent commercial examples include the plant of St1 in Kajaani and plans for NordFuel in Haapavesi. Nevertheless, it is likely that further research in the area can lead to improved yields of cellulosic ethanol, new ways to valorize hydrolysis lignin, and reduction of production costs. In this way, further technological advancements can contribute to a more competitive industry.

34

For production of high concentrations of ethanol, high sugar yields are needed, which requires high solids loadings (Kristensen et al., 2009; Koppram et al., 2014). However, saccharification and fermentation typically become more difficult when the loadings of solids increase. There are probably several reasons for that, but one contributing factor would be higher concentrations of inhibitors of enzymes and microorganisms, a problem that was addressed further by the research presented in Papers IV and V.

There are several comparisons of different conditioning methods designed to reduce problems with microbial inhibitors (Larsson et al., 1999; Cantarella et al., 2004; Guo et al., 2013; Fernandes-Klajn et al., 2018). However, several of these do not take conditioning with reducing agents into account. Previous research shows that conditioning with reducing agents has several attractive features that most other conditioning methods lack (Alriksson et al. 2011; Cavka et al., 2011; Cavka and Jönsson, 2013; Cavka et al., 2015b). Therefore, it is of interest to compare treatment with reducing agents with other detoxification methods, especially with regard to chemical effects. There is a techno-economical evaluation of treatment with sulfite (Cavka et al., 2015b), but there is no comparison of different reducing agents. Although several common industrial reducing agents have been examined, hydrogenation, which is widely used in industry, has not been investigated previously. The approach used is focused on low concentrations of reducing agents to avoid a costly chemical recovery step, and mild reaction conditions, which are compatible with enzymes and microbes so that treatments can be carried out directly in the bioreactor without separate process steps. Firstly, the loadings of sodium sulfite and sodium dithionite were investigated in relation to the WIS content of a softwood slurry in a hybrid hydrolysis and fermentation scenario (Paper IV). In a second experimental series, different conditioning methods (sodium sulfite, sodium dithionite, sodium borohydride, hydrogenation, alkali treatment, and addition of activated charcoal) were compared (Paper IV). Thirdly, an evaluation of the impact of the different conditioning methods on slurry and hydrolysate was performed by determining the concentrations of inhibitors (Paper IV). A fourth series of experiments addressed how sulfite and dithionte affected the solid phase and enzymatic saccharification. Finally, an investigation of how pH, temperature, and loading of dithionite affected ethanol production from softwood hydrolysate was carried out (Paper V). Analyses of inhibitors, glucose consumption rates, ethanol yields, and toxicity were carried out.

The results revealed the positive impact of sodium dithionite and sodium sulfite on the fermentation process (Fig. 1, Paper IV). However, slurry with a WIS content of 15 or 20% required higher dosages of reducing agents than

35

slurry with 10% WIS (Fig. 1, Paper IV). Fig. 2 in Paper IV shows the correlation between WIS content and the dosage of reducing agent. It was observed that a ratio of mM reducing agent per percent of WIS of around 1 resulted in higher glucose consumption rate and improved fermentability (Fig. 2, Paper IV). Analytical enzymatic saccharification and determination of the sulfur content in the solid phase after treatment with sulfite and dithionite surprisingly showed sulfonation of the solid phase and improved saccharification, despite low concentrations of reducing agents and very mild reaction conditions (Table 6, Paper IV).

The determination of inhibitors showed that neither sodium sulfite nor sodium dithionite had any major effects on the concentrations of furan aldehydes, whereas sodium borohydride, calcium hydroxide, and treatment with activated charcoal decreased the concentrations of furan aldehydes (Table 8, Paper IV). Interestingly, the concentration of formaldehyde clearly decreased after treatment with sulfur oxyanions, and also after treatment with calcium hydroxide (Table 8, Paper IV). Hydrogenation resulted in efficient removal of coniferyl aldehyde and vanillin.

The results presented in Paper V demonstrate the importance of treatment conditions, such as temperature, pH, and loading of dithionite, on the fermentability of hydrolysates. Results show that lower or higher pH than typical fermentation pH (5.5) was of no advantage. Results also show that increased temperature to some extent can compensate for using lower concentrations of dithionite, which is likely advantageous in an industrial context.

36

9. Conclusions and further perspectives

Biorefining of lignocellulosic feedstocks is promising for reducing the dependence of fossil fuels and the emission of green-house gases. Fundamental knowledge about biochemical conversion of lignocellulose can be useful for improving the production of biofuels and green chemicals by making processing of biomass more competitive. The pretreatment step has large impact on biochemical conversion of lignocellulosic biomass to bio- based commodities. Hydrothermal pretreatment with and without externally added acid catalysts is one of the best investigated pretreatment methods. Still, further research is needed to understand basic effects of temperature, residence time, acid addition, and redox environment. The results of the studies showed fundamental aspects of hydrothermal pretreatment of sugarcane bagasse, Norway spruce and wheat straw providing a better understanding of the impact of pretreatment conditions on inhibitor formation, enzymatic digestibility, and fermentability. The investigation showed the significance of fundamental parameters, such as residence time, temperature, severity factor, and the addition of sulfuric acid, on the efficiency of pretreatment and on yields. Additionaly, the presence of gases, such as oxygen gas, affects the pretreatment conditions and could play an important role in the control and performance of hydrothermal pretreatments with and without acid catalysis. Further research and investigation is needed in this area to achieve higher yields of sugars and ethanol. Increased interest for high-solids loadings makes research on inhibition phenomena and process solutions aiming at decreasing inhibition problems more important. The results in this thesis may provide guidance for the development of more efficient biochemical conversion of lignocellulosic biomass with optimized hydrothermal pretreatment conditions. Nevertheless, further research and development is required for using lower dosages of enzymes and microorganisms, and for recycling of the biocatalysts. Challenges such as valorization of hydrolysis lignin and separation of real lignin and pseudo-lignin need to be further investigated. A broadening of the raw material supply, including further investigations on lignocellulosic residues and food wastes, is needed. Studies focusing on environmental aspects, such as green-house gas emissions, recycling of process water, and purification of effluents, are also needed.

37

10. Acknowledgments

I would like to thank all people who contributed and supported me during these studies. Firstly, I would like to express my gratitude to my main academic supervisor, Professor Leif Jönsson, for giving me the chance to become a PhD student, for his support and guidance throughout this journey and for sharing his knowledge It would have been impossible for me to complete this work without his mentoring and moral support. I want to extend a special thanks to my co-supervisors Carlos Martín and Björn Alriksson, the most helpful people imaginable, for great guidance and helpful criticism through my research studies and for sharing their experience and laboratory skills. You have always been there for me! A big thank goes to all of my colleagues within our research group for the great time. Special thanks to Stefan Stagge for help with inhibitor analyses, Guochao Wu for guidance about fermentation, and Madhavi Latha Gandla and Jenny Lundqvist for their invaluable support in the laboratory. Last but not least, I would like to thank my colleague and room-mate in the office in Örnsköldsvik, Jessica Gard Timmerfors for being a friend, supporting through the hard times, laughing and discussing along my PhD journey. I want to acknowledge collaborations with RISE Processum, RISE Research Institutes of Sweden, SEKAB E-Technology, and Lantmännen Agroetanol, and financial support from the Swedish Energy Agency, Bio4Energy, and Kempe Foundations for the research presented in this thesis. I would like also to express my gratitude to collaborators at MoRe Research, Biopolymer Analytical Platform (Junko Takahashi Schmidt), the Vibrational Spectroscopy Core Facility (András Gorzsás), and the NMR Core Facility (Mattias Hedenström). A big thank to all people and organizations on Domsjö Development Area in Örnsköldsvik for welcoming me and making my PhD journey just awesome! A very special thanks to my twin sister, Tania, and my brother, Kostas, and all family members for their love. A big thanks goes to my love, Daniel who has been with me on this journey in highs and lows. You deserve the best! Last but certainly not least, I would like to thank my parents for their endless support and help. My parents deserve credit for all my achievements. They have supported me whole heartedly in all my endeavours. You have taught me that even impossible dreams begin with a single step! Ιωάννη Ιλανίδη και Μελποµένη Ιλανίδου σας ευχαριστώ πολύ για όλα! Τα λόγια είναι πολύ λίγα για να περιγράψω όλα αυτά που κάνατε για εµένα και για την ευκαιρία που µου δώσατε να έρθω και να συνεχίσω τις σπουδές µου στην Σουηδία. Σας αγαπώ πολύ!

38

11. References

1. AFDC, Alternative Fuels Data Center. Available online: https://www.afdc.energy.gov/data/ (accessed on 25 March 2021).

2. Alriksson, B., Cavka, A., Jönsson, L.J., (2011). Improving the fermentability of enzymatic hydrolysates of lignocellulose through chemical in-situ detoxification with reducing agents. Bioresour. Technol. 102:1254–1263.

3. Alriksson, B., Sjöde, A., Nilvebrant, N.O., Jönsson, L.J., (2006). Optimal conditions for alkaline detoxification of dilute-acid lignocellulose hydrolysates. Appl. Biochem. Biotechnol. 129–132:599–611.

4. Arato, C., Pye, E.K., Gjennestad, G., (2005). The lignol approach to biorefining of woody biomass to produce ethanol and chemicals. Appl. Biochem. Biotechnol. 121:871-882.

5. Batista, G., Souza, R.B.A., Pratto, B., dos Santos-Rocha, M.S.R., Cruz, A.J.G., (2019). Effect of severity factor on the hydrothermal pretreatment of sugarcane straw. Bioresour. Technol. 275:321–327.

6. Berlin, A., Maximenko, V., Gilkes, N., Saddler, J., (2007). Optimization of enzyme complexes for lignocellulose hydrolysis. Biotechnol. Bioeng. 97:287–296.

7. Bezerra, T.L., Ragauskas, A.J., (2016). A review of sugarcane bagasse for second-generation bioethanol and biopower production. Biofuels, Bioprod. Bioref. 10:634–647.

8. Boerjan, W., Ralph, J., Baucher, M., (2003). Lignin biosynthesis. Ann. Rev. Plant Biol. 54:519–546.

9. Bommarius, A.S., Sohn, M., Kang, Y., Lee, J.H., Realff, M.J., (2014). Protein engineering of cellulases. Curr. Opin. Biotechnol. 29:139–145.

10. Brandt, A., Gräsvik, J., Hallett, J.P., Welton, T., (2013). Deconstruction of lignocellulosic biomass with ionic liquids. Green Chem. 15: 550–583.

11. Brienzo, M., Fikizolo, S., Benjamin, Y., Tyhoda, L., Görgens, J., (2017). Influence of pretreatment severity on structural changes, lignin content

39

and enzymatic hydrolysis of sugarcane bagasse samples. Renew. Energy 104:271–280.

12. Brownell, H.H., Saddler, J.N., (1984). Steam-explosion pretreatment for enzymatic hydrolysis. Biotechnol. Bioeng. Symp. 14:55–68.

13. Cantarella, M., Cantarella, L., Gallifuoco, A., Spera, A., Alfani, F., (2004). Comparison of different detoxification methods for steam-exploded poplar wood as a substrate for the bioproduction of ethanol in SHF and SSF. Proc. Biochem. 39:1533–1542.

14. Cavka, A., Alriksson, B., Ahnlund, M., Jönsson, L.J., (2011). Effect of sulfur oxyanions on lignocellulose-derived fermentation inhibitors. Biotechnol. Bioeng. 108:2592–2599.

15. Cavka, A., Jönsson, L.J., (2013). Detoxification of lignocellulosic hydrolysates using sodium borohydride. Bioresour. Technol. 136:368– 376.

16. Cavka, A., Stagge, S., Jönsson, L.J., (2015a). Identification of small aliphatic aldehydes in pretreated lignocellulosic feedstocks and evaluation of their inhibitory effects on yeast. J. Agric. Food Chem. 63:9747–9754.

17. Cavka, A., Martín, C., Alriksson, B., Mörtsell, M. and Jönsson, L.J., (2015b). Techno-economic evaluation of conditioning with sodium sulfite for bioethanol production from softwood. Bioresour. Technol. 196:129- 135.

18. Chang, V., Holtzapple M., (2000). Fundamental factors affecting biomass enzymatic reactivity. Appl. Biochem. Biotechnol. 537:84–86.

19. Chen, X., Li, H., Sun, S., Cao, X., Sun, R., (2018). Co-production of oligosaccharides and fermentable sugar from wheat straw by hydrothermal pretreatment combined with alkaline ethanol extraction. Ind. Crops Prod. 111:78-85.

20. Chum, H., Johnson, D., Black, S., Overend, R., (1990). Pretreatment – Catalyst effects and the combined severity parameter. Appl. Biochem. Biotechnol. 24:1–14.

21. Contreras, M.d.M., Romero, I., Moya, M., Castro, E., (2020). Olive- derived biomass as a renewable source of value-added products. Process Biochem. 97:43–56.

40

22. Costa, T.H.F., Kadic, A, Bengtsson, O., Lidén, G, Eijsink, V.G.H., Horn, S.J., (2020). Demonstration-scale enzymatic saccharification of sulfite- pulped spruce with addition of hydrogen peroxide for LPMO activation. Biofuels Bioprod. Bioref. 14:734–745.

23. Cruz, S.H., Dien, B.S., Nichols, N.N., Saha, B.C., Cotta, M.A., (2012). Hydrothermal pretreatment of sugarcane bagasse using response surface methodology improves digestibility and ethanol production by SSF. J. Ind. Microbiol. Biotechnol. 39:439–447.

24. Demirbas, M.F., Balat, M., (2007). Biomass pyrolysis for liquid fuels and chemicals: A review. J. Sci. Ind. Res. 66:797-804.

25. Dias, M.O.S.; Lima, D.R.; Mariano, A.P., 2018. Chapter 10 - Techno- economic analysis of cogeneration of heat and electricity and second- generation ethanol production from sugarcane. In: Chandel, A.K., Silveira M.H.L., (eds.). Advances in Sugarcane Biorefinery. Elsevier, Amsterdam. p. 197.

26. Du, B., Sharma, L.N., Becker, C., Chen, S.-F., Mowery, R.A., van Walsum, G.P, Chambliss, C.K., (2010). Effect of varying feedstock – Pretreatment chemistry combinations on the formation and accumulation of potentially inhibitory degradation products in biomass hydrolysates. Biotech. Bioeng. 107:430–440.

27. Eom, I.Y., Kim, K.H., Kim, J.Y., Lee, S.M., Yeo, H.M., Choi, I.G., Choi, J.W., (2011). Characterization of primary thermal degradation features of lignocellulosic biomass after removal of inorganic metals by diverse solvents. Bioresour. Technol. 102:3437–3444.

28. Faix, O., (1991). Classification of lignins from different botanical origins by FT-IR spectroscopy. Holzforschung 45:21–22.

29. Faix, O., (1991). Classification of lignins from different botanical origins by FT-IR spectroscopy. Holzforschung 45:21-22.

30. Faix, O., Böttcher, J.H., (1992). The influence of particle size and concentration in transmission and diffuse reflectance spectroscopy of wood. Holz. Roh. Werkst. 50:221–226.

31. FAOSTAT - Food and Agriculture Organization Corporate Statistical Database. Available online: http://www.fao.org/faostat/en/ #search/wheat (accessed 24 September, 2020).

41

32. Favaro, L., Jansen, T., van Zyl, W.H., (2019). Exploring industrial and natural Saccharomyces cerevisiae strains for the bio-based economy from biomass: The case of bioethanol. Crit. Rev. Biotechnol. 39:800–816.

33. Fernandes-Klajn, F., Romero-García, J.M., Díaz, M.J., Castro, E., (2018). Comparison of fermentation strategies for ethanol production from olive tree pruning biomass. Ind. Crops Prod. 122:98–106.

34. Fockink, D.H., Sanchez, J.H., Ramos, L.P., (2018). Comprehensive analysis of sugarcane bagasse steam explosion using autocatalysis and

dilute acid hydrolysis (H3PO4 and H2SO4) at equivalent combined severity factors. Ind. Crops Prod. 123:563–572.

35. Galbe, M., Zacchi., (2002). A review of the production of ethanol from softwood. Appl. Microbiol. Biotechnol. 59:618–628.

36. Galbe, M., Zacchi G., (2012). Pretreatment: the key role to efficient utilization of lignocellulosic materials. Biomass Bioenerg. 46:70–78.

37. Galbe, M., Wallberg, O., (2019). Pretreatment for biorefineries: A review of common methods for efficient utilisation of lignocellulosic materials. Biotechnol. Biofuels 12: 294.

38. Gandla, Μ.L., Martín, C., Jönsson, L.J., (2018). Analytical enzymatic saccharification of lignocellulosic biomass for conversion to biofuels and bio-based chemicals. Energies 11:2936.

39. García-Vargas, M.C., Contreras, M.d.M., Castro, E., (2020). Avocado- derived biomass as a source of bioenergy and bioproducts. Appl. Sci. 10:8195.

40. Garrote, G., Cruz, J.M., Dominguez, H., Parajo, J.C., (2008). Non- isothermal autohydrolysis of barley husks: Product distribution and antioxidant activity of ethyl acetate soluble fractions. J. Food Eng. 84:544–552.

41. Geddes, C.C., Mullinix, M.T., Nieves, I.U. Peterson, J.J., Hoffman, R.W., York, S.W., Yomano, L.P., Miller, E.N., Shanmugan, K.T., Ingram, L.O., (2011). Simplified process for ethanol production from sugarcane bagasse using hydrolysate-resistant Escherichia coli strain MM160. Bioresour. Technol. 102:2702–2711.

42

42. Gierer, J., (1970). Reactions of lignin during pulping: A description and comparison of conventional pulping processes. Svensk Papperstidn. 73:571–596.

43. Gräsvik, J., Winestrand, S., Normark, M., Jönsson, L.J., Mikkola, J.P., (2014). Evaluation of four ionic liquids for pretreatment of lignocellulosic biomass. BMC Biotechnol. 14:20.

44. Gremos, S., Zarafeta, D., Kekos, D., Kolisis, F., (2011). Direct enzymatic acylation of cellulose pretreated in BMIMCl ionic liquid. Bioresour. Technol. 102:1378–1382.

45. Guo, X., Cavka, A., Jönsson, L.J., Hong F., (2013). Comparison of methods for detoxification of spruce hydrolysate for bacterial cellulose production. Microb. Cell Fact. 12:93.

46. Haynes, W.M., (ed.), (2011). CRC Handbook of Chemistry and Physics (92nd ed.). CRC Press, Boca Raton, FL.

47. Hon, D.N.S., Shiraishi, N., (1991). Wood and Cellulosic Chemistry. Marcel Dekker, New York.

48. Horn, S.J., Vaaje-Kolstad, G., Westereng, B., Eijsink, V.G., (2012). Novel enzymes for the degradation of cellulose. Biotechnol. Biofuels 5:45.

49. Hu, F. and Ragauskas, A.J., (2012). Pretreatment and lignocellulosic chemistry. Bioenerg. Res. 5:1043–1066.

50. Jiang, Y.; Xin, F.; Lu, J.; Dong, W.; Zhang, W.; Zhang, M.; et al. (2017) State of the art review of biofuels production from lignocellulose by thermophilic bacteria. Bioresour. Technol. 245:1498–1506.

51. Johansen, K.S., (2016). Lytic polysaccharide monooxygenases: The microbial power tool for lignocellulose degradation. Trends Plant Sci. 21:926–936.

52. Jönsson, L.J., Alriksson, B., Nilvebrant, N.-O., (2013). Bioconversion of lignocellulose: inhibitors and detoxification. Biotechnol. Biofuels 6:16.

53. Jönsson, L.J., Martín, C., (2016). Pretreatment of lignocellulose: Formation of inhibitory by-products and strategies for minimizing their effects. Bioresour. Technol. 199:103–112.

43

54. Karatzos, S.K., Edye, L.A., Doherty, W.O.S., (2012). Sugarcane bagasse pretreatment using three imidazolium-based ionic liquids; mass balances and enzyme kinetics. Biotechnol. Biofuels 5:62.

55. Kieber, R.J., Mopper, K., (1990). Determination of picomolar concentrations of carbonyl compounds in natural waters, including seawater, by liquid chromatography. Environ. Sci. Technol. 24:1477– 1481.

56. Kim, S., Dale, BE., (2004). Global potential bioethanol production from wasted crops and crop residues. Biomass Bioenerg. 26:361–375.

57. Kim, Y., Kreke, T., Hendrickson, R., Parenti, J., Ladisch, M.R., (2013). Fractionation of cellulase and fermentation inhibitors from steam pretreated mixed hardwood. Bioresour. Technol. 135:30–38.

58. Klemm, D., Heublein, B., Fink, H.P., Bohn, A., (2005). Cellulose: Fascinating biopolymer and sustainable raw material. Angew. Chem. Int. Ed. Engl. 44:3358–3393.

59. Ko, J.K., Um, Y., Park, Y.C., Seo, J.H., Kim, K.H., (2015). Compounds inhibiting the bioconversion of hydrothermally pretreated lignocellulose. Appl. Microbiol. Biotechnol. 99:4201–4212.

60. Kollaras, A., Koutouridis, P., Moore, M.J.B., Bell, P.J.L., Philip, J.L., Attfield, P.V., (2011). Adding value to sugarcane bagasse using a xylose metabolising yeast. Int. Sugar J. 113:777-781.

61. Koppram, R., Tomás-Pejó, E., Xiros, C., Olsson, L., (2014). Lignocellulosic ethanol production at high-gravity: challenges and perspectives. Trends Biotechnol. 32:46–53.

62. Kricka, W., Fitzpatrick, J., Bond, U., (2014). Metabolic engineering of yeasts by heterologous enzyme production for degradation of cellulose and hemicellulose from biomass: A perspective. Front. Microbiol. 5:174.

63. Kristensen, J.B., Felby, C., Jørgensen, H., (2009). Yield-determining factors in high-solids enzymatic hydrolysis of lignocellulose. Biotechnol. Biofuels 2:11.

64. Kubicek, CP., (2013). Systems biological approaches towards understanding cellulase production by Trichoderma reesei. J. Biotechnol. 163:133-142.

44

65. Kudahettige, R.L, Holmgren, M., Imerzeel, P., Sellstedt, A., (2012). Characterization of bioethanol production from hexoses and xylose by the white rot fungus Trametes versicolor. Bioenerg. Res. 5:277–285.

66. Kumar, R., Wyman, C.E., (2014). Strong cellulase inhibition by mannan polysaccharides in cellulose conversion to sugars. Biotechnol. Bioeng. 111:1341–1353.

67. Larsson, S., Reimann, A., Nilvebrant, N.-O., Jönsson, L.J., (1999). Comparison of different methods for the detoxification of lignocellulose hydrolysates of spruce. Appl. Biochem. Biotechnol. 77-79:91-103.

68. Larsson, S., Quintana-Sáinz, A., Reimann, A., Nilvebrant, N.-O., Jönsson, L.J., (2000). Influence of lignocellulose-derived aromatic compounds on oxygen-limited growth and ethanolic fermentation by Saccharomyces cerevisiae. Appl. Biochem. Biotechnol. 84:617–632.

69. Leu, S.Y., Zhu, J.Y., (2013). Substrate-related factors affecting enzymatic saccharification of lignocelluloses: Our recent understanding. Bioenerg. Res. 6:405–415.

70. Li, Z., Xiao, D., Kong, Y., Ge, Y., (2015). Enhancing lead adsorption capacity by controlling the chain length of alkyl amine grafted lignin. Bioresources 10:2425–2432.

71. Li, F., Xie, G., Huang, J., Zhang, R., Li, Y., Zhang, M., et al., (2017). OsCESA9 conserved-site mutation leads to largely enhanced plant lodging resistance and biomass enzymatic saccharification by reducing cellulose DP and crystallinity in rice. Plant Biotechnol. J. 15:1093-104.

72. Limayem, A., Ricke, S.C., (2012). Lignocellulosic biomass for bioethanol production: Current perspectives, potential issues and future prospects. Prog. Energ. Combust. 38:449–467.

73. López-Linares, J.C., Gómez-Cruz, I., Ruiz, E., Romero, I., Castro, E., (2020). Production of ethanol from hemicellulosic sugars of exhausted olive pomace by Escherichia coli. Processes 8:533.

74. Lynd, L.R, Liang, X., Biddy, M.J., Allee, A., Cai, H., Foust, T., Himmel, M.E., Laser, M.S., Wang, M., Wyman, C.E., (2017). Cellulosic ethanol: Status and innovation. Curr. Opin. Biotechnol. 45:202–211.

45

75. Lynd, L.R., Weimer, P.J., van Zyl, W.H., Pretorious, I.S., (2002). Microbial cellulose utilization: fundamentals and biotechnology. Microbiol. Mol. Biol. Rev. 66:506–577.

76. Manochio, C., Andrade, B.R., Rodriguez, R.P., Moraes, B.S., (2017). Ethanol from biomass: A comparative overview. Renew. Sust. Energ. Rev. 80:743–755.

77. Marques, G., del Rio, J.C., Gutierrez, A., (2010). Lipophilic extractives from several nonwoody lignocellulosic crops (flax, hemp, sisal, abaca) and their fate during alkaline pulping and TCF/ECF bleaching. Bioresour. Technol. 101:260–267.

78. Martín, C., Wu, G., Wang, Z., Stagge, S., Jönsson, L.J., (2018) Formation of microbial inhibitors in steam-explosion pretreatment of softwood impregnated with sulfuric acid and sulfur dioxide. Bioresour. Technol. 262:242–250.

79. Martinez, A., Rodriguez, M.E., York, S.W., Preston, J.F., Ingram, L.O.,

(2000). Effects of Ca(OH)2 treatments (’’overliming’’) on the composition and toxicity of bagasse hemicellulose hydrolysates. Biotech. Bioeng. 69:526–536.

80. Mboowa, D., 2021. A review of the traditional pulping methods and the recent improvements in the pulping processes. Biomass Conv. Bioref. https://doi.org/10.1007/s13399-020-01243-6

81. McKendry, P., (2002). Energy production from biomass (part 1): Overview of biomass. Bioresour. Technol. 83:37–46.

82. Meier, D., Ante, R., Faix, O., (1992). Catalytic hydrolysis of lignin: Influence of reaction conditions on the formation and composition of liquid products. Bioresour. Technol. 40:171–177.

83. Meyer, A.S., Rosgaard, L., Sørensen, H.R., (2009). The minimal enzyme cocktail concept for biomass processing. J. Cereal Sci. 50:337–344.

84. Min, D., Wei, L., Zhao, T., Li, M., Jia, Z., Wan, G., Zhang, Q., Qin, C. Wang, S., (2018). Combination of hydrothermal pretreatment and sodium hydroxide post-treatment applied on wheat straw for enhancing its enzymatic hydrolysis. Cellulose 25:1197–1206.

46

85. Mitchell, V.D., Taylor, C.M., Bauer, S., (2014). Comprehensive analysis of monomeric phenolics in dilute acid plant hydrolysates. BioEnerg. Res. 7:654–669.

86. Nakasu, P.Y.S., Ienczak, L.J., Costa, A.C., Rabelo, S.C., (2016). Acid post- hydrolysis of xylooligosaccharides from hydrothermal pretreatment for pentose ethanol production. Fuel 185:73–84.

87. Nanda, S., Mohammad, J., Reddy, S.N., Kozinski, J.A., Dalai, A.K., (2014). Pathways of lignocellulosic biomass conversion to renewable fuels. Biomass Convers. Bior. 4:157-191.

88. Nelson, R.S., Stewart, Jr. C.N., Gou, J.Q., Holladay, S., Gallego-Giraldo, L., Flanagan, A., et al., (2017). Development and use of a switchgrass (Panicum virgatum L.) transformation pipeline by the BioEenergy Science Center to evaluate plants for reduced cell wall recalcitrance. Biotechnol. Biofuels. 10:309.

89. Nelson, M.L., O’Connor, R.T., (1964). Relations of certain infrared bands to cellulose crystallinity and crystal lattice type. Part I. Spectra of lattice types I, II, III and amorphous cellulose. J. Appl. Polym. Sci. 8:1311–1324.

90. Nilvebrant, N.O., Persson, P., Reimann, A., De Sousa, F., Gorton, L., Jönsson, L.J., (2003). Limits for alkiline detoxification of dilute-acid lignocellulose hydrolysates. Appl. Biochem. Biotechnol. 107:615–628.

91. Normark, M., Pommer, L., Gräsvik, J., Hedenström, M., Gorzsás, A., Winestrand, S., Jönsson, L.J., (2016). Biochemical conversion of torrefied Norway spruce after pretreatment with acid or ionic liquid. Bioenergy Res. 9:355–368.

92. Öhgren, K., Bura, R., Lesnicki, G., Saddler, J., Zacchi, G., (2007). A comparison between simultaneous saccharification and fermentation and separate hydrolysis and fermentation using steam-pretreated corn stover. Process Biochem. 42:834–839.

93. Overend, R.P., Chornet, E., (1987). Fractionation of lignocellulosics by steam-aqueous pretreatments. Phil. Trans. R. Soc. London, A321, 523– 536.

94. Panesar, P.S., Marwaha, S.S, Kennedy, J.F., (2006). Zymomonas mobilis: An alternative ethanol producer. J. Chem. Technol. Biot. 81:623–635.

47

95. Parajó, J., Domínguez, H., Domínguez, J., (1997). Xylitol production from Eucalyptus wood hydrolysates extracted with organic solvents. Process Biochem. 32:599–604.

96. Park, S., Baker, J.O., Himmel, M.E., Parilla, P.A., Johnson, D.K., (2010). Research Cellulose crystallinity index: measurement techniques and their impact on interpreting cellulase performance. Biotechnol. Biofuels. 3:10.

97. Persson, P., Larsson, S., Jönsson, L.J., Nilvebrant, N.-O., Sivik, B., Munteanu, F., Thörneby, L., Gorton, L., (2002). Supercritical fluid extraction of a lignocellulosic hydrolysate of spruce for detoxification and to facilitate analysis of inhibitors. Biotech. Bioeng. 79:694-700.

98. Pienkos, P.T., Zhang, M., (2009). Role of pretreatment and conditioning processes on toxicity of lignocellulosic biomass hydrolysates. Cellulose 16:743–762.

99. Pu, Y., Zhang, D., Singh, P.M., Ragauskas, A.J., (2007). The new forestry biofuels sector. Biofuels Bioprod. Biorefin. 2:58-73.

100. Puligundla, P., Smogrovicova, D., Mok, C., Obulam, V.S.R., (2019). A review of recent advances in high gravity ethanol fermentation. Renew. Energ. 133:1366-1379.

101. Qing, Q., Yang, B., Wyman, C.E., (2010). Xylooligomers are strong inhibitors of cellulose hydrolysis by enzymes. Bioresour. Technol. 101:9624–9630.

102. Ragauskas, A.J., (2013). Materials for Biofuels. Word Scientific, Singapore.

103. Reshamwala, S., Shawky, B.T., Dale, B.E., (1995). Ethanol production from enzymatic hydrolysates of AFEX-treated coastal Bermuda grass and switchgrass. Appl. Biochem. Biotechnol. 51:43–55.

104. Rocha, G.J.M., Martín, C., Silva, V.F.N., Gόmez, E.O., Gonçalves, A.R., (2012). Mass balance of pilot-scale pretreatment of sugarcane bagasse by steam explosion followed by alkaline delignification. Bioresour. Technol. 111:447–452.

105. Rødsrud, G., Lersch, M., Sjöde, A., (2012) History and future of world’s most advanced biorefinery in operation. Biomass Bioenerg. 46:46-59.

48

106. Romero-García, J.M., Niño, L., Martínez-Patiño, C., Álvarez, C., Castro, E., Negro, M.J., (2014). Biorefinery based on olive biomass. State of the art and future trends. Bioresour. Technol. 159:421-432.

107. Rowell, R.M., Pettersen, R., Han, J.S., Rowell, J.S., Tshabalala, M.A., (2005). Cell wall chemistry. In: Rowell R.M. (ed.) Handbook of Wood Chemistry and Wood Composites. CRC Press, Boca Raton, FL.

108. Ruiz, H.A., Rodríguez-Jasso, R.M., Fernandes, B.D., Vicente, A.A., Teixeira, J.A., (2013). Hydrothermal processing, as an alternative for upgrading agriculture residues and marine biomass according to the biore nery concept: a review. Renew. Sust. Energ. Rev. 21:35–51.

109. Saini, J.K., Saini, R., Tewari, L., (2015). Lignocellulosic agriculture wastes as biomass feedstocks for second-generation bioethanol production: Concepts and recent developments. 3 Biotech 5:337–53.

110. Sánchez, C., (2009). Lignocellulosic residues: Biodegradation and bioconversion by fungi. Biotechnol. Adv. 27:185–194.

111. Sánchez, O.J., Cardona C.A., (2008). Trends in biotechnological production of fuel ethanol from different feedstocks. Bioresour. Technol. 99:5270–5295.

112. Sannigrahi, P., Ragauskas, A., Miller, S., (2008). Effects of two-stage dilute acid pretreatment on the structure and composition of lignin and cellulose in Loblolly pine. Bioenerg. Res. 1:205–214.

113. Santo, M.E., Rezende, C.A., Bernardinelli, O.D., Pereira Jr, N., Curvelo, A.A.S., Azevedo, E.R., Guimaraes, F.E.G., Polikarpov, I., (2018). Structural and compositional changes in sugarcane bagasse subjected to hydrothermal and organosolv pretreatments and their impacts on enzymatic hydrolysis. Ind. Crops Prod. 113:64–74

114. Sassner, P., Galbe, M., Zacchi, G., (2008). Techno-economic evaluation of bioethanol production from three different lignocellulosic materials. Biomass Bioenerg. 32:422–430.

115. Schmatz, A.A., Tyhoda, L., Brienzo, M., (2020). Sugarcane biomass conversion in uenced by lignin. Biofuels Bioprod. Bioref. 14:469–480.

116. Sedlak, M., Ho, N.W.Y., (2004). Production of ethanol from cellulosic biomass hydrolysates using genetically engineered Saccharomyces yeast

49

capable of co fermenting glucose and xylose. Appl. Biochem. Biotechnol. 114:403–416.

117. Shinde, S.D., Meng, X., Kumar, R., Ragauskas, A.J., (2018). Recent advances in understanding the pseudo-lignin formation in a lignocellulosic biorefinery. Green Chem. 20:2192–2205.

118. Silveira, M.H.L., Morais, A.R.C., Lopes, A.M.C., Olekszyszen, D.N., Bogel- Łukasik, R., Andreaus, J., Ramos, L.P., (2015). Current pretreatment technologies for the development of cellulosic ethanol and biore neries. Chem. Sus. Chem. 8:3366–3390.

119. Singh, S.S., Lim, L.T., Manickavasagan, A., (2020). Imaging and spectroscopic techniques for microstructural and compositional analysis of lignocellulosic materials: a review. Biomass Conv. Bioref. https://doi.org/10.1007/s13399-020-01075-4

120. Singleton, V.L., Orthofer, R., Lamuela-Raventos, R.M., (1999). Analysis of total phenols and other oxidation substrates and antioxidants by means of Folin-Ciocalteu reagent. Method. Enzymol. 299:152–178.

121. Sjöström, E., (1993). Wood Chemistry Fundamentals and Applications, 2nd edition. Academic Press, Dan Diego.

122. Sluiter, J.B., Ruiz, R.O., Scarlata, C.J., Sluiter, A.D., Templeton, D.W., (2010). Compositional analysis of lignocellulosic feedstocks. 1. Review and description of methods. J. Agric. Food Chem. 58:9043–9053.

123. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., et al., (2012). Determination of structural carbohydrates and lignin in biomass. NREL/TP-510-42618. National Renewable Energy Laboratory, Golden, CO.

124. Sluiter, A., Hyman, D., Payne, C., Wolfe J., (2008) Determination of insoluble solids in pretreated biomass material. NREL / TP-510-42627. National Renewable Energy Laboratory, Golden, CO.

125. Stagge, S., Cavka, A., Jönsson, L.J., (2015). Identification of benzoquinones in pretreated lignocellulosic feedstocks and inhibitory effects on yeast. AMB Express 5:62.

126. Stenius, P., Vuorinen, T., (1999). Direct characterization of chemical properties of fibers. In: Sjöström, E., Alén, R. (eds.) Analytical Methods in

50

Wood Chemistry, Pulping and Papermaking. Springer-Verlag, Berlin. pp. 149–192.

127. Sues, A., Millati, R., Edebo, L., Taherzadeh, M.J., (2005). Ethanol production from hexoses, pentoses, and dilute-acid hydrolyzate by Mucor indicus. Yeast Res. 5:669–676.

128. Sullivan, J., Douek M., (1994). Determination of carbohydrates in wood, pulp and process liquor samples by high-performance anion-exchange chromatography with pulsed amperometric detection. J. Chromatogr. A 671:339-350.

129. Susmozas, A., Martín-Sampedro, R., Ibarra, D., Eugenio, M.E., Iglesias, R., Manzanares, P., Moreno, A.D., (2020). Process strategies for the transition of 1G to advanced bioethanol production. Processes 8:1310.

130. Teugjas, H., Väljamäe, P., (2013). Product inhibition of cellulases studied with 14C- labeled cellulose substrates. Biotechnol. Biofuels 6:104.

131. Van Dyk, J.S., Pletschke, B.I., (2012). A review of lignocellulose bioconversion using enzymatic hydrolysis and synergistic cooperation between enzymes—Factors affecting enzymes, conversion and synergy. Biotechnol. Adv. 30:1458–1480.

132. Vanholme, B., Desmet, T., Ronsse, F., Rabaey, K., Van Breusegem, F., De Mey, M., Soetaert, W., Boerjan, W., (2013). Towards a carbon-negative sustainable bio-based economy. Front. Plant Sci. 4:174.

133. Vanholme, R., Morreel, K., Darrah, C., Oyarce, P., Grabber, J.H., Ralph, J., Boerjan, W., (2012). Metabolic engineering of novel lignin in biomass crops. New Phytol. 196:978–1000.

134. Varnai, A., Huikko, L., Pere, J., Siika-Aho, M., Vilkari, L., (2011). Synergistic action of xylanase and mannase improves the total hydrolysis of softwood. Bioresour. Technol. 102:9096–9104.

135. Wang, Z., Gräsvik, J., Jönsson, L.J., Winestrand, S., (2017). Comparison - - - of [HSO4] , [Cl] and [MeCO2] as anions in the pretreatment of aspen and spruce with imidazolium-based ionic liquid. BMC Biotechnol. 17:82.

136. Wang, Z., Wu, G., Jönsson, L.J., (2018). Effects of impregnation of softwood with sulfuric acid and sulfur dioxide on chemical and physical

51

characteristics, enzymatic digestibility, and fermentability. Bioresour. Technol. 247:200–208.

137. Ximenes, E., Kim, Y., Mosier, N., Dien, B., Ladisch, M., (2011). Deactivation of cellulases by phenols. Enzyme Microb. Technol. 48:54– 60.

138. Xu, N., Zhang, W., Ren, S., Liu, F., Zhao, C., Liao, H., et al., (2012). Hemicelluloses negatively affect lignocellulose crystalllinity for high

biomass digestibility under NaOH and H2SO4 pretreatments in Miscanthus. Biotechnol. Biofuels 5:58.

139. Yamashita, Y., Kurosumi, A., Sasaki, C., Nakamura, Y., (2008). Ethanol production from paper sludge by immobilized Zymomonas mobilis. Biochem. Eng. J. 42:314–319.

140. Yang, B., Dai, Z., Ding, S.Y., Wyman, C.E., (2011). Enzymatic hydrolysis of cellulosic biomass. Biofuels 2:421–450.

141. Yu, Z., Jameel, H., Chang, H., Park, S., (2011). The effect of delignification of forest biomass on enzymatic hydrolysis. Bioresour. Technol. 102:9083–9.

142. Zabed, H., Sahu, J.N., Boyce, A.N., Faruq, G., (2016). Fuel ethanol production from lignocellulosic biomass: An overview on feedstocks and technological approaches. Renew. Sust. Energ. Rev. 66:751–774.

143. Zhai, R., Hu, J., Saddler, J.N., (2016). What are the major components in steam pretreated lignocellulosic biomass that inhibit the efficacy of cellulase enzyme mixtures? ACS Sust. Chem. Eng. 4:3429−3436.

144. Zhong, N., Chandra, R., Saddler, J.N., (2019). Sulfite post-treatment to simultaneously detoxify and improve the enzymatic hydrolysis and fermentation of a steam-pretreated softwood lodgepole pine whole slurry. ACS Sust. Chem. Eng. 7:5192–5199.

52