Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The impact of endogenous acetylation on the deconstruction of Populus trichocarpa wood during pretreatment Johnson, Amanda M. 2015

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2015_september_johnson_amanda.pdf [ 11.65MB ]
JSON: 24-1.0166522.json
JSON-LD: 24-1.0166522-ld.json
RDF/XML (Pretty): 24-1.0166522-rdf.xml
RDF/JSON: 24-1.0166522-rdf.json
Turtle: 24-1.0166522-turtle.txt
N-Triples: 24-1.0166522-rdf-ntriples.txt
Original Record: 24-1.0166522-source.json
Full Text

Full Text

The impact of endogenous acetylation on the deconstruction of Populus trichocarpa wood during pretreatment   by  Amanda M. Johnson  B. Sc. The University of Ottawa, 2009  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Forestry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2015 © Amanda M. Johnson, 2015 ii  Abstract Wood is a renewable, sustainable and economic resource. Non-cellulosic wood polysaccharides are acetylated in a species-dependent and spatially-regulated manner. Acetyl groups comprise approximately 5% of the dry weight of poplar wood. Biologically, acetyl groups increase xylan chain solubility and may therefore influence secondary cell wall formation by affecting the association of hemicelluloses with the other cell wall components, such as cellulose, lignin, and the pectic and proteinaceous constituents of the adjacent primary cell wall. In this research, we found that acetyl content correlated positively with lignin (R = 0.28) and negatively with cellulose (R = -0.41) in wood samples from a common garden of over 200 unrelated P. trichocarpa individuals. During the pretreatment of biomass, acetyl groups are hydrolyzed from wood to form free acetic acid in the reaction media. The present research examined the relationships between wood composition and pretreatment sugar yield of 19 black cottonwood (P. trichocarpa) genotypes with varying levels of acetylation. The results clearly show a strong correlation (R > 0.77) between acetic acid and polysaccharide dissolution. Fast-reacting xylan had a degree of acetylation of 0.35, while slow-reacting xylan had a degree of acetylation of 0.73. Hydrophobic interactions could explain the negative correlation between lignin content and acetic acid released (R = -0.59). This research highlights the impact acetylation may have on the large-scale industrial utility of plants.   iii  Preface Trees in the Black Cottonwood Growth Trial are clones from a common garden established by the British Columbia Ministry of Forests, Lands and Natural Resource Operations. Harvest of the trees, wood processing, and the initial analysis for wood chemistry traits was conducted as part of the Genome BC PopCan project.   Experimental design, and method optimization and pretreatment vessel innovation are accredited to A.M. Johnson and S.D. Mansfield.   iv  Table of Contents Abstract…………………………………………………………………..………………………………..……ii Preface…………………………………………………………………………………………………………iii Table of Contents………………………………………….…………………………………………………iv List of Tables………………………………………………………………………………………..………..vii List of Figures……………………………………………………………………...………………….…….viii List of Symbols……………………………………………………………………..…………………………x Acknowledgements……………………………………………………………………………….…………xii Chapter 1 Introduction .................................................................................................................... 1 1.1 Poplar ................................................................................................................................. 1 1.1.1 Ecology ....................................................................................................................... 1 1.1.2 Poplar research using Black Cottonwood Growth Trial ............................................. 2 1.2 Poplar Wood, Cell Walls and Acetylation ......................................................................... 3 1.2.1 Wood anatomy and development ................................................................................ 3 1.2.2 Components of the secondary cell wall ...................................................................... 4 1.2.3 Modification of xylans: acetyl groups and glucuronic acid ........................................ 8 1.2.4 Wood chemistry variation within tree ....................................................................... 10 Between growth rings ........................................................................................ 10 Non-normal wood .............................................................................................. 11 Green wood versus dry wood ............................................................................ 12 1.2.5 Storage and acetylation ............................................................................................. 12 1.3 Renewable Energy and Cellulosic Ethanol ...................................................................... 13 1.3.1 Overview ................................................................................................................... 13 1.3.2 Thermochemical conversion of biomass during pretreatment .................................. 15 Types of pretreatment ........................................................................................ 15 Deconstruction during pretreatment: mechanisms and products ....................... 17 Pretreatment effects on cell wall ultrastructure ................................................. 20 Impact on downstream hydrolysis and fermentation ......................................... 21 1.3.3 Chemical substituents affect hemicellulose dissolution ........................................... 22 v  1.4 Scope and Aim ................................................................................................................. 23 Chapter 2 Materials and Methods ................................................................................................. 25 2.1 Sampling and Processing Wood from Growth Trial ........................................................ 25 2.2 Survey of Wood Acetylation ........................................................................................... 26 2.2.1 Saponification of acetyl groups from poplar wood ................................................... 26 2.2.2 Effects of storage and processing methods on acetyl content ................................... 27 2.3 Chemical Transformations of P. trichocarpa Wood During Pretreatment ...................... 28 2.3.1 Dilute acid hydrothermal pretreatment ..................................................................... 28 2.3.2 Pretreatment in acetic acid ........................................................................................ 30 2.3.3 Temperature gradient ................................................................................................ 31 2.3.4 Post-pretreatment processing .................................................................................... 31 2.3.5 Calculations and statistical analysis .......................................................................... 33 Chapter 3 Results .......................................................................................................................... 36 3.1 Survey of Growth Trial for Wood Acetylation ................................................................ 36 3.1.1 Method development: saponification of O-acetyl groups from poplar wood ........... 36 3.1.2 Variability of acetylation over time and samplings .................................................. 38 3.1.3 Acetylation results .................................................................................................... 38 3.1.4 Acetyl levels by provenance ..................................................................................... 39 3.1.5 Correlations analysis ................................................................................................. 44 3.2 Chemical Transformations of P. trichocarpa Wood During Pretreatment ...................... 47 3.2.1 Reaction bombs evaluated ........................................................................................ 47 3.2.2 Chemical transformations of poplar wood at different severities ............................. 48 Acetic acid enhances sugar solubilisation and dehydration .............................. 48 Higher temperature enhances sugar solubilisation and dehydration ................. 51 Sugar solubilisation and dehydration at increasing severities .......................... 53 3.2.3 Chemical transformation of different genotypes ...................................................... 64 Material balance of wood from candidate genotypes ........................................ 64 Pretreatment performance by genotype ............................................................. 66 Correlations between wood chemistry and compounds formed ........................ 69 3.3 Other Results .................................................................................................................... 75 3.3.1 NMR assessment ....................................................................................................... 75 vi  3.3.2 Expression of putative O-acetyltransferases ............................................................. 75 Chapter 4 : Discussion .................................................................................................................. 76 4.1 Survey of a Black Cottonwood Growth Trial for Acetylation ......................................... 76 4.1.1 Method development ................................................................................................ 76 4.1.2 Phenotypic associations ............................................................................................ 76 4.1.3 Spacial and temporal variability in acetylation ......................................................... 78 4.2 Chemical Transformations of P. trichocarpa Wood During Pretreatment ...................... 78 4.2.1 Small-scale assay versus industrial-scale pretreatment ............................................ 78 4.2.2 Mode of heat transfer in pretreatment assay ............................................................. 79 4.2.3 Chemical transformations by varying treatment severity ......................................... 81 Acetic acid enhances sugar release during pretreatment ................................... 81 Temperature selection ........................................................................................ 82 Selection of dilute acid pretreatment regime ..................................................... 82 4.2.4 Chemical transformation of different genotypes ...................................................... 85 Zero-acetyl negative control ............................................................................. 85 Assessing sugar release by genotype ................................................................ 87 Lignin impedes the release of acetyl groups as acetic acid .............................. 87 Sugar release by genotype: the numbers ........................................................... 88 Acetylation renders xylan fast- or slow-reacting .............................................. 89 Chapter 5 : Conclusions & Future Research ................................................................................. 92 5.1 Summary .......................................................................................................................... 92 5.2 Future Work ..................................................................................................................... 94 5.2.1 Substituent effects: glucuronic acid .......................................................................... 94 5.2.2 Localization of O-acetyl groups ................................................................................ 95 5.2.3 Genetic basis for O-acetylation ................................................................................. 96  References…………………………………………………………………..……………………..…………97 Appendices……………………………………………………………………………….…………………108    vii  List of Tables Table 1. Average acetyl content in clones from 34 provenances in Totem Field ......................... 42 Table 2. Association of wood quality and biomass traits………………………………..……....59 Table 3. Dissolution and degradation of wood at various dilute acid pretreatments. ................... 59 Table 4. Fate of acetyl groups following 12 pretreatments .......................................................... 63 Table 5. Chemical composition of the wood of selected individuals ........................................... 65 Table 6. Dissolved xylose and glucose following pretreatment. .................................................. 68 Table 7. Association between wood components before and after pretreatment, ........................ 72  Appendix Table 1. Inject-to-inject repeatability of Dionex AS50 Autosampler ........................ 110 Appendix Table 2. Effect of saponifying wood meal on cell wall constituents. ........................ 111 Appendix Table 3. Effect of acetone extraction on acetyl content. ............................................ 112 Appendix Table 4. Effect of baking overnight at 105°C on acetyl content ................................ 113 Appendix Table 5. Effect of storage in the cold room on acetyl content.. ................................. 114 Appendix Table 6. Provenance drainages of trees used in the study.. ........................................ 118 Appendix Table 7. Repeatability of glucose, xylose and acetyl release ..................................... 124 Appendix Table 8. Composition of wood residue ollowing pretreatment regimes.. .................. 128 Appendix Table 9. Acetyl content for identical wood samples analyzed two years apart.. ....... 129   viii  List of Figures Figure 1. Structure of hardwood O-acetyl-4-O-methylglucuronoxylan ......................................... 8 Figure 2. Model of the secondary cell wall before and after pretreatment. .................................. 15 Figure 3. Carbohydrates in poplar wood and their conversion to inhibitors ................................ 20 Figure 4. Saponification mechanism: Nucleophilic acyl substitution. ......................................... 27 Figure 5. Dilute acid pretreatment vessel.. ................................................................................... 30 Figure 6. Flowchart for the quantitative analysis of compounds before and after pretreatment. . 33 Figure 7. Saponification protocol optimization ............................................................................ 37 Figure 8. Acetyl of extractive-free wood (% w/w) by provenance. .............................................. 41 Figure 9. Acetyl of extractive-free wood (% w/w) by geographic cluster. ................................... 43 Figure 10. Release of xylose (a) and glucose (b), and their degradation products (c) t.. ............. 50 Figure 11. Temperature optimization of pretreatment regime. ..................................................... 53 Figure 12. Pretreatment optimization: photograph of solid and liquid phases.. ........................... 58 Figure 13. Dissoled xylose and glucose following five pretreatment regimes. ............................ 60 Figure 14. Dissolved furfural and 5-hydroxymethyl-2-furaldehyde t. ......................................... 61 Figure 15. Fate of acetyl groups on poplar wood following five dilute acid pretreatments.. ....... 62 Figure 16. Fate of acetyl groups in nineteen genotypes following dilute acid pretreatment. ....... 69 Figure 17. Relationship of total lignin to total xylose and glucose dissolved. . ........................... 73 Figure 18. Relationship of acetic acid to dissolved xylose, glucose and degradation products. .. 74  Appendix Figure 1. Sample HPX87-H chromatographs……………………………………….109 Appendix Figure 2. Difference between two technical replicates in acetyl content. .................. 115 Appendix Figure 3. Acetyl content (% w/w) in selected angiosperms and gymnosperms. ........ 116 Appendix Figure 4. Map showing provenance of trees used in this study……………………..114 Appendix Figure 5. Biomass accumulation and appearance of tree cookies .............................. 119 Appendix Figure 6. Biological repeatability of acetylation ........................................................ 120 ix  Appendix Figure 7. Expression of six putative O-acetyltransferases in developing xylem. ...... 121 Appendix Figure 8. NMR of poplar total cell wall material………………………………........119 Appendix Figure 9. Sample CarboPak PA1 anion exchange column………………………….120 Appendix Figure 10. Pressure versus time in three stainless steel reaction vessels.. ................. 125 Appendix Figure 11. Effect of heat treatment on a sugar standard solution .......................... 126 Appendix Figure 12. Insoluble lignin in wood residue following each pretreatment regime. ... 127 Appendix Figure 13. Relationship of dissolved wood and acetic acid to oligomeric xylose. .... 130  x  List of Symbols °C Degrees Celsius %SM Percent solid material following pretreatment % w/v Percent expressed as weight over volume % w/w Percent expressed as weight of component over total weight % v/v Percent expressed as volume over volume AA Acetic acid Acet or OAc O-acetyl groups Ara Arabinose DBH Diameter at Breast Height FA Formic acid F.G. O-Formyl groups Gal Galactose Glu Glucose H2SO4 Sulphuric acid HMF 5-hydroxymethyl-2-furaldehyde HPLC High Performance Liquid Chromatography Hrs Hours L.I. Insoluble lignin L.S. Soluble lignin M Molar (moles per litre) Man Mannose M.C. Moisture content (in % w/w) Mg/g Milligrams per gram Min Minutes  NaOH Sodium hydroxide nm Nanometer (10-9 metre) O:M Oligomer-to-monomer ratio R Pearson product-moment correlation coefficient R2 Coefficient of determination Rha Rhamnose SD Standard deviation of the sample, 𝑆𝐷 =    ??? ????  XOS Xylooligosaccharides xi  Xyl Xylose  xii  Acknowledgements Many thanks to S. M. Mansfield for his research ideas and guidance, and C. J. Douglas and J. N. Saddler for providing sound scientific advice. Lab members S. Barkwill, R. Chandrasekhar, L. Da Ros, E. Gonzalez-Vigil, J. F. Hart, X. Huang, A. Kenawy, S. Lazarova, V. J. Maloney, B. Marco de Lima, G. McNair, Y. Mottiar, E. Murphy, F. Raiche De Araujo, M. Roach, A. Skyba, S. Smigiel, N. Ukrainetz, F. Unda, S. M. von Loessl, and D. Yao were important collaborators. Brainstorming sessions with R. Chandra, K. Semple, S. Burkhardt, B. Coetzee, P. A. Chung, A. McKown and M. Piddocke helped carry the research to completion. Research funding from the Working on Walls (WoW) NSERC Create Training Program is gratefully acknowledged.  1  Chapter 1 …Introduction 1.1 Poplar 1.1.1 Ecology There are 30 tree species in the genus Populus, comprising approximately ten percent of the family Salicaceae (Isebrands & Richardson, 2014). Poplar species are classified into six sections: Populus, the white poplars and aspens; Aigeiros, the cottonwoods and black poplar; Tacamahaca, the balsam poplars; Leucoides, the swamp poplars, Abaso, Mexican poplar; and Turanga, Afro-asian poplars. 40% of Earth’s natural poplar stands are in Canada, amounting to approximately 28.3 million hectares (Isebrands & Richardson, 2014). Poplar grows rapidly and can produce large volumes of wood in a relatively short time relative to other tree species, especially those in North America1.  The largest of the poplar trees is Populus balsamifera spp. trichocarpa Torrey & Gray, reaching 50 m in height and accruing diameters of ~1.5 m (Pojar et al., 1994). Also known as black cottonwood, P. trichocarpa grows west of the Cascade Mountains at latitudes spanning from Baja, California to Alaska. Black cottonwood is a dominant species at low to medium elevations, mainly on the sandy, porous soils of stream banks and flood plains (Little et al., 1980). P. trichocarpa is an early-successional, shade-intolerant species with grey bark that is smooth on young trees and becomes ridged and furrowed with age. Leaves are broadly ovate; shiny dark green above, and glaucous light grey-green underneath. P. trichocarpa is the three-                                                1 33.3 m3/ha/yr for poplar hybrid Populus trichocarpa x Populus deltoids 53-242 (Mike Carlson, unpublished data).   2  carpelled sister species to P. balsamifera, with which it frequently interbreeds in areas where the ranges overlap.   For a variety of reasons, P. trichocarpa is a model species in plant biology. The widespread distribution of the sister species P. balsamifera in North America is a testament to its vigour across biomes2. The growth characteristics of poplar make it amenable to breeding and genetic research, as the trees often reach reproductive maturity in 10 years. Additionally, poplar can be coppiced, clonally propagated, and genetically transformed by Agrobacterium with relative ease (Mellerowicz et al., 2001). In 2006, it was the first woody plant to have its genome sequenced (Tuskan et al., 2006).  Poplar is currently valued for the production of wood products plywood, veneer, fibreboard, particleboard, oriented strandboard, wood-cement composites, wood-plastic composites, glued structural products, and pulp and paper (Isebrands & Richardson, 2014). Other products, such as pallets, matchsticks, wooden pencils (pencil stock) and upholstered furniture frames, are also made from poplar wood (Wood Database). 1.1.2 Poplar research using Black Cottonwood Growth Trial  The present research stemmed from a multidisciplinary, large-scale genomics project, PopCan, which aimed to advance poplar trees as a Canadian bioenergy feedstock by highlighting genotype-phenotype interactions. This comprehensive project used association and expression analyses to elucidate the trait diversity in both P. balsamifera and P. trichocarpa. A common                                                 2The natural range of P. balsamifera covers vast areas across North America from southwestern Alaska to Newfoundland. 3  garden experiment of wild, unrelated black cottonwood (P. trichocarpa) genotypes was extensively phenotyped and analyzed.  The black cottonwood growth trial was initially established in Surrey, British Columbia in 2000 by the British Columbia Ministry of Forests. Since then, this trial has helped elucidate key aspects of poplar phenotypic variability, such as variation in cell wall chemistry and ultrastructural traits (Porth et al., 2013). Genome-wide association studies also identified hundreds of unique biomass, ecophysiology and phenological related genes (McKown et al., 2014). Three processes contributing to genetic variation were hybridization, limited gene flow, and natural selection (Geraldes et al., 2014). Moreover, moderate to high broad-sense heritability were observed for phenology, biomass and ecophysiology traits, and single nucleotide polymorphisms for 16 wood chemistry and ultrastructural traits individually explained 3-7% of phenotypic variance (Porth et al., 2013). The plasticity of Populus trichocarpa is apparent in the phenotypic diversity of Totem Field. 1.2 Poplar Wood, Cell Walls and Acetylation 1.2.1 Wood anatomy and development Xylem tissue conducts water and solutes from the ground to the photosynthetic tissues of the tree and provides mechanical support. Xylem is fibrous because it is largely comprised of fibres, a specialized cell type. These fibres are long, needle-like, and their walls are heavily lignified and thickened for mechanical support. Other specialized cell types in xylem, vessel elements, render xylem porous. Vessel elements have relatively large diameters to allow for the conduction of water. Both vessel elements and fibres are prosenchymatous, or non-living, at 4  maturity. Xylem also contains parenchyma cells, which function in sapwood to store and transport carbohydrates.  Development of vessel elements and fibres occurs in five stages: initiation, elongation, cell wall thickening, lignification, and programmed cell death. The first step is the periclinal division of a secondary meristem called the vascular cambium. This forms cambial initials, which grow and terminally differentiate into vessel elements or fibres. Cell wall thickening involves the deposition of cell wall material outside the plasma membrane in layers known as lamellae. Lignification adds reinforcement to the thickened cell wall. The final step in the terminal differentiation of vessel elements and fibres is programmed cell death. Following completion of these five processes, the skeleton of the cell is formed.  1.2.2 Components of the secondary cell wall  During cell elongation, the primary cell wall is deposited outside the cell membrane. When the cell has reached its final size, secondary cell wall biosynthesis and deposition increases the thickness of the cell wall fifty-fold (Fengel & Wegener, 1983). The secondary cell wall is primarily composed of the macromolecules cellulose, lignin and hemicellulose.  Cellulose is made of β-D-glucose and comprises roughly half of the secondary cell wall. The configuration of glucose into chains, crystalline lattices, and winding helices of cellulose microfibrils is elegant. The smallest repeating unit of cellulose is cellobiose, structurally two glucose molecules rotated 180° relative to one another and linked by a β-(1→4)-glycosidic bond. Depending on plant species, the cellulose polymer can reach up to ten thousand glucose units in length. The equatorial positioning of the ring substituents allows individual cellulose chains to come into close proximity to one another. This facilitates another level of organization of 5  cellulose in the third dimension: the crystalline lattice. Intermolecular hydrogen bonds form between OH-groups of adjacent cellulose chains, forming a microfibril about 3 nm in diameter (Albersheim et al., 2010). Water molecules form layers between adjacent cellulose chains as they hydrogen bond to hydroxyl groups on cellulose.  Another level of cellulose organization is the arrangement of cellulose microfibrils along the fibre longitudinal axis. Based on this arrangement, three distinct layers can be delineated in the secondary cell wall. These cell wall layers are termed S1, S2 and S3 to reflect the order of deposition. The S2 layer makes up the largest proportion of the secondary cell wall and is primarily responsible for the load-bearing capacity of the secondary cell wall. In the S2 layer, cellulose microfibrils are aligned parallel to one another and wind as a right-handed helix around the longitudinal axis of the cell. The steepness of the helical winding is measurable, and is indicative of the microfibril angle. The microfibril angle affects at least three physical properties of wood: stiffness, shrinkage upon drying, and overall flexural properties. Fibres with a larger microfibril angle have a lower modulus of elasticity. For example, juvenile wood has a higher microfibril angle than mature wood, and as such is more flexible and can bend in response to wind. In contrast, mature wood generally has a lower microfibril angle and increased stiffness. This enhances the load-bearing capacity of the wood. The microfibril angle of the S2 layer of poplar is about 16° (Bendtsen & Senft, 1986). In the S1 and S3 layers of the secondary cell wall, cellulose microfibrils are arranged in both right- and left-handed helices along the fibre wall. These layers have a higher microfibril angle than the S2 layer. In addition to cellulose, the S1, S2 and S3 layers contain the polymers hemicellulose and lignin. Lignin is the second most abundant biopolymer in wood, comprising 20-30% of the secondary cell wall by weight. Lignin maintains the strength, hydrophobicity, and pathogen 6  resistance of the tree. The three main phenylpropanoid subunits comprising lignin are syringyl, guaiacyl and p-coumaryl. Lignin subunits are polymerized in muro following cellulose and hemicellulose deposition. While cellulose microfibrils extend in two-dimensional space, lignin is amorphous and fills the cell wall in three dimensions. In lignin, the dominant inter-unit linkages are ether bonds, joining the aromatic head of one phenylpropane unit to the aliphatic tail of another. β-O-4 and β-5 linkages are examples of ether linkages commonly found between lignin phenylpropane subunits. Other prominent linkages are head-to-head carbon-carbon or ether bonds (5-5 and 5-O-4, respectively), or tail-to-tail carbon-carbon bonds (5-5).  After cellulose and lignin, hemicelluloses make up most of the remainder of the cell wall structure. Comprising 15-30% of the secondary cell wall, hemicelluloses are defined as the alkaline-soluble constituents of a cell wall from which pectic substances have been removed. Hemicelluloses belong to a family of matrix polysaccharides made of pentose, hexose, hexuronic acid and deoxy-hexose sugar building blocks. Like cellulose, hemicelluloses consist of a backbone of sugars linked by β-1,4-glycosidic bonds. Unlike cellulose, hemicelluloses are of variable composition, shorter in length (DP < 500) and highly branched. Branching impedes hydrogen bonding, rendering hemicelluloses neither load-bearing nor crystalline in structure.  Four hemicelluloses are commonly found in angiosperms: xyloglucan, mixed linkage glucans, mannans, and xylan (Albersheim et al., 2010). Xyloglucan is the prevalent hemicellulose in the primary cell wall of all vascular plants. The structure of xyloglucan is a β-(1→4)-linked glucan chain with α-(1→6)-linked xylose branches. Mixed-linkage glucans, found in the primary walls of grasses, are so-called because of the mixed bond type. Glucosyl subunits in mixed-linkage glucans are either β-(1→3)- or β-(1→4)-linked, so that the final hemicellulose polymer takes on a crimped form. Mannans, particularly glucomannans, are the main 7  hemicelluloses fortifying cell walls of conifers. Xylans are the main hemicelluloses in the secondary cell wall of hardwoods. Xylans have a backbone of β-(1→4)-linked xylose residues. The xylose subunits are frequently substituted with O-acetyl groups, 4-O-methylglucuronic acid, and glucuronic acid in dicot secondary walls (Figure 1). The dominant xylan in commelid monocots is glucuronoarabinoxylan. Glucuronoarabinoxylan is glucurono- and arabinosylated, and occasionally esterified with ferulic acid at the O-5 position of arabinofuranosyl residues (Hatfield 1999). Ferulates decorating glucuronoarabinoxylan can dimerize to form intra- and intermolecular cross links with hemicellulose and with lignin. Thus, in grasses, lignin and hemicellulose are covalently linked in what has been termed the lignin-carbohydrate complex (Vanholme et al., 2010). Another feature of xylans is a shared reducing-end motif: the oligosaccharide β-D-Xylp-(1→4)-β-D-Xylp-(1→3)-α-L-Rhap-(1→2)-α-D-GalpA-(1→4)-D-Xylp (Rennie & Scheller, 2014). Xylan, lignin and cellulose are the polymers maintaining the structural integrity of the secondary cell wall. Small amounts of pectin, extractives and ash are also present.  Pectin is a polymer rich in D-galacturonic acid. Overall, pectin comprises about 2% (w/w) of poplar wood, however, pectin is prevalent in tensionwood (see and in the middle lamella. Non-structural compounds found in wood include extractives and ash. Extractives, whose primary purpose is defence against herbivory, can make up between 1-5% poplar wood dry weight. They are concentrated in ray parenchyma and can contain terpenes and terpenoids, fats and waxes, phenols, lignans and quinones, and tannins and flavonoids. Ash, which comprises 0.2 to 0.5% of wood from temperate zones, is largely made of calcium, potassium, and magnesium. 8   Figure 1. Structure of hardwood O-acetyl-4-O-methylglucuronoxylan showing O-Acetyl and 4-O-Methylglucupyranosyluronic acid substituents (Patil et al., 2012). Bolded numbers indicate substituent position on the pyranosyl subunit.  1.2.3 Modification of xylans: acetyl groups and glucuronic acid Acetyl groups occur on carbons 2 or 3 of xylose and are present at roughly every second xylosyl residue in xylan (Busse-Wicher 2014). This modification could serve two biological functions. First, acetylation could facilitate the transport and insertion of xylan precursors into the cell wall. Unlike cellulose, which is synthesized and deposited in muro, hemicellulose subunits are synthesized in the Golgi apparatus and subsequently transported to the cell wall. Maintaining solubility of hemicellulose precursors in the aqueous cytoplasm is important. Acetyl groups prevent intramolecular hydrogen bonding, preventing aggregation of xylan precursors in transit to the cell wall. The second biological function xylan acetylation serves is to alter mechanical properties of the assembled secondary cell wall. Xylan acetylation increases chain stiffness and decreases hydrogen bonding (Chen et al., 2005). Chemical acetylation of wood is used to improve its mechanical properties. Acetylation increases the modulus of rupture and modulus of elasticity of wood (Youngquist et al., 1986).   O-Acetyl 4-O-methyl glucuronic acid Xylose Xylose Xylose Xylose 9  Acetyl groups are also hydrophobic and readily interact with lignin. The positioning of acetyl groups towards lignin in an assembled cell wall was recently shown (Busse-­‐‑Wicher et al., 2014). The xylan backbone is ribbon-like and interacts with the hydrophilic surfaces of cellulose. Consequently, hydrophobic acetyl groups face away from the cellulose microfibril, positioned toward the lignified region of the cell wall. Thus, acetyl groups are responsible for the amphiphilic nature of xylan, mitigating the interfacial tension that would otherwise occur at the lignin-cellulose boundary (Utracki, 2002). Acetyl groups increase the strength of the cell wall by hydrophobically interacting with lignin to improve cohesion between lamellae.   Another xylan modification is glycosidic linkage of glucuronic acid (GlcA) or 4-O-methylglucuronoxylan (4-O-MeGlcA) to xylose (Figure 1). Like acetyl groups, uronic acid groups are added to carbons 2 or 3 of xylopyranosyl residues. The proportion of xylosyl residues substituted with uronic acids is approximately 1:8. Of these substitutions, the proportion of GlcA to 4-O-MeGlcA in glucuronoxylan is approximately 3:5. Xylan is also substituted with O-formyl groups, arabinose and, more rarely, galactose and xylose (Scheller & Ulvskov, 2010; Ebringerova & Heinze, 2000).  A recent knockout-and-rescue experiment suggested that acetyl groups and glucuronic acid moieties on xylan are functionally equivalent (Xiong et al., 2015). However, kinetic studies on wood pulping pointed out that the acetyl groups and glucuronic acid moieties have distinct biochemical properties and would therefore have different functions in the secondary cell wall (Kuitunen et al., 2013). O-acetyl groups may, however, exhibit similar biochemical reactivity to methylester groups, found on pectin (Pawar et al., 2013).  10  Genetic studies have identified proteins responsible for O-acetylation of xylan. In primary cell walls where load-bearing is less important, acetylation can be reduced without growth penalties. A study by Gille et al. examined 125 Arabidopsis thaliana ecotypes and found a three-fold difference in xyloglucan acetylation with no decrease in ecological fitness (Gille et al., 2011). Cell walls of A. thaliana reduced wall acetylation mutants (rwa), containing 20% less acetyl units, were shown to be structurally sound (Manabe et al., 2011). Conversely, evidence suggests that acetylation is necessary for maintaining the structural integrity of the secondary cell wall. Mutants in the gene ESKIMO1 (esk1), with a 32% decrease in acetylation, accumulated less biomass (Xiong et al., 2013;Yuan et al., 2013). In Arabidopsis, Trichome Birefringence-like 29 (TBL29) proteins add acetyl groups to xylan, with Reduced Wall Acetylation (RWA) proteins possibly serving as acetyl-CoA transporters (Rennie & Scheller, 2014). TBL29/ESK1 T-DNA insertion lines, with a 50% reduction in acetylation, had an irregular xylem phenotype (Xiong et al., 2013). These two studies suggest that acetylation imparts important structural and physiological properties to xylan, which ultimately manifests in larger secondary cell wall phenotypes. 1.2.4 Wood chemistry variation within tree Between growth rings Tree rings are the result of secondary growth from the vascular cambium. A ring is formed in one growing season, and comprises earlywood and latewood. Earlywood is formed in the spring and is specialized for water transport, having larger vessel diameters and thinner cell walls. Latewood is formed in the summer and consists of a higher proportion of fibres for mechanical support. The boundary between successive years’ growth is visible where the dark 11  latewood of one season meets the light earlywood of the next. Carbon can be differentially partitioned from year to year in response to climate variables such as CO2 and water availability (Runion et al., 1999). Therefore, growth ring number is an important source of wood chemistry variation.  Specifically, the cellulose, lignin and acetyl content of wood can vary from pith to bark. In a study of poplar clones, cellulose increased 15% (w/w) between the innermost and 11th growth ring (Sheng-zuo & Wen-zhong, 2003). Another study examining poplar wood found that lignin content decreased by one-third and glucan content increased by one-third in growth rings measured from pith to bark (DeMartini & Wyman, 2011). Xylan content in the same study followed no trend. Acetyl content is lower in the innermost heartwood rings of a tree than in sapwood (Krilov & Lasander, 1988). From the above studies, it can be deduced that wood chemistry is largely dependent upon the location of sampling within the tree. Non-normal wood Wood is non-homogenous, with growth-related ‘defects’ exhibiting different chemical and mechanical properties. These are referred to as non-normal wood. Two examples of non-normal wood are tension wood and knots.  Tension wood development can be induced by mechanical stress or by gravitational stimuli on the upper side of branches or leaning stems of angiosperm trees (Isebrands & Parham, 1974). Tension wood is characterized by thicker cell walls, a higher fibre-to-vessel ratio and an additional cell layer often referred to as gelatinous layer (Mizrachi et al., 2014). The gelatinous layer (G layer) is composed mainly of cellulose (70% w/w). Polysaccharides xyloglucan, pectin, and mannan make up the remainder of the G layer at proportions of 15, 8, and 2 %, respectively 12  (Al-Haddad et al., 2013). Tension wood has a lower microfibril angle compared to normal wood, and a higher elastic energy (Clair et al., 2011). Another type of non-normal wood exists in knots. In aspen, knots have been shown to contain 9% (w/w) extractives, about three times higher than that in stem wood (Pietarinen et al., 2006). Green wood versus dry wood Wood chemistry can also change depending on the physiological state of the wood. Green wood is wood that has been freshly cut. The lumina of vessels in green wood are partially filled with water so that green wood contains 150% weight in water relative to its dry weight (Simpson & TenWolde, 1999). In dry wood, all water has been removed from cell lumina. This is known as the equilibrium moisture content, and amounts to approximately 10% in juvenile poplar wood. At equilibrium moisture content, water is restricted to—and bound to—the cell wall through hydrogen bonds with cellulose and hemicellulose. When the water content of wood is reduced to 0% by oven-drying, hornification occurs. This is when removal of water from between adjacent cellulose microfibrils in the crystalline lattice results in an irreversible reduction of cellulose hygroscopicity. Following hornification, new hydrogen bonds form between adjacent cellulose chains to replace those previously occupied by water molecules. 1.2.5 Storage and acetylation Several studies suggest that acetyl groups are released from wood as acetic acid vapours when wood is stored in a humid environments (Arni et al., 1965; Bouveng, 1961). Birch, oak and Douglas-fir lose over half their acetyl content after two years of storage under warm, damp 13  conditions (Packman, 1960). Even at ambient conditions with lower humidity, a 25% - 50% reduction in acid content of beech and ash has been shown following 10 months in storage (Ucar & Ucar, 2008).  The emission of acetic acid during storage does not seem to depend on the total acetyl content of wood, but rather depends on the type of wood. For instance, oak wood chips with somewhat lower acetyl content than beech emit more acetic acid (Choon & Roffael, 1990). Moisture content of the wood likely has one of the largest effects on acetyl loss, as spring-felled green woods were shown to release more acid than the corresponding winter or autumn specimens (Arni et al., 1965). 1.3 Renewable Energy and Cellulosic Ethanol 1.3.1 Overview  There has been a global shift towards renewable energy in order to preserve the climate for future generations. Renewable fuels contributed 5.8% of the total energy consumed in Canada in 2012 (Canada’s Energy Future 2013). The large-scale implementation of renewable fuels requires remodelling current policies, programs and regulations. Moreover, there are technological barriers for the biochemical conversion of biomass into bio-fuel on a large scale. The oil and gas industry has undergone rapid transformation in technologies used to extract fossil fuels in recent years. These challenges and changes in other industries inspire the question, can similar innovations expedite cellulosic ethanol production in the renewable fuels sector? Generally, carbohydrates in lignocellulosic biomass can be biochemically converted to ethanol in four steps: size reduction, pretreatment, hydrolysis, and fermentation. Following the 14  mechanical breakdown of biomass into suitably-sized particles, a pretreatment regime must be applied, where the biomass ultrastructural and molecular properties are altered to increase enzymatic accessibility (Figure 2). Hydrolysis uses enzymes, or acid, to convert polysaccharides into monomeric sugars, which then are converted by microorganisms to value-added products by fermentation, yielding ethanol.  The economical production of lignocellulosic ethanol requires low-cost feedstocks that are tailored for the above processes. Poplar has emerged as one such promising biofuel feedstock. However, as with other types of lignocellulosic biomass, lignin is a key technical barrier for the poplar-to-ethanol bioconversion (Mansfield et al., 2012). To overcome technical and economical barriers in lignocellulosic ethanol production, various pretreatments can be applied.   15   Figure 2. Pretreatment schematic showing the deconstruction of lignocellulosic biomass. Enlarged is O-acetyl substituent on hemicellulose, which is hydrolyzed to acetic acid during pretreatment (diagram modified from, accessed July 2015).  1.3.2 Thermochemical conversion of biomass during pretreatment  Types of pretreatment Pretreatments use physical and/or chemical means to deconstruct biomass, usually with the addition of a catalyst. Most pretreatment used in lignocellulosic bioenergy production are modified pulping processes. The most common pretreatments for lignocellulosic biomass are steam-catalyzed explosion, dilute acid, ammonia fiber explosion, and organosolv.  Dilute acid is a leading pretreatment platform. In this method, sulphuric acid (or less commonly nitric acid, hydrochloric acid and phosphoric acid) catalyzes breakdown of the Cellulose Hemicellulose Lignin                         Physical Chemical	  Biological	   PRETREATMENT O-­‐acetyl	  group Acetic	  acid 16  biomass. High temperature and pressure are also required. Dilute acid pretreatment dissolves hemicellulose and redistributes lignin, while cellulose is generally retained in the crystalline state. Dilute acid pretreatment works effectively on most feedstocks. Poplar inherently contains a high proportion of hemicellulose and syringyl-rich lignin and is particularly well-suited to dilute acid pretreatment. One study of poplar wood pretreatment showed that 90% of xylan dissolves following one-minute at 180°C and 0.9% sulphuric acid (Esteghlalian et al., 1997). Moreover, dilute acid pretreatment is low-cost and requires no apparatus other than a sealable reactor, and only small amounts of the comparatively inexpensive catalyst sulphuric acid. For these reasons, dilute acid pretreatment was the method of choice for the current research.  One form of dilute acid pretreatment is steam explosion. This uses low concentrations of chemicals and consumes relatively low amounts of energy (Bura et al., 2009). During steam catalyzed pretreatment, biomass3 is subjected to steam at 160-290°C for a short duration (1 – 5 minutes), and then decompressed rapidly. Steam pretreatment is suitable for feedstocks such as poplar and agricultural residues, which contain a high proportion of hemicellulose, and syringyl and parahydroxy-derived lignin, which are less recalcitrant to bioconversion (Brownell & Saddler, 1987). Following steam pretreatment, lignin is partially solubilized and virtually all hemicelluloses are solublized (Mosier et al., 2005). Similar to steam pretreatment, ammonia fiber explosion (AFEX) uses high temperature and pressure to alter the compact structure of the biomass, but is catalyzed by liquid ammonia instead of SO2. AFEX uses chemical and physical means to increase the surface area of biomass, partially solubilize lignin, hydrolyze hemicellulose, and decrystallize cellulose (Bals et al., 2010). This type of pretreatment is commonly employed on agricultural residues and forage                                                 3Biomass can be SO2-impregnated for SO2-catalyzed steam explosion.  17  grasses. For example, AFEX-pretreated corn stover is 90% converted to glucose following hydrolysis at a moderate enzyme loading (Kim et al., 2002). AFEX has the advantages of a high recovery rate, negligible degradation of cellulose and hemicellulosic sugars, and a product that does not need to be neutralized prior to enzymatic hydrolysis (Teymouri et al., 2005). However, AFEX is ineffective on softwoods and only partially effective on hardwoods (McMillan, 1994). Organosolv pretreatment uses organic solvents such as ethanol, acetone and acetic acid. Since the aromatic lignin polymer is more soluble in organic solvents than in aqueous solvents, this pretreatment removes more lignin from the substrate. This results in higher enzymatic digestibility of otherwise recalcitrant softwood residues (Pan et al., 2005; Pan et al., 2006). For example, organosolv fractionation of mountain pine beetle-killed lodgepole pine yielded 97% cellulose-to-glucose conversion within 48 hours (Pan et al., 2007). Lignin extracted from organosolv processes has an added advantage, as it can be used to manufacture valuable co-products such as bio-based thermoplastics (Kadla et al., 2004). Earth-metal or acid salts have also been supplemented to the solvent to act as a catalyst and improve overall efficacy of the orgonosolv pretreatment. Deconstruction during pretreatment: mechanisms and products Pretreatment removes hemicellulose and amorphous cellulose from the cell wall matrix by hydrolyzing β-(1→4)-glycosidic bonds and breaking hydrogen bonds. The addition of heat and presence of a catalyst during pretreatment provides sufficient energy to protonate the oxygen atom in a glycosidic bond. Once the transition state has been reached, nucleophilic attack of an adjacent carbocation by an aqueous hydroxyl group is the final step in the reaction. Hydrolysis of xylose to xylooligomers has an activation energy of ~170 kJ/mol, while hydrolysis of 18  xylooligomers to xylose, and subsequently dehydration of xylose to furfural, requires 130 kJ/mol (Borrega et al., 2011). The activation energies for hydrolysis of glucan chains are similar. During pretreatment, some lignin is hydrolyzed via the cleavage of β-O-4 linkages (Yelle et al., 2013). Products of lignin deconstruction during pretreatment include free phenolics such as hydroxybenzaldehyde, syringaldehyde, and vanillin (Fengel & Wegener, 1983). During pretreatment, ester-linked substituents are hydrolyzed by nucleophilic acyl substitution to their corresponding carboxylic acid.  In poplar glucuronoxylan, acetyl groups are hydrolyzed to acetic acid and formyl groups to formic acid. In grass glucuronoarabinoxylan, ferulates are hydrolyzed to ferulic acid.  The sequence of cell wall deconstruction depends on the pretreatment platform. In dilute acid pretreatment, the first stage of pretreatment releases non-stuctural cell wall moieties (extractives, ash and soluble sugars) into the liquid phase (Conner, 1984). Of the structural cell wall material, xylan is removed first, as polysaccharide chains of less than twenty-five residues (Pu et al., 2013). As pretreatment progresses, polysaccharides are hydrolyzed to oligomers of three to nine residues in length. Next, oligosaccharides from xylan and glucan are hydrolyzed to monosaccarides. Galacturonyl side chains are decarboxylated to form CO2, and acetyl side chains are hydrolyzed to form free acetic acid (Leschinsky et al., 2009).  Hydrolysis acts further on monomeric sugars in the acyclic form. β-elimination, hydride shift, and/or (1→2)-enolization reactions form furfural from pentose sugars and uronic acid (Danon et al., 2014). Similarly, 5-hydroxymethyl-2-furaldehyde (HMF) is formed from hexose sugars by these same reactions. Furthermore, the addition of two water molecules to HMF produces levulinic acid and formic acid (Figure 3). 19  The degradation products mentioned above can combine with free phenolics to form aromatics such as 3-dihydroxybenzoic acid, 1-(3,4-dihydroxy-6-methylphenyl)-2-hydroxyethanone, 5,6-2-methyl-benzofuran, and 3-hydroxy-6-hydroxymethyl-2-methylchromone (Fengel & Wegener 1983). These products are pseudo-lignin, which can be erroneously counted as Klason lignin following pretreatment (Sannigrahi et al., 2011). Other compounds resulting from pretreatment include levoglucosan, erythrose, methylglyoxal, glycolaldehyde and glycolic acid (Conner, 1984; Fengel & Wegener, 1983). There are many uses for the xylose extracted from wood during pretreatment. It can be used to produce value-added products, including bio-polymers, artificial sweeteners and organic acids (Werpy et al., 2004). Alternatively, the xylose can be fermented to ethanol or other fermentation products, but these processes are inefficient to date (Harner et al., 2015).   20    Figure 3. Carbohydrates in poplar wood and their conversion to inhibitors during hydrothermal pretreatment. Hemicellulose is hydrolyzed into acetic acid and its constituent sugars glucuronic acid, arabinose, xylose, mannose, galactose and glucose. The former three are dehydrated into furfural, while the latter three are dehydrated into 5-hydroxymethyl-2-furaldehyde. Further pretreatment causes 5-hydroxymethyl-2-furaldehyde dehydration into levulinic acid, and furfural and 5-hydroxymethyl-2-furaldehyde degradation into formic acid (Palmqvist & Hahn-Hägerdal, 2000). Pretreatment effects on cell wall ultrastructure In assessing suitability of a particular biomass for enzymatic hydrolysis, cell wall ultrastructure is equally as important as cell wall chemistry. Pretreatment increases surface area, improving accessibility of enzymes to their substrates. An increase in paracrystalline cellulose has also been shown (Sun et al., 2014). A decrease in cellulose DP and decrease in amorphous 21  cellulose has also been reported (Yang & Wyman, 2008). Porosity increases as hemicellulose is dissolved (Ishizawa et al., 2007). These changes in cell wall ultrastructure improve the efficiency of enzymatic hydrolysis. Impact on downstream hydrolysis and fermentation Different pretreatments vary temperature, time, pressure and/or catalyst to obtain a biomass with a defined degree of deconstruction. Pretreatment uses extreme temperature, pressure and acidity to remove hemicelluloses from lignocellulosic biomass. The resulting wood residue has more-accessible cellulose and altered lignin. The next step in bioethanol production is the enzymatic hydrolysis of the residue using biopolymer-specific enzymes. After enzymatic hydrolysis, glucose is liberated from cellulose and metabolized into ethanol by microorganisms. Efficient sugar fermentation is dependent upon the health of fermentation microorganisms. The pretreatment liquor must provide them with a suitable environment to metabolize, grow and reproduce (Keating et al., 2006). In most cases, pretreatment slurries are washed and neutralized to physiological pH prior to enzymatic hydrolysis (Mosier et al., 2005). Levels of furfural, HMF and acetic acid must be kept below thresholds of 100 mM (Larsson et al., 1999). Pretreatment can be tailored to meet the specific requirements of enzymatic hydrolysis and microbial fermentation. In an ideal pretreatment, most of the hemicellulose-derived sugars are recovered and lignin is easily fractionated, resulting in a cellulosic residue readily hydrolyzed by commercial enzymes (Chen et al., 2012). An oligomer-to-monomer ratio of greater than ten is ideal, as it is associated with less sugar degradation (Pu et al., 2013). It is clear that pretreatment affects all downstream processes of bioconversion.  22  The other major factor affecting bioconversion is the nature of the starting biomass. Sugar release following enzymatic hydrolysis has been shown to be inversely proportional to lignin content and proportional to the ratio of syringyl-to-guiacyl subunits (Studer et al., 2011). Lignin and xylans have a high affinity for the enzyme cellulase and can bind it non-productively, resulting in saccharification losses (Zhang et al., 2012). Acetyl groups have also been shown to block access of cellulases to β-(1→4)-glycosidic linkages in cellulose, decreasing sugar yields (Pan et al., 2006; Chang & Holtzapple, 2000). In the context of pretreatment outcomes, structural and compositional aspects of hemicellulose will be discussed. 1.3.3 Chemical substituents affect hemicellulose dissolution In poplar wood, hemicelluloses comprise glucuronoxylan (20-30% w/w), mannan (2-5% w/w) and xyloglucan (~3% w/w). The three exhibit different reactivities during dilute acid pretreatment. The relative proportions of xylan, mannan and glucan components, as well as how they are modified, can lead to varying pretreatment results. Glucuronoxylan is more accurately termed O-acetyl-(4-O-methyl)glucuronoxylan to account for its two main substituents, O-acetyl and 4-O-methylglucuronic acid. The principal aim of pretreatment is to remove O-acetyl-(4-O-methyl)glucuronoxylan from wood by cleavage of β-(1→4)-glycosidic bonds. The cleavage of β-(1→4)-glycosidic bonds and, therefore, the removal of hemicellulose from wood is affected by both O-acetyl and 4-O-methylglucuronic acid substituents. Removal of O-acetyl-(4-O-methyl)glucuronoxylan from wood occurs in two simultaneous phases: a fast phase and a slow phase. During the fast phase, fast-reacting xylan is hydrolyzed; next, during the slow phase, slow-reacting xylan is removed. Structures for the two types of xylan have not been proposed (Shen & Wyman, 2011). It is highly plausible that 23  conformational, steric and inductive effects induced by chemical substituents determine whether xylan is fast- or slow-reacting. An early study suggested that slow-reacting xylan is ‘embedded within’ lignin to a higher degree than fast-reacting xylan (Conner 1984). The acetic acid formed during pretreatment can catalyze hydrolysis of β-(1→4)-glycosidic linkages. However, several studies suggest the acetyl group facilitates hyrophobic interactions with lignin (Busse-­‐‑Wicher et al., 2014), which would slow xylan removal during pretreatment. 1.4 Scope and Aim The research for this thesis was conducted on the premises that there is natural variation in acetylation of poplar cell walls, acetylation affects sugar release during pre-hydrolysis, and selection of genotypes varying in acetyl content could elucidate mechanisms of cell wall deconstruction.  The objectives of the first part of the project were to survey a black cottonwood common garden experiment for acetyl content and select trees varying in acetyl content. This required developing a method for rapidly and accurately quantifying acetyl groups in woody biomass. The acetyl content of over 200 individual trees representing nearly 100 genotypes was examined.  The second part of the project aimed to investigate the impact of various pretreatment regimes on sugar release following pretreatment of P. trichocarpa wood. To conduct this assessment an optimized microscale pretreatment regime was established. Samples were pretreated over temperature and time gradients, as well as concentration gradients of catalysts acetic and sulphuric acid. 24  Finally, the fate of cell wall components in selected poplar genotypes following pretreatment was determined. Genotypes varying in inherent acetyl content were pretreated under optimized conditions. This elucidated patterns of cell wall deconstruction, including the contribution of the acetyl group. This third experiment extends the knowledge base of acetylation and pretreatment by varying acetyl content non-intrusively in P. trichocarpa.    25  Chapter 2 Materials and Methods 2.1 Sampling and Processing Wood from Growth Trial Trees used for this study were Populus balsamifera subsp. trichocarpa (Torr. & A. Gray ex Hook.) clones originating from a common garden established by the British Columbia Ministry of Forests. A total of 854 clones were collected from 188 provenances spanning 44-60°N along 36 river drainages in the Pacific Northwest coast (Appendix Figure 4, Appendix Table 6). The physiological age of trees from which scions were collected was 10 to 25 years. Scions were propagated in a stool bed in Surrey, British Columbia, at the Ministry of Forests and Range Nursery. From the stool bed, a common garden experiment was established in April 2000. Trees were planted in 1 × 1m plots. A split-plot design with five blocks enabled clones to be grouped according to provenance, and provenances according to drainages. This minimized the shading of slow-growing individuals from faster-growing individuals. In the first growing season only, trees were irrigated four times daily for 30 minutes (Xie et al., 2009). A large-scale black cottonwood phenotyping experiment was established at the University of British Columbia. Stecklings for 456 genotypes were grown under greenhouse conditions and planted in Totem Field in June 2008. The identity of the trees was confirmed using Illumina iSelect Infinium sequencing (Geraldes et al., 2014). Trees were planted in a common field using a 1.5 × 1.5m random block design. In the first year after planting, trees were watered when necessary using a drip irrigation system, and the field was mowed regularly to prevent weed overgrowth (McKown et al., 2013). 26  After four years growth, two trees from each clone were harvested during daylight hours in March 2012. Cookies were cut 6” above the base of the tree and air-dried. Wood free of bark and pith was ground to pass a 40-mesh using a Wiley mill. Acetone-soluble extractives were removed by 16-hour reflux in a Soxhlet apparatus, and the samples were stored at room temperature prior to analysis. Initial carbohydrate composition was determined by High Pressure Anion Exchange Liquid Chromatography (HPAELC) following a two-stage acid hydrolysis reaction. Acid-soluble lignin was determined spectrophotometrically according to TAPPI Useful Method A205, while acid-insoluble lignin was determined gravimetrically. 2.2 Survey of Wood Acetylation 2.2.1 Saponification of acetyl groups from poplar wood Saponification was the method used to quantify acetyl groups. Saponification is a nucleophilic acyl substitution, reacting a base with an ester to yield an alcohol and a salt. When protons are added to solution, the salt forms a carboxylic acid (Figure 4). The reaction mechanism proceeds in three steps: First, a hydroxide ion adds to the carbonyl group of I, forming the orthoester intermediate II. Next, the alkoxide (O-Rʹ′) becomes the leaving group as the ester reforms, and the irreversible deprotonation of III forms an alcohol (HO-Rʹ′) and a sodium salt (IV). Finally, acidification of the solution results in the formation of the carboxylic acid V (acetic acid where R = CH3). Acetic acid was measured by HPLC. A saponification protocol (Browning, 1967) was optimized by varying sample mass, NaOH concentration, and reaction time. 0.2M NaOH was added to wood at a 2% solids loading, to establish a reaction with an initial pH of approximately 12. Butyric acid was used as the internal standard, and the reactant incubated at 60°C with constant shaking (500 rpm) for 75 27  minutes. Following incubation, each sample was acidified to pH 2 ± 1 with 72% (w/w) sulphuric acid and cooled in an ice bath for 5 minutes. The tubes were centrifuged at 13 000 g for 2 minutes, the supernatant filtered through a 0.45 um filter into an HPLC vial. Acetic acid content was quantified on an HPX-87H column (Aminex, USA) by High Pressure Liquid Chromatography (HPLC; Dionex) equipped with an ASI-100 Autosampler, a P60 HPLC Quartinary Gradient Pump, and a PDA-100 Photodiode Array Detector (Dionex, USA). The mobile phase was 5 mM sulphuric acid flowing at 0.7 mL/min, and the detector measured absorbance at 205 nm. This first variation of the saponification method was used in determining the acetyl content in solid wood samples. Peaks were integrated manually (Appendix Figure 1).   Figure 4. Saponification mechanism showing the nucleophilic acyl substitution reaction between hydroxide and an ester carbon. Above, O-acetyl group is saponified with sodium hydroxide. I, ester; II, othoester intermediate; III, carboxylic acid; IV, salt; V, carboxylic acid (modified from Dr. I. R. Hunt).  2.2.2 Effects of storage and processing methods on acetyl content Acetyl in untreated versus extractives-free poplar wood was compared. First, the acetone extractives content of a bulk poplar sample was determined, following TAPPI method T 204 cm-28  97, to be 2.0% w/w. Next, samples were acetone-extracted as outlined in 2.1. A control sample was not extracted. Both samples were oven-dried at 105°C overnight prior to analysis. The 2.0% w/w extractives content was corrected for in the control group. For each treatment, seven technical replicates were performed. Since samples had been baked at least twice prior to acetyl analysis4, a quality-control experiment pertaining to oven-drying samples was performed. We tested acetyl content in wood following successive overnight baking at 105°C. Samples were placed in an oven at 105°C for 0, 1, 2, 3, and 4 nights, allowing for re-equilibration to ambient air temperature for 10 hours between.  The next quality-control experiment sought to determine whether wood stored green differs in acetyl content from air-dried wood. We sampled two corresponding wood samples; one sample had been dried soon after felling and stored at room temperature, and the other sample was green wood, which had been stored in the cold room (4°C, 60% relative humidity) for two years. Moisture content was measured after storage and ranged from 7 to 11% (w/w).  2.3 Chemical Transformations of P. trichocarpa Wood During Pretreatment 2.3.1          Dilute acid hydrothermal pretreatment Logs which had been stored at 4°C and 60% relative humidity for two years were re-sampled by cutting a disk approximately 1” from the base. Non-extracted, never-dried wood free of bark and pith was ground to pass 40-mesh using a Wiley-mill and stored at -20°C prior to use                                                 4The same wood samples were concurrently being tested for wood chemistry traits; thus, when a given sample was analyzed for acetyl content it had already been oven-dried and equilibrated to room temperature at least once.  29  (Cullis et al., 2004). Pretreatment conditions were first optimized by varying temperature, time, and sulphuric acid concentration. Poplar wood flour, 250 mg, was mixed with water or dilute sulphuric acid (0.1%, 0.3% or 0.6% w/w) at 5% (w/w) solids loading. Reaction vessels were stainless steel cylinders constructed from HPLC columns (300 mm long x 7.8 mm diameter) with a bolt-screw fitting on either end (Figure 5).   Prior to closing the reactors, both fittings were sealed with Teflon tape to prevent steam from escaping. Sealed reactors were vortexed for ~15 seconds, and subsequently shaken end-wise to ensure dispersion of wood in the liquid phase. Preincubation was conducted at 60° C for 60 minutes to ensure impregnation of the wood sample with dilute acid or water. The pretreatment was then carried out in a Lindberg / Blue M GO1330A laboratory gravity oven (Thermo Scientific, USA) set to 180°C for 10, 30 or 60 (± 0.5) minutes. After pretreatment, the reactors were cooled by immediately submerging in an ice bath for one minute.  The pressure inside the reactor was measured by attaching a 2,000 psi pressure gauge (Ashcroft, USA) to bold-screw fitting via 30 cm of stainless steel tubing. The stainless steel tubing enabled non-intrusive monitoring of reactor pressure during pretreatment. Time zero was taken to be the time at which reactors were placed in the oven.   Five pretreatment vessels were adapted in-house for these experiments; thus, samples were pretreated in batches of five. Batchwise repeatability was assessed by comparing the standard deviation of glucose, xylose and acetyl release following pretreatment of an identical sample in three different batches (Appendix Table 7). Three technical replicates were performed for each sample. To control for minor aberrations in pretreatment time, technical replicates were always pretreated in separate batches. 30   Pretreatment liquor and wood residue were transferred quantitatively to 15 mL falcon tubes and wrapped in aluminum foil to prevent UV-catalyzed degradation of furans. Samples were stored at 4°C for a maximum of three days prior to HPLC analysis.    Figure 5. Dilute acid pretreatment vessel. 1, plastic screw cap for bolt screw fitting; 2, bolt screw fitting; 3, stainless steel reaction chamber. Asterisks indicate where Teflon tape was wrapped as a sealant (image modified from BC Tech).   2.3.2      Pretreatment in acetic acid   Pretreatment was carried out with added acetic acid (3, 6 and 9% (v/v)) in order to investigate acid-reinforced autohydrolysis of poplar wood. To 250 mg of poplar wood, 100 mL of aqueous acetic acid solution was added. As a control, the treatment was also carried out in water only. Reaction vessels were 200 mL serum bottles that were sealed with a rubber stopper and an aluminum crimp cap. Samples were preincubated for 60 minutes at 60°C prior to pretreatment at 121°C in an autoclave for 60 minutes. Analysis of oligomers, monomers and degradation products is described in 2.3.4. 31  2.3.3         Temperature gradient  To investigate the effects of temperature on sugar release and degradation, pretreatment was carried out at ten degree increments between 160°C and 200°C. No acid catalyst was added. Pretreatment time was 60 minutes. Solids loading, reaction vessels, preincubation and analysis of the liquid product are described in 2.3.4.  2.3.4         Post-pretreatment processing  Wood was processed and the solid and liquid phases separately analyzed for components (Figure 6 shows a concept schematic). After pretreatment, a sample of the liquid product was directly analyzed by HPAELC to quantify the free monosaccharides. Separation was achieved on a Carbopak-PA1 anion exchange resin using AS50 Autosampler, GS50 Gradient Pump, and ED50 Electrochemical detector (Dionex, USA). Isocratic elution in deionized water occurred over 35 minutes; next, a linear gradient ramping to 0.5M NaOH for 10 minutes was employed to wash out strongly adsorbing components. From 45 to 60 minutes, the solvent composition was pure water to equilibrate the column for the next injection.  For oligosaccharide quantification, the hydrolysates were subject to secondary acid hydrolysis in 2.5% (w/w) sulphuric acid at 121°C for 60 minutes. Total sugars (oligomers and monomers) in the hydrolysate were quantified using the same HPAELC protocol employed for monomeric sugars. Oligosaccharides in a hydrolysate were calculated as the difference between its total sugars and its previously quantified monosaccharide compostion. Peaks were manually integrated (Appendix Figure 9). 32  Separation of the acetic acid, formic acid, furfural and 5-hydroxymethyl-2-furaldehyde was achieved on an Aminex HPX-87H column (BioRad, USA) at 60°C (Appendix Figure 1b). Dionex AS3500 Inert Variable-Loop Autosampler was used with GP40 Gradient Pump and AD20 Absorbance detector (Dionex, USA). Determination of acetyl bound to dissolved xylan in the liquid phase was conducted using a variation of the first saponification method: to the liquid phase, 2.0M NaOH was added in a 1:10 (v/v) ratio. Subsequent heating, acidification steps were identical as for the solid wood samples described in 2.2.1 The same HPLC apparatus and program were used as for the saponification of solid wood, but the temperature was decreased to 50°C to aid in the separation of interfering peaks (Appendix Figure 9d). The residue was washed with (100 mL) hot distilled water into a pre-weighed, medium-coarseness, scintered-glass crucible (Pyrex). Following drying and gravimetric analysis, wood residue was assessed for acetyl content and acid-soluble sugars. Acetyl bound to wood residue was determined using the first saponification method (see 2.2.1). Acid-soluble sugars and insoluble lignin were quantified following a two-stage acid hydrolysis (Klason). 33   Figure 6. Flowchart for the quantitative analysis of compounds in wood before and after pretreatment.  2.3.5         Calculations and statistical analysis  Molar stoichiometrics were applied to account for mass loss following the hydrolysis of polysaccharides into monosaccharides (0.90 for hexose sugars and 0.88 pentose sugars). Similarly, conversion constants were used for acetic acid to acetyl (0.98); hexose to HMF (0.70), and pentose to furfural (0.64). An extractives content of 2.0%, determined in 2.2.2, was factored in to all pretreatment calculations.   34  Sugar release was calculated using the formulae: (1) 𝑆𝑢𝑔𝑎𝑟  𝑟𝑒𝑙𝑒𝑎𝑠𝑒 =    ™???    ™    ™????????✤    ™????   ( ™ )? ™?    ™?    ™??????   ( ™ ) ×100% (2) 𝐴𝑑𝑗𝑢𝑠𝑡𝑒𝑑  𝑠𝑢𝑔𝑎𝑟  𝑟𝑒𝑙𝑒𝑎𝑠𝑒 =   ™???    ™    ™????????✤    ™????   ( ™ )™???    ™    ™?    ™??????   ( ™ ? ) × ?™??    ™?    ™??????   (?) ×100%            Wood quality, wood chemistry, and growth traits were correlated by computing Pearson product-moment correlation coefficients and their associated probabilities using SAS statistical package (SAS Institute, 2012). The Pearson’s correlation is represented by ɳ?∙? =  ?   ?𝑀? 𝜎? ,  where X and Y are the two variables to be correlated,   ?   ?𝑀?  is the standard deviation of the mean values of X for each value of Y, and 𝜎?  is the total standard deviation of X (Wright, 1921). The correlation coefficient is between -1 and +1, representing perfect negative and perfect positive correlations, respectively. The CORR procedure derives a p-value under the null hypothesis that the corresponding population correlation is zero. In all cases, the sample size was between 209 and 240. Since the 95% confidence interval is given by the formula 95% CI = ±1.96 (𝑆𝐷/ 𝑛), the large sample size shortens the interval. This increases the power for detecting associations, increasing the probability of rejecting H0 when it is false.  Three technical replicates were performed on each of 19 wood samples. Measured parameters pertain either to untreated (raw) wood, solid wood residue following pretreatment, or monomeric and oligomeric components in the liquid phase following pretreatment. The analysis was performed with two datasets. The first dataset included adjusted values; for example, the proportion of glucose as oligomers relative to the total amount of glucose in the raw wood. The second dataset included the simple concentration of components and is included in the appendix. 35  Pretreatment outcomes were correlated with wood compositional traits using the CORR procedure described above.   36  Chapter 3 Results 3.1  Survey of Growth Trial for Wood Acetylation 3.1.1 Method development: saponification of O-acetyl groups from poplar wood Determination of acetyl groups in poplar wood by saponification required establishing the most suitable NaOH concentration, reaction time, and solids loading. Wood was heated in alkaline solution, acidified, and the resulting compounds separated by ion exchange/reversed phase chromatography. At a concentration of 0.2M NaOH, all acetyl groups were removed from the wood (Figure 7a). The reaction of acetyl groups in wood with 0.2M NaOH at 60°C was complete after 20 minutes (Figure 7b). Sample masses between 5 and 70 mg gave the same acetic acid per unit weight (Figure 7c), and therefore a mass of 30.0 mg was chosen for acetyl quantification.  Quality-control measures were taken throughout the research. Peak separation was adequate, with acetic acid eluting before the internal standard butyric acid (14.9 and 21.4 minutes, respectively; Appendix Figure 1c). Inject-to-inject repeatability on the ASI-100 Autosampler (Dionex, USA) was greater than 99.7% (Appendix Table 1). We tested the stability of acetic acid in the saponifed solution following storage at room temperature, and acetic acid content did not vary for the first 96 hours (Figure 7d). Treatment of wood with alkali did however result in minor losses in hemicellulose and lignin (Appendix Table 2). After establishing these optimized conditions and quality-control measures, the acetyl content of over 200 trees was determined on extractives-free wood. For each specimen, two 37  technical replicates were performed; if they did not agree within 5% of the overall mean for acetyl content, the sample was analyzed a third time (Appendix Figure 2).   Figure 7. Saponification protocol optimization for P. trichocarpa wood meal showing acetyl groups by NaOH concentration (a), time of reaction (b), mass (c) and storage time at room temperature (d). All reactions were carried out at 60°C.  38  3.1.2 Variability of acetylation over time and samplings Since acetyl content of wood decreases with prolonged periods of storage in humid environments (Krilov & Lasander, 1988), we conducted a number of experiments to assess acetyl content before and after the various storage and processing methods employed in this study. First, we compared extracted and unextracted wood and found no significant difference in acetyl content (Appendix Table 3). Next, we determined experimentally that drying wood at 105±5ºC overnight does not affect acetyl content (Appendix Table 4). Finally, we tested acetyl content following two different drying and storage conditions. Green wood stored in the cold room for two years was compared with air-dried wood stored at room temperature for two years. Samples stored in the cold room for one year had on average 0.9% less acetyl groups (w/w) than similar samples dried and stored at room temperature soon after felling (Appendix Table 5). Moreover, logs from the same individuals were taken out of storage two years later and a new cookie tested for acetyl content (Appendix Table 9). In 14 of 19 cases, the acetyl level of wood sampled two years later differed.  3.1.3 Acetylation results Woody material from diverse taxa was also tested for acetyl content (Appendix Figure 3) to benchmark Populus in a collection of wood from species commonly utilized in the wood industry. Acetyl content ranged from near zero in Western Redcedar to over 6% (w/w) in beech. The black cottonwood specimen contained 5.5% (w/w) acetyl groups. Next, the systematic survey of over 200 individuals in the Black Cottonwood Growth Trial in Totem Field was conducted.  39  The average acetyl content of 5-year-old P. trichocarpa wood was 5.2 ± 0.3 % (w/w ± SD). This equals a degree of acetylation of approximately 0.6. The wood specimen with the minimum acetyl content (QBKR16-3/TP-17-22) had 3.6% (w/w) acetyl groups, and the wood sample with the greatest acetyl content (BLCG28-1/TO-13-16) had 6.7% (w/w) acetyl groups. The 96 pairs of surveyed biological replicates differed from each other by, on average, 0.3% (w/w) acetyl (Appendix Figure 6).  3.1.4 Acetyl levels by provenance 35 different provenances were represented in the wood samples analyzed for acetyl content. Acetyl content by provenance is shown in Figure 8 and Table 1. Black Creek produced individuals with significantly higher acetyl content than other provenances. The individual with the highest acetyl content overall originated from Black Creek (BLCG28-1/TO-13-16). The individual with the lowest acetyl came from Baker Creek, however, this individual was an outlier relative to other three trees from Baker Creek. Other outliers were seen in the Chuckwalla, Hope North, Kitimat and Yale drainages.  To further examine acetylation by provenance, trees were averaged by broad geographic cluster, previously established by (Geraldes et al., 2014). The three geographic clusters represented trees from central interior BC, southern BC and Oregon (clusters 3, 4 and 5, respectively, in Figure 9; see also Appendix Figure 4). Acetylation did not differ significantly across these geographic clusters5. Two outliers in central interior BC had lower acetyl than the remaining genotypes in this cluster. Cluster 5, from Oregon, contained two outliers with respect to wood acetylation.                                                 5Trees from the northern provenances of the common garden experiment were not sampled in this study. 40  Of approximately 200 trees tested for acetyl content, 19 were selected for the next phase of pretreatment experiments. These individuals were identified as having the highest, lowest, and mid-levels of total cell wall acetylation in the population of 200 trees first screened. Their cell wall composition is discussed in  Figure 8. Acetyl of extractive-free wood (% w/w) by provenance for over 200 trees examined. Top and bottom of the box are the first and third quartiles; band inside the box represents the median, whiskers extend from the minimum to maximum value in each category. Outliers are plotted as open circles. The number of trees per provenance ranges from three to 19 (see Table 1). Large interquartile range of QBK is due to one outliers. 42  Table 1. Average acetyl content ± standard deviation in clones from 34 provenances in Totem Field at UBC. n = number of clones tested. For provenance drainage abbreviation key, see Appendix Table 6. provenance n acetyl (% w/w  ±  SD) BEL 3 5.26 ± 0.20 BLC 3 5.98 ± 0.25 CHK 7 5.03 ± 0.38 CHW 3 5.38 ± 0.34 CMB 4 5.38 ± 0.44 CNY 4 5.36 ± 0.25 DEN 10 5.43 ± 0.19 ELA 4 4.95 ± 0.52 FNY 4 4.95 ± 0.14 GLC 3 5.15 ± 0.25 HAL 6 5.21 ± 0.35 HAR 12 5.09 ± 0.29 HOM 17 5.26 ± 0.38 HOP 8 5.11 ± 0.18 HRS 12 5.11 ± 0.51 JAS 8 5.15 ± 0.39 KLN 19 5.05 ± 0.22 KTM 5 5.22 ± 0.37 LAF 3 4.81 ± 0.30 LIL 6 5.47 ± 0.13 LON 3 5.31 ± 0.03 MCH 7 5.31 ± 0.36 PHL 4 5.35 ± 0.08 QBK 4 5.00 ± 0.98 QFR 3 5.18 ± 0.14 QLK 3 5.13 ± 0.21 SKW 8 5.28 ± 0.24 SLM 7 5.24 ± 0.15 SQM 6 5.39 ± 0.47 STH 4 4.93 ± 0.42 TOB 4 5.36 ± 0.16 VND 4 5.14 ± 0.27 WEL 4 5.18 ± 0.22 YAL 5 5.02 ± 0.19    43   Figure 9. Acetyl of extractive-free wood (% w/w) by geographic cluster for over 200 trees examined. Top and bottom of the box are the first and third quartiles; band inside the box represents the median, whiskers extend from the minimum to maximum value in each category. Outliers are plotted as open circles. Number of trees for provenance clusters 3, 4 and 5 are 20, 23 and 194, respectively. Cluster 3 represents trees from central interior BC; cluster 4, southern BC; cluster 5, Oregon.44  3.1.5 Correlations analysis Table 2 shows correlations between wood quality, wood chemistry and growth traits. Of 276 possible pairs of wood quality, wood chemistry, and growth traits, 102 were significant (α = 0.05).  In the first six columns in Table 2, the wood quality traits density, microfibril angle, fibre length, fibre width, fibre coarseness and cellulose crystallinity are displayed. Density is positively correlated with cellulose crystallinity (R = 0.15) and carbon isotope discrimination (R = 0.16). Microfibril angle weakly positively correlates with acetylation (R = 0.16), rhamnose content (R = 0.15), and basal diameter (R = 0.14). Microfibril angle is weakly negatively correlated with fibre width (R = -0.17) and cellulose crystallinity (R = -0.25). Fibre length and fibre width are weakly positively correlated (R = 0.20). Fibre length is positively associated with growth traits (0.20 < R < 0.28) and fibre width is negatively associated with growth traits (-0.38 < R < -0.20). Fibre length is negatively associated with wood chemistry traits rhamnose and galactose content (R = -0.28 and -0.20, respectively) and positively associated with glucose content (R = 0.14). Fibre width is positively associated with mannose content (R = 0.18). Fibre width is negatively correlated with density (R = -0.20). Fibre coarseness, the weight per unit length of fibre, is positively correlated with growth traits basal diameter, DBH, height, and stem volume (R = 0.19, 0.15, 0.15 and 0.17, respectively). Cellulose crystallinity is negatively associated with acetyl content (R = -0.18), lignin and arabinose content (R = -0.19 and -0.15) and carbon isotope discrimination (R = -0.13). The next nine columns in Table 2 represent correlations between wood chemistry traits and other parameters measured in this study. Acetyl and lignin are positively correlated (R = 0.28). 45  Lignin is negatively associated with galactose and glucose content (R = -0.18 and -0.24, respectively). Arabinose content is moderately associated with other cell wall sugars (0.43 < R < 0.75), and negatively associated with the lignin-to-cellulose and lignin-to-xylose ratios (R = -0.48 and -0.63, respectively). Rhamnose in the cell wall correlates with acetyl, galactose, xylose and mannose (R = 0.19, 0.50, 0.26 and 0.15, respectively). Rhamnose seems to be negatively associated with total and stem biomass (R = -0.14). Galactose is negatively associated with acetyl content (R = -0.21) and positively associated with glucose, xylose and mannose content (R = 0.49, 0.37 and 0.18, respectively). Galactose is inversely proportional to lignin-to-glucose and lignin-to-xylose ratios (R = -0.46 and -0.40, respectively). 46     Dens MFA Fib_L Fib_W Fib_C Crys Lignin Ara Rha Gal Glu Xyl Man Lig	  :	  cel Lig	  :	  xyl Bas_D DBH Height Tot_M Stem_M Stem_V δ13C Col RowAcetyl -­‐0.07 0.16 -­‐0.08 -­‐0.06 -­‐0.04 -­‐0.18 0.28 -­‐0.04 0.19 -­‐0.21 -­‐0.41 0.04 0.02 0.41 0.12 0.02 -­‐0.03 -­‐0.12 -­‐0.03 -­‐0.07 -­‐0.04 0.02 0.06 0.12Density	  (kg/m3) 0.04 0.07 -­‐0.20 -­‐0.03 0.15 -­‐0.11 0.01 -­‐0.06 0.04 0.06 0.03 0.00 -­‐0.13 -­‐0.12 -­‐0.01 -­‐0.06 -­‐0.06 -­‐0.06 -­‐0.08 -­‐0.06 0.16 0.06 0.09MFA	  (°) -­‐0.09 -­‐0.17 0.06 -­‐0.25 0.09 0.01 0.15 -­‐0.07 -­‐0.12 0.06 0.01 0.12 0.00 0.14 0.09 0.02 0.11 0.10 0.11 0.10 0.00 0.01Fibre_Length	  (mm) 0.20 0.09 0.12 -­‐0.07 -­‐0.12 -­‐0.28 -­‐0.20 0.14 -­‐0.02 0.04 -­‐0.13 -­‐0.05 0.20 0.28 0.27 0.27 0.28 0.21 0.02 0.03 -­‐0.01Fibre_Width	  (µm) -­‐0.04 -­‐0.03 0.06 0.00 -­‐0.01 -­‐0.03 0.03 -­‐0.04 0.18 0.03 0.10 -­‐0.38 -­‐0.24 -­‐0.20 -­‐0.25 -­‐0.24 -­‐0.30 0.05 0.05 0.08Fibre_Coarseness	  (mg/m) 0.07 -­‐0.07 -­‐0.11 -­‐0.01 0.01 -­‐0.07 -­‐0.08 -­‐0.05 -­‐0.01 0.00 0.19 0.15 0.15 0.12 0.11 0.17 -­‐0.01 -­‐0.01 0.03Cellulose_Crystallinity -­‐0.19 -­‐0.15 -­‐0.08 -­‐0.02 0.02 -­‐0.10 -­‐0.07 -­‐0.09 0.03 0.08 0.08 0.01 0.08 0.06 0.06 -­‐0.13 -­‐0.02 -­‐0.06Lignin	  (%	  w/w) -­‐0.01 0.08 -­‐0.18 -­‐0.24 0.07 0.06 0.55 0.34 -­‐0.09 -­‐0.11 -­‐0.04 -­‐0.11 -­‐0.07 -­‐0.07 0.12 0.09 -­‐0.08Arabinose	  (%	  w/w) 0.43 0.62 0.61 0.75 0.57 -­‐0.48 -­‐0.63 -­‐0.02 -­‐0.01 0.02 -­‐0.03 0.00 -­‐0.01 -­‐0.01 0.01 0.19Rhamnose	  (%	  w/w) 0.50 0.02 0.26 0.15 0.00 -­‐0.14 -­‐0.09 -­‐0.07 -­‐0.10 -­‐0.14 -­‐0.14 -­‐0.12 0.13 0.00 0.10Galactose	  (%	  w/w) 0.49 0.37 0.18 -­‐0.46 -­‐0.40 -­‐0.06 -­‐0.04 0.02 -­‐0.10 -­‐0.07 -­‐0.06 -­‐0.01 0.07 0.14Glucose	  (%	  w/w) 0.75 0.55 -­‐0.86 -­‐0.80 0.08 0.12 0.19 0.12 0.16 0.11 -­‐0.10 -­‐0.09 -­‐0.02Xylose	  (%	  w/w) 0.72 -­‐0.57 -­‐0.81 0.08 0.09 0.12 0.06 0.09 0.07 -­‐0.11 -­‐0.09 0.07Mannose	  (%	  w/w) -­‐0.38 -­‐0.53 -­‐0.02 0.03 0.07 -­‐0.02 0.04 0.00 -­‐0.12 -­‐0.17 0.05Lignin	  :	  cellulose 0.90 -­‐0.06 -­‐0.11 -­‐0.14 -­‐0.11 -­‐0.13 -­‐0.10 0.08 0.06 0.04Lignin	  :	  xylose -­‐0.09 -­‐0.12 -­‐0.14 -­‐0.10 -­‐0.12 -­‐0.09 0.10 0.06 -­‐0.01Basal_Diameter 0.87 0.75 0.90 0.87 0.95 -­‐0.16 -­‐0.16 -­‐0.19Diameter	  at	  Breast	  Height	  (DBH) 0.82 0.89 0.91 0.85 -­‐0.16 -­‐0.13 -­‐0.21Height	  (m) 0.76 0.84 0.79 -­‐0.16 -­‐0.12 -­‐0.32Total_Biomass	  (kg) 0.96 0.92 -­‐0.16 -­‐0.07 -­‐0.22Stem_Weight	  (kg) 0.91 -­‐0.17 -­‐0.11 -­‐0.32Stem_Volume	  (cm3) -­‐0.20 -­‐0.17 -­‐0.28Carbon	  isotopic	  signature	  (δ13C,	  ‰) 0.18 0.12Wood	  QualityWood	  ChemistryGrowthWood	  Quality Growth SpacialWood	  ChemistryTable 2. Association of wood quality and biomass traits as determined by the Pearson correlation coefficient. Bolded values indicate significance at the 95% confidence level. Sample size was between 209 and 240 individuals. Acetyl is in percent extractives-free oven-dry weight. All data except acetyl content courtesy of PopCan. 47  3.2 Chemical Transformations of P. trichocarpa Wood During Pretreatment 3.2.1 Reaction bombs evaluated A protocol was established whereby small amounts of poplar wood could be pretreated, and the products monitored in a quantitative manner in HPLC column reaction bombs. There were five bombs, allowing for the pretreatment of five samples in a batch. The repeatability of the experiment was assessed by pretreating the seventeen wood samples in three different reaction vessels and comparing differences in xylose, glucose and acetyl release across technical replicates (Appendix Table 7). Xylose release had a relative standard deviation of 16%, on average across three technical replicates. For glucose release, the average relative standard deviation was 19% across three technical replicates. Acetyl quantification gave relative standard deviations of 23, 21, 20%, respectively, for dissolved xylooligosaccharides, acetic acid release, and wood residue. Total acetylation across technical replicates had a relative standard deviation of 7.2%. Using this approach, we concluded that the reaction vessels could be used to obtain repeatable results for sugar release following pretreatment.  The time required to reach the target temperature was approximately 80 minutes, at which time pressure in the columns was approximately 200 psi (Appendix Figure 10). Hence, the pretreatment was non-isothermal. Wood flour was added to the reaction vessel first. Dilute acid was added by pipetting along the vessel wall in order to remove residual wood flour.  The wood-dilute acid slurry was mixed by vortexing, and centrifugal force was used to bring contents to one end of the reaction vessel. Three separate columns in three separate batches followed highly similar barometric curves (Appendix Figure 10). 48  The optimized HPLC program efficiently separated dissolved sugars in pretreatment liquor (Appendix Figure 9c & d). A chromatograph of the sugar degradation products, furfural and 5-hydroxymethyl-2-furaldehyde, is shown in Appendix Figure 1b.  To further investigate monomeric sugar degradation at high temperatures and acid concentrations, a standard sugar solution of arabinose, rhamnose, galactose, glucose xylose and mannose in 4% sulphuric acid was heated to 99°C and 180°C for 60 minutes (Appendix Figure 11). The standard showed minimal degradation at 99°C. Approximately half the sugars degraded following pretreatment at 180°C.  3.2.2 Chemical transformations of poplar wood at different severities Acetic acid enhances sugar solubilisation and dehydration Acetic acid was added at concentrations varying from 0% to 9% (94 mg/g) to observe its effect on sugar release and degradation during pretreatment (Figure 10). In the control, 61 mg/g xylose dissolved. At 3% acetic acid, total xylose release was similar, but monomeric xylose doubled relative to the control. Increasing acetic acid to 6%, 82 mg/g xylose dissolved. At 9% acetic acid, xylose in solution was similar to that released following the 6% acetic acid treatment. Furfural, the product of xylose dehydration, increased consistently from 3 mg/g to12 mg/g to 21 mg/g to 38 mg/g at acetic acid concentrations of 0, 3, 6, and 9% (v/v), respectively (Figure 10c).  Glucose release during pretreatment was also enhanced by acetic acid (Figure 10b). Total glucose release increased from 12 mg/g to 18 mg/g to 23 mg/g to 25mg/g at the four increasing acetic acid concentrations. Notably, the proportion of glucose monomers increased up to a 49  concentration of 6% acetic acid, then declined at 9% acetic acid. This is explained by the increase in 5-hydroxymethyl-2-furaldehyde detected at the higher concentration (Figure 10c).  Technical replicates varied substantially following pretreatment in the autoclave, as shown by the large error bars in Figure 10. The autoclave was not the heating method used for the remainder of the samples, but rather a Lindberg / Blue M GO1330A laboratory gravity oven (see 2.3.1).  50   Figure 10. Release of xylose (a) and glucose (b), and their degradation products (c) following pretreatment in acetic acid. Samples were autoclaved at 121°C for 60 minutes, following a 60-minute preincubation at 60°C. Error bars represent standard deviation for three technical replicates. HMF, 5-hydroxymethyl-2-furaldehyde. 51 Higher temperature enhances sugar solubilisation and dehydration Pretreatment severity can also be increased by elevating the reaction temperatures (Figure 11). As temperature is increased from 160 to 200°C, xylose release followed an inverted parabola. At 160°C, 60 mg/g xylose dissolved, while increasing the temperature to 170°C nearly doubled xylose release and also hydrolyzed a small amount of xylose into monomers. At 180°C, monomeric xylose release doubled and oligomeric xylose increased to 112 mg/g. The highest total xylose release was 140 mg/g and occurred at 190°C, with over one-third xylose being in monomeric form. At 200°C, total xylose in solution decreased to 60 mg/g, with two-thirds xylose being in monomeric form.  Correspondingly, furfural production at each of the five temperatures is shown in Figure 11b. Furfural formation is negligible after pretreatment at temperatures less than 180°C. Following pretreatment at 190°C, 17 mg/g furfural is formed, relative to the extractives-free weight of the starting wood. Pretreatment at the highest temperature (200°C) yielded 41 mg/g furfural.  Dissolved glucose increased from 6 mg/g to 23 mg/g as temperature increased from 160°C to 200°C (Figure 11d). Oligomeric glucose peaked at 190°C then decreased slightly at 200°C (14 mg/g to 12 mg/g). At the same time, monomeric glucose increased from 6 mg/g at 190°C to 11 mg/g at 200°C. This decrease in oligomeric xylose and drastic increase in monomeric xylose at 200°C was accompanied by some dehydration of glucose to HMF, amounting to 0.4% of the starting wood mass. At 180°C, 9 mg/g oligomeric and 1 mg/g monomeric glucose were detected. 52  The pH of the pretreatment liquor decreased in a near-linear fashion from 3.8 to 3.2 as temperature increased (Figure 11c). This is explained by the increase in acetic acid (Figure 11b). Formic and levulinic acids were not detected. Pretreatment at 160°C yielded 6 mg/g acetic acid. Acetic acid increased to 8, 12, 29 and 42 mg/g as temperatures increased to 170, 180, 190 and 200°C, respectively.    53   Figure 11. Temperature optimization of pretreatment regime showing xylose oligomer and monomer release (a), acetic acid and sugar degradation products furfural and 5-hydroxymethyl-2-furaldehyde and furfural (b), pH (c), and glucose oligomer and monomer release (d). Samples were pre-incubated at 60°C for 1 hour. Liquid phase was water, 5% solids loading, pretreatment time 60 minutes. Shown are the averages of three replicates. Error bars represent standard deviation. Sugar solubilisation and dehydration at increasing severities To establish optimal pretreatment conditions, a bulk poplar sample was subjected to pretreatment regimes of varying time and acid concentrations. Time of pretreatment was either 10, 30 or 60 minutes, and sulphuric acid concentration was 0, 0.1, 0.3 or 0.6% (w/w). Thus, 12 54  possible severities were tested, listed in Table 3. A colour change in wood residue (a) and pretreatment liquor (b) became more pronounced with increasing severity (Figure 12). After pretreatment at varying times and acid concentrations, the hydrolysate was analyzed for monomeric and oligomeric sugars (Figure 13). Ten minutes was not sufficient to liberate appreciable amounts of xylose or glucose (regimes 1-4, Figure 13). Under these mild pretreatment conditions, between 7 and 12% of the wood dissolved (Table 3). The following wood components were not accounted for, explaining the uncertainties in Table 3: acid-soluble lignin, arabinose, galactose and rhamnose. Thirty minutes’ pretreatment with water (regime 5) released 0.6 % (w/w) oligomeric glucose (Figure 13). Adding 0.1% acid (regime 6) increased oligomeric xylose to 5% (w/w) and oligomeric glucose to 1.1% (w/w). Following pretreatment at severity 7, an appreciable 10% (w/w) monomeric and oligomeric xylose dissolved, accompanied by 1.6% (w/w) monomeric glucose and 2.0% oligomeric glucose (Figure 13). At 0.6% acid, monomeric sugars surpassed oligomeric sugars. Monomeric and oligomeric xylose release was 15 and 8% (w/w), respectively, for regime 8. Concurrently, monomeric glucose release was 2.5% (w/w), and oligomer glucose release was 1.5% (w/w) for regime 8.  The water-only pretreatment for 60 minutes (regime 9) yielded nearly 20% (w/w) oligomeric xylose and 1.5% (w/w) oligomeric glucose (Figure 13). However, under these conditions dissolved xylose or glucose in monomeric form were negligible. Adding 0.1% acid (regime 10) resulted in a near-inversion of the xylose oligomer-to-monomer ratio, whereby oligomeric xylose was 5% and monomeric xylose was 17%. Following regime 10, the glucose oligomer-to-monomer ratio also inverted relative to regime 9, with glucose oligomers at 1.8% 55  (w/w) and monomers 5% (w/w). 0.3% acid for 60 minutes (regime 11) yielded near-zero oligomeric xylose close to zero and monomeric xylose 13% (w/w). Following regime 11, dissolved glucose amounted to 9% (w/w) monomers and 1.9% (w/w) oligomers. The harshest pretreatment was 60 minutes at 0.6% acid (regime 12). This yielded near-zero dissolved xylose, 7.6% (w/w) glucose as monomers, and 1.7% (w/w) glucose as oligomers.  In general, as severity increased, xylose oligomer-to-monomer ratios decreased. When all the xylose dissolved, monomers and oligomer decreased, and furfural increased (Figure 13, Figure 14). Cellulose was only partially dissolved, as glucose release reached a maximum at regime 11, with 9% monomeric glucose and 2% oligomeric glucose. Under conditions 10-12 some degradation of hexose sugars to 5-hydroxymethyl-2-furaldehyde was seen, ranging from 0.5 to 2.5 mg/g (Figure 14). Regime 6 achieved 14.4% (w/w) wood dissolution, while regime 7 dissolved 28.0% of wood (Table 3).  Dilute acid pretreatment of P. trichocarpa wood flour (regime 7) resulted in recovery of all original hemicellulose in the liquid phase. As severity increased to regime 8, the recovery was still high, but most of the sugars were found in the oligomeric form. Oligomeric xylose increased from 7.5 to 10.5 % w/w, and monomeric sugars increased from 9 to 16% w/w. The concentration of monomeric glucose originating from cellulose hydrolysis increased from 1.5 to 2.0% w/w. Oligomeric glucose increased slightly from 1.5 to 1.6% w/w. Higher total sugar release was accompanied by similar oligomer-to-monomer ratios, but with more degradation products. The wood residue was analyzed for its constitutive sugars (Appendix Table 8). Under mild pretreatment conditions, the residue contained lignin, cellulose and hemicellulose. Increasing severity to regimes 11-12, the hemicellulose was virtually removed from the wood residue so that it contained only lignin and cellulose. 56  The degradation of pentose sugars and uronic acids to furfural is shown in Figure 14a. Likewise, the degradation of hexose sugars to 5-hydroxymethyl-2-furaldehyde is shown in Figure 14b. Degradation of sugars is negligible for all acid concentrations at 10 and 30 minutes. Furans are formed after 60 minutes in all acid-catalyzed pretreatments, increasing with higher acid concentration. At 0.1% acid, furfural amounted to 1% of the starting biomass. At 0.3% acid, furfural amounted to about 3% of the starting biomass. At 0.6% acid, furfural amounted to 5% of the starting biomass. 5-hydroxymethyl-2-furaldehyde (HMF) formation amounted to 0.05, 0.1 and 0.25% of the starting biomass for 0.1, 0.3 and 0.6% acid, respectively.  Next, acetyl groups were traced from their initial position on wood xylan, to dissolved xylan, and finally, to free acetic acid (Figure 15, Table 4). For any given pretreatment pretreatment regime, as sulphuric acid concentration increased, so did acetic acid release. Correspondingly, acetyl bound to wood residue decreased. Acetyl bound to dissolved xylooligosaccharides did not follow any apparent trend. Acetylated xylooligosaccharides were not detected for any of the 10-minute pretreatments. At 30 minutes, acetylated xylooligosaccharides were detected in liquor of samples treated with 0.1% and 0.3% acid (regimes 6 and 7, respectively). Regime 7 had approximately equal proportions of bound, dissolved and free acetic acid. In the non-catalyzed 30-minute control and the 0.6% acid-treated sample (regimes 5 and 8, respectively), no acetylated xylooligosaccharides were detected. From the increase in free acetic acid and decrease in acetylated wood residue with increasing severity, it can be inferred that there is a shift in acetyl groups from the wood residue after regime 5 to free acetic acid after regime 8. After 60 minutes, the hot water only control (regime 9) had one third acetyl groups bound to dissolved xylooligosaccharides. Adding 0.1% acid shifted the equilibrium towards free acetic acid, with only 5 mg/g as acetylated xylooligosaccharides and 3 57  mg/g bound to wood residue. At regimes 11 and 12, all acetyl groups were hydrolyzed to acetic acid. The optimal pretreatment condition chosen from these results was 0.3% acid and 30 minutes (regime 7). The optimized pretreatment dissolved two-thirds of the xylose in the wood substrate: approximately 5% (w/w) total glucose dissolved. The xylose oligomer-to-monomer ratio was approximately 15:1. Degradation of sugars into furans was near-zero. Acetyl groups were divided equally between the solid wood, dissolved xylooligosaccharides, and free acetic acid.   58   Figure 12. Solid and liquid phases of wood biomass following pretreatment at 180°C over time and acid gradients. Top, pretreated wood samples in sintered glass crucibles after washing. Bottom, pretreatment liquor following saponification. Conditions of pretreatment are: (a) left to right, 0, 0.1, 0.3 and 0.6% acid; top to bottom, 10, 30 and 60 minutes. (b) left to right, 10 minutes in 0% acid to 60 minutes in 0.6% acid. Samples were pre-incubated at 60°C for 60 minutes.     a b 59  Table 3. Dissolution and degradation of wood at various dilute acid pretreatment regimes at 180°C. Soluble lignin and minor sugars arabinose, rhamnose and galactose are not included in the mass balance. O:M, oligomer-to-monomer ratio. Regime Time (min) H2SO4 (% w/w) Dissolved  Glu & Xyl  (% w/w) Wood residue constituents (% w/w) Mass lost to degradation (% w/w) Xylose O:M Glucose O:M 1 10 0.0 3.3 93 < 5.0 0.62 1.1 2 10 0.1 3.4 91 < 5.0 1.8 1.1 3 10 0.3 3.3 88 < 5.0 7.1 1.0 4 10 0.6 3.3 88 < 5.0 3.2 1.2 5 30 0.0 3.8 88 < 5.0 100 2.2 6 30 0.1 14 84 < 5.0 4.5 2.1 7 30 0.3 28 74 < 5.0 1.1 1.1 8 30 0.6 38 72 < 5.0 0.53 0.5 9 60 0.0 28 72 < 5.0 20 4.8 10 60 0.1 37 70 < 5.0 0.45 0.34 11 60 0.3 34 69 7.4 0.18 0.21 12 60 0.6 21 68 23 0.83 0.29   60   Figure 13. Xylose and glucose in the water-soluble fraction of poplar wood pretreated for different times at varying concentrations of sulphuric acid (H2SO4). Oligomeric xylose (a), oligomeric glucose (b), monomeric xylose (c) and monomeric glucose release (d), expressed as % weight.  Pretreatment conditions were 180°C and 5% solids loading. Samples were pre-incubated at 60°C for 1 hour. Data are the average of three replicates. Error bars represent standard deviation.   61    Figure 14. Sugar degradation products furfural (a) and 5-hydroxymethyl-2-furaldehyde (b) following hydrothermal pretreatment of P. trichocarpa wood at various pretreatment times and sulphuric acid (H2SO4) concentrations, expressed as mg/g untreated wood. Pretreatment conditions were 180°C and 5% solids loading. Samples were pre-incubated at 60°C for 1 hour. Data are the average of three (or sometimes two) replicates. Error bars represent standard deviation.   62     Figure 15. Fate of acetyl groups on poplar wood following five dilute acid pretreatments. Grey line shows total acetyl groups in raw material; blue square marker, O-acetyl groups on wood; blue circle marker, free acetic acid; red triangle marker, O-acetyl groups bound to dissolved xylooligosaccharides. For acid concentration and time of pretreatment for each regime, see Table 3.   0	  35	  70	  3	  	   	  	  	  	  	  6	   	  	  	  	  	  	  	  	  	  7	  	  	  	  	  	  	  	  	  	  10	   	  	  	  12	  Regime	  Acetyl	  groups	  (mg/g)	  63  Table 4. Fate of acetyl groups following 12 regimes of differing severity. Pretreatment conditions are specified inTable 3. XOS, dissolved xylooligosaccharides. WR, wood residue. St.Dev., standard deviation. Acetic acid is the average of three technical replicates. Values for XOS and WR are a single technical replicate. Regime Initial acetyl groups Acetic Acid Acetyl bound to XOS Acetyl bound to WR Acetyl balance  (mg/g) (mg/g) (mg/g) (mg/g) (mg/g) 1 61 0.60 0.0 50 10.4 2 61 1.4 0.0 48 11.6 3 61 1.4 0.0 50 9.6 4 61 1.6 0.0 50 9.4 5 61 0.60 0.0 50 10.4 6 61 3.0 12 41 5.0 7 61 18 15 24 4.0 8 61 37 0.0 5.4 18.6 9 61 4.1 21 50 -14.1 10 61 40 5.5 3.0 12.5 11 61 54 0.0 0.0 7.0 12 61 58 0.0 0.0 3.0    64  3.2.3 Chemical transformation of different genotypes        Material balance of wood from candidate genotypes Table 5 details constituents of nineteen selected genotypes. Acid-soluble sugars (excluding uronic acids), lignin, and acetyl and formyl groups were quantified. Individual 1 (“AJ ID” in Table 5) had the highest lignin content of 27.4% (w/w). The lowest lignin content of 16.9% (w/w) was found in individual 20, accompanied by the lowest acetyl and the highest glucose of 55.2% (w/w). Individual 15 had the lowest glucose content of 41.7% (w/w). Xylose ranged from 15.8% (w/w) in individual 14 to 19.9% (w/w) in individual 12.  Genotypes had been originally selected to span the natural range in acetylation, determined in 2012 to be 3.6 to 6.7% (w/w). However, when the same samples were removed from storage in the cold room and re-tested for acetyl content two years later, acetyl values were different (Appendix Table 9). Therefore, the wood samples used ranged from 5.2 to 6.7% (Table 5).65  Table 5. Chemical composition of the wood of selected individuals in percent extractives-free dry weight. Moisture content is also included. ID, identification number; Glu, glucose; Xyl, xylose; Acet, acetyl groups; Man, mannose; Form, formyl groups; Gal, galactose; Ara, arabinose; Rha, rhamnose; H2O, moisture content. Lignin is the sum of acid-soluble and acid-insoluble lignin components. Values are an average of two technical replicates. Genotype /Location  Internal ID AJ           ID Glu Lignin Xyl Acet Man Form Gal Ara Rha H2O Material Balance BELC18-5/TO-4-29 529 1 51.6 24.7 16.1 6.24 1.83 0.38 0.67 0.46 0.28 9.0 102 WLOW15-7/TO-51-28 302 2 47.1 23.0 18.8 6.15 2.73 0.66 0.56 0.50 0.37 10.3 100 HRSO27-2/TO-37-36 561 4 44.8 21.3 17.9 6.23 3.52 0.57 0.49 0.43 0.30 10.0 95.5 BLCG28-1/TO-27-32 111 5 44.4 21.5 18.4 6.36 3.50 0.42 0.48 0.38 0.34 10.7 95.8 SLMB28-2/TO-54-37 182 6 44.2 21.4 19.7 6.48 2.99 0.62 0.52 0.48 0.35 10.1 96.7 QFRS16-3/TO-20-15 162 7 46.1 21.1 17.8 6.24 2.90 0.21 0.50 0.42 0.33 8.6 95.6 LILB26-3/TO-4-1 221 8 48.1 23.1 18.9 5.87 2.48 0.48 0.37 0.34 0.06 10.0 99.7 NBON29-2/TO-40-11 167 9 45.2 21.4 17.3 6.21 3.57 1.57 0.57 0.36 0.29 8.5 96.5 VNDL27-3/TO-37-20 534 10 45.0 23.2 16.7 6.13 3.64 0.61 0.53 0.50 0.27 7.8 96.6 BLCG28-3/TO-1-34 94 11 43.1 20.9 18.2 6.42 2.95 0.47 0.47 0.47 0.32 8.9 93.3 CARS29-3/TO-44-1 306 12 42.3 23.3 19.9 6.51 3.26 0.94 0.46 0.56 0.35 10.0 97.2 KLNG20-2/TO-54-29 104 13 45.0 22.1 18.6 6.37 3.21 0.36 0.54 0.53 0.36 9.3 97.0 HRSP27-4/TO-8-19 278 14 49.0 21.7 15.8 5.50 2.44 0.66 0.65 0.45 0.34 7.7 96.5 ELAD25-4/TO-2-35 25 15 41.7 22.9 17.5 5.97 3.27 0.75 0.59 0.61 0.40 8.5 93.7 AMER13-1/TO-31-14 736 16 45.5 19.6 18.2 6.44 3.23 0.57 0.54 0.54 0.37 9.5 95.0 QBKR16-3/TO-17-22 658 17 49.6 20.2 16.7 5.33 3.04 0.26 0.61 0.47 0.34 8.2 96.5 SKWD24-4/TO-12-27 196 18 50.6 20.0 17.1 5.68 2.35 1.16 0.60 0.35 0.31 8.3 98.2 BLCG28-1/TO-13-16 689 19 47.1 20.5 19.2 6.67 3.25 0.50 0.53 0.45 0.36 8.2 98.6 STHB21-4/TO-17-6 282 20 55.2 16.9 16.6 5.18 1.91 1.16 0.77 0.49 0.32 7.4 98.5 66        Pretreatment performance by genotype The next portion of the project sought to elucidate the effect of inherent acetylation of poplar wood on pretreatment sugar release. Select individuals were pretreated according to regime 7 (180°C, 0.3% H2SO4 for 30 minutes). Table 6 shows sugar release for each of the 19 genotypes. The total xylose release among genotypes ranged from 63.0 to 148 mg/g, while the total glucose release varied from 6.29 to 21.8 mg/g. Total dissolved sugar averaged 117 mg/g xylose and 14.8 mg/g glucose. There was a 64 mg/g difference in total xylose release between the lowest and highest yielding individuals. Glucose release differed by 11.8 mg/g between genotypes. Standard deviation between technical replicates ranged from 2 to 38%. The genotype STHB21-4 (TO-17-6) yielded the highest xylose (148 mg/g) and glucose (21.8 mg/g). Genotypes VNDL27-3 (TO-37-20) and ELAD25-4 (TO-2-35) gave the lowest total xylose yields (84.0 and 63.0 mg/g, respectively). However, the latter was excluded due to high standard error among technical replicates. The total glucose release from low xylose-yielding individuals VNDL27-3 (TO-37-20) and ELAD25-4 (TO-2-35) were 11.7 and 6.29 mg/g, respectively. Oligomeric xylose and glucose release were strongly correlated (R = 0.80, Table 7), such that individuals with high xylose release also tended to have high glucose release. Dividing sugar release into monomers and oligomers gave further insight into differences between genotypes. Monomeric xylose release ranged from 3 - 11 mg/g. Oligomeric xylose amounted to 60 – 140 mg/g. Monomeric glucose ranged between 0.2 and 1.6 mg/g, and its oligomeric counterpart from 6 – 20 mg/g. From Table 3, the average xylose oligomer-to-monomer ratio was 15, and the average glucose oligomer-to-monomer ratio was 19. Thus, for every xylose molecule dissolved in monomeric form, there are 15 xylose molecules dissolved in oligomeric form. Similarly, for every dissolved monomeric glucose molecule, there are 19 67  oligomeric glucose molecules. One individual, QFRS16-3 gave exceptionally high oligomer-to-monomer ratios for both xylose and glucose (23 and 39, respectively). Acetic acid liberated during pretreatment was 5.6 to 13 mg/g, making the solutions 0.3% to 0.6% (v/v) in acetic acid (Figure 16). The proportion of acetyl groups in free form versus bound to xylooligosaccharides versus bound to WR is similar to that seen for regime 7 in Figure 15. Proportions of acetyl groups varied with genotype. No acetylated xylooligosaccharides were detected in genotype 15. Genotypes 2 and 12 had the highest levels of acetylated xylooligosaccharides. Acetyl on wood residue was highest in genotypes 4, 5, 10 and 15. Acetic acid release was highest in genotypes 8, 16 and 19 (all 13 mg/g) and lowest in genotype 15 (5.6 mg/g).   Figure 18a plots xylose released during pretreatment versus acetic acid concentration. Both xylose oligomers and monomers correlate strongly linearly with acetic acid (R = 0.91 and 0.95, respectively in Appendix Table 10). The oligomeric xylose versus acetic acid curve follows a hyperbolic shape that plateaus at around 140 mg/g. Monomeric glucose correlates strongly linearly with acetic acid (R = 0.91, Figure 18b). Oligomeric glucose and acetic acid are strongly correlated (R = 0.89), following a parabolic shape. Figure 18c shows degradation products HMF and furfural versus acetic acid, where no correlation is apparent.   68  Table 6. Xylose and glucose release and their respective oligomer-to-monomer ratios (O:M) following pretreatment. Wood from 19 P. trichocarpa genotypes was tested. Values are in mg/g ± standard deviation. Data are an average of three technical replicates.    Genotype Location Xylose  Glucose   Total (mg/g) O:M Total (mg/g) O:M STHB21-4 TO-17-6 148 ± 3.50 11.7 21.8 ± 3.45 17.8 LILB26-3 TO-4-1 140 ± 25.8 15.2 18.4 ± 5.73 21.2 AMER13-1 TO-31-14 136 ± 3.64 16.8 16.7 ± 0.45 23.5 CARS29-3 TO-44-1 135 ± 10.8 14.9 15.1 ± 0.87 18.7 SKWD24-2 TO-12-27 134 ± 14.3 15.9 18.2 ± 2.41 21.7 BLCG28-1 TO-13-16 129 ± 10.9 14.3 13.2 ± 1.56 14.6 WLOW15-7 TO-51-28 126 ± 26.1 14.8 15.1 ± 5.82 17.8 KLNG20-2 TO-54-29 120 ± 24.6 17.7 14.5 ± 3.98 23.1 HRSP27-4 TO-8-19 118 ± 8.15 16.3 15.8 ± 3.44 20.5 HRSO27-2 TO-37-36 118 ± 27.6 15.2 15.0 ± 4.39 17.3 BLCG28-1 TO-27-32 115 ± 30.3 13.6 15.8 ± 4.38 14.9 SLMB28-2 TO-54-37 115 ± 6.12 15.5 10.0 ± 0.44 18.4 QBKR16-3 TO-17-22 113 ± 61.0 13.6 16.3 ± 11.1 14.9 QFRS16-3 TO-20-15 113 ± 6.06 22.9 16.7 ± 0.96 38.9 BELC18-5 TO-4-29 108 ± 23.1 14.1 14.6 ± 5.42 14.5 BLCG28-3 TO-1-24 107 ± 14.9 16.4 10.6 ± 1.97 17.3 NBON29-2 TO-40-11 100 ± 38.4 12.5 14.7 ± 4.87 13.0 VNDL27-3 TO-37-20 84.0 ± 10.0 14.9 11.7 ± 1.32 16.2 ELAD25-4 TO-2-35 63.0 ± 53.2 12.8 6.29 ± 5.71 13.4 69    Figure 16. Fate of acetyl groups in nineteen genotypes pretreated at regime 7 as per Table 3. XOS, O-acetyl groups bound to dissolved xylooligosaccharides. WR, O-acetyl groups bound to wood residue. Values are the average of three technical replicates.        Correlations between wood chemistry and compounds formed during           pretreatment Table 7 shows which aspects of wood chemistry are correlated during pretreatment. Of 300 possible combinations, 154 were significantly correlated at the 95% confidence level. Key outcomes of this analysis were that (1) the formation of acetic acid is impeded by lignin endogenous to wood, and (2) that acetic acid is correlated with higher sugar release. Other significant findings are listed below.  From Table 7, monomeric and oligomeric sugars correlated strongly with each other (0.51 < R < 0.96) with the notable exception of galactose. Acetic acid was inversely correlated with undissolved wood (R = -0.63). Specifically, acetic acid was inversely correlated with xylose 0	  3.5	  7	  15	   17	   8	   10	   19	   20	   2	   9	   7	   14	   13	   11	   18	   6	   1	   5	   4	   16	   12	  Ace?c	  acid	  XOS	  WR	  Acetyl	  groups	  (%	  w/w)	  Genotype	  70  in wood residue (R = -0.77). Acetic acid was positively associated with oligomeric glucose, xylose and mannose (R = 0.83, 0.86 and 0.63, respectively). Monomers of these sugars also correlated positively with total acetic acid. Pearson correlation coefficients of acetic acid with rhamnose, galactose, glucose, xylose and mannose are 0.71, 0.67, 0.87, 0.83 and 0.72, respectively. Acetic acid was associated with a significant decrease in the proportion of both oligomeric xylose and glucose dissolved (R = -0.74 and -0.83, respectively). Some factors are inherently related and confound results. For example, glucose and xylose content in wood are highly positively correlated (R = 0.75). Similarly, acetyl content is positively correlated with lignin content (R = 0.28) and negatively correlated with glucose content (R = -0.41). Monomeric constituents of the liquid phase positively and significantly correlated with each other (0.61 < R < 0.94). Some monomeric parameters did not strongly correlate, such as those involving arabinose and formic acid (R < 0.43). Oligomeric sugars correlated positively with each other (0.50 < R < 0.91), with the notable exception of galactose (R < 0.19). Acetyl in wood residue signifies more solid wood residue remaining after pretreatment; that is, a smaller proportion of dissolved wood. Fewer hemicellulosic sugars in the solid phase denotes more in the oligomeric and monomeric phases (R < -0.74). There is a negative relationship between wood residue and oligomers/monomers (Table 7). Levels of furans, furfural and HMF, were negligible. These two compounds are not associated with any of the measured parameters except each other (R = 0.85). Oligomer-to-monomer ratios for xylose and glucose are also shown (Table 3). Generally, glucose and xylose oligomer-to-monomer ratios were correlated (R = 0.59), such that if a genotype had a high glucose oligomer-to-monomer ratio, it was also likely to have a high xylose oligomer-to-monomer ratio. The average oligomer to monomer ratio was 15.2 for xylose, and 18.8 for glucose. The highest xylose oligomer to monomer ratio of 22.9 was obtained from 71  genotype QFRS16-3 (TO-20-15). The lowest xylose oligomer-to-monomer ratio of 11.7 was obtained from genotype STHB21-4 (TO-17-6). Glucose release from these two individuals was the reverse of xylose release. QFRS16-3 (TO-20-15) had the highest glucose oligomer-to-monomer ratio (38.9), while STHB21-4 (TO-17-6) exhibited a lower-than-average glucose oligomer-to-monomer ratio (17.8). The lowest glucose oligomer-to-monomer ratio of 13.0 was obtained from NBON29-2 (TO-40-11). Plotting sugar release versus lignin content, it is apparent that higher lignin is associated with lower sugar release (Figure 17). Both correlations are significant at the 95% confidence level72  Table 7. Association between wood components before and after pretreatment, as determined by Pearson correlation coefficients. Pink represents negative correlations, and blue positive correlations. Bolded values denote significance at a 95% confidence level. Values are an average of three technical replicates for 19 wood samples.  %SM, percent solid material remaining following pretreatment; Glu, glucose; Xyl, xylose; Man, mannose; Acet, acetyl groups attached either to solid wood residue or to dissolved xylooligosaccharides; Ara, arabinose; Rha, rhamose; AA, acetic acid; FA, formic acid; O:M Ratio, oligomer-to-monomer ratio. aDenotes that sugar values have been adjusted to the amount of the corresponding constituent in the starting material. L.I. L.S. F.G. M.C. %SM Glua Xyla Mana Aceta Gala Glua Xyla Mana Aceta Araa Rhaa Gala Glua Xyla Mana AA AAa FA Xyl GluLignin, total 0.59 0.25 -0.27 0.45 -0.42 -0.67 0.51 0.55 0.51 0.44 -0.39 -0.54 -0.06 -0.39 0.05 -0.27 -0.44 -0.53 -0.42 -0.15 -0.41 -0.59 -0.42 0.16 0.37Lignin, insoluble 0.50 -0.13 -0.05 0.18 -0.42 0.40 0.22 0.47 0.39 -0.36 -0.53 -0.25 -0.35 -0.02 -0.16 -0.18 -0.46 -0.45 -0.22 -0.49 -0.54 -0.40 0.17 0.25Lignin, soluble -0.20 0.12 0.42 -0.52 0.10 0.15 0.24 0.16 -0.35 -0.42 -0.25 -0.34 0.14 0.05 -0.08 -0.45 -0.36 -0.21 -0.32 -0.31 -0.28 0.12 0.19Formyl groups -0.26 0.20 0.16 -0.20 -0.23 0.04 0.00 0.16 0.17 -0.02 0.34 0.01 -0.01 0.24 0.13 0.12 -0.03 -0.08 0.02 -0.06 0.06 0.08Moisture Content -0.16 -0.55 -0.01 -0.11 0.15 0.48 0.07 -0.15 -0.06 -0.10 0.20 -0.16 -0.06 -0.21 -0.20 -0.21 0.10 -0.12 -0.19 0.01 0.12%SM -0.28 0.09 -0.46 0.52 0.16 -0.41 -0.58 -0.68 -0.59 -0.20 -0.34 -0.10 -0.60 -0.72 -0.74 -0.63 -0.62 -0.77 0.58 0.48Glua -0.14 -0.17 -0.45 -0.77 0.16 0.43 0.16 0.22 -0.42 0.06 0.02 0.47 0.40 0.29 0.40 0.54 0.54 -0.07 -0.22Xyla 0.43 0.81 -0.17 -0.79 -0.85 -0.44 -0.80 -0.57 -0.69 -0.85 -0.72 -0.74 -0.47 -0.77 -0.77 -0.31 0.72 0.77Mana 0.13 -0.14 -0.36 -0.16 0.48 -0.12 0.16 0.19 -0.35 -0.21 0.10 0.45 -0.16 -0.17 -0.03 -0.18 0.09Aceta 0.15 -0.70 -0.89 -0.66 -0.76 -0.39 -0.71 -0.61 -0.77 -0.84 -0.71 -0.90 -0.91 -0.55 0.71 0.77Gala 0.19 -0.10 -0.01 0.13 0.56 0.06 0.32 -0.20 -0.20 -0.19 -0.10 -0.28 -0.51 -0.12 -0.03Glua 0.80 0.50 0.77 0.51 0.68 0.85 0.90 0.74 0.58 0.83 0.81 0.48 -0.65 -0.74Xyla 0.71 0.91 0.53 0.75 0.73 0.87 0.96 0.76 0.86 0.92 0.55 -0.81 -0.82Mana 0.65 0.61 0.75 0.40 0.55 0.82 0.92 0.63 0.64 0.41 -0.77 -0.56Aceta 0.60 0.69 0.77 0.77 0.86 0.67 0.73 0.76 0.37 -0.79 -0.78Araa 0.75 0.62 0.31 0.53 0.54 0.37 0.31 -0.05 -0.72 -0.54Rhaa 0.77 0.72 0.83 0.86 0.71 0.74 0.43 -0.87 -0.81Gala 0.76 0.67 0.49 0.67 0.70 0.30 -0.70 -0.78Glua 0.86 0.72 0.87 0.94 0.68 -0.69 -0.80Xyla 0.90 0.83 0.91 0.61 -0.86 -0.81Mana 0.72 0.76 0.58 -0.83 -0.71Acetic acid 0.92 0.55 -0.74 -0.83Acetic acida 0.73 -0.68 -0.77Formic acid -0.25 -0.37O:M Xylose Oligomer/Monomer Ratio 0.91Untreated W.Solid Wood Res.Oligomers in LiquidMonomers in LiquidO:M RatiosMonomers in LiquidOligomers in LiquidSolid Wood ResidueUntreated Wood73      Figure 17. Relationship of total lignin to total xylose and glucose dissolved during pretreatment. Both regressions are significant at α = 0.05.   R²	  =	  0.24266	  R²	  =	  0.27979	  5	  15	  25	  40	  90	  140	  16	   18	   20	   22	   24	  Xylose	  Glucose	  Lignin	  content	  (%	  w/w)	  Xylose	  dissolved	  (mg/g)	   Glucose	  dissolved	  (mg/g)	  74   Figure 18. Relationship of acetic acid to xylose (a), glucose (b) and degradation products (c) in pretreatment liquor. Each marker represents one technical replicate of the nineteen genotypes tested.75  3.3 Other Results 3.3.1 NMR assessment Appendix Figure 8 is a 2D 13C–1H-correlated (HSQC) NMR spectrum for total poplar cell walls. The location of the acetyl groups on mannan and xylan on poplar wood are clearly illustrated. Lignin is also represented and shows good peak dispersion. Grey contours corresponding to xylose 2 or 3 (labelled X2 and X3, respectively) show that a considerable fraction of xylose residues are not O-acetylated. Light green contours represent the O-acetylated C2 of mannose. Dark green contours represent O-acetylated C2 and C3 of xylose. The 2-O-Ac-β-D-Manp2 contour is smaller than the 3-O-Ac- β -D-Xylp3 contour, which is smaller than the 2-O-Ac- β -D-Xylp2 contour. This suggests that in total cell walls, 2-O-acetylated mannose is less abundant than 3-O-acetylated xylose, which is less abundant than 2-O-acetylated xylose. 4-O-methyl-α-D-glucuronic acid and α-L-arabinofuranosyl units were also detected. There is a small signal for the α-L-arabinofuranosyl unit.   3.3.2 Expression of putative O-acetyltransferases We also looked at developing xylem expression values for published hemicellulose O-acetyltransferases. Regressing gene expression values against acetyl levels, very low and insignificant Pearson product-moment correlation coefficients were obtained (R < 0.1, data not shown). Expression in the ten most acetylated and ten least-acetylated wood samples were averaged and compared (Appendix Figure 7). Six genes varied in expression relative to one another, however, there was no significant difference between the putative O-acetyltransferase expression patterns in the ten most- and least-acetylated samples. 76  Chapter	  4 :	  Discussion	  4.1 	  Survey of a Black Cottonwood Growth Trial for Acetylation 4.1.1 Method development A small-scale, repeatable method to saponify acetyl groups from poplar wood was developed. Varying sodium hydroxide concentration from 0.01 to 1.0 M showed that 0.2M NaOH was the most dilute solution that resulted in complete saponification of O-acetyl groups (Figure 7a). This concentration was chosen to minimize endwise degradation and alkaline hydrolysis of polysaccharides (Fengel & Wegener 1983). Measuring acetic acid in solution after allowing saponification to proceed for various time periods (Figure 7b) showed that twenty minutes was sufficient for the release of all acetyl groups in the lignocellulosic samples. Finally, it was shown that a reasonable sample mass was between 5 mg and 75 mg of wood (Figure 7c). Following these preliminary experiments, we chose the following reaction conditions: 0.2M NaOH, 30 mg wood, and 75 minutes at 60ºC. 4.1.2 Phenotypic associations During secondary cell wall development, cell expansion is followed by polysaccharide deposition, then lignification and cell death (Kaneda et al., 2010). Cellulose is synthesized in muro, while the hemicelluloses such as xylans are synthesized in short chains in the Golgi and polymerized in muro (Pauly et al., 2013). Lignin phenylpropane subunits are also polymerized in muro. A novel finding of this study is that secondary cell wall acetyl content is significantly correlated with glucose and lignin (R = -0.41 and 0.28, respectively). These findings have implications for secondary cell wall deposition.  77  First, acetyl is negatively correlated with glucose content in secondary cell walls (R = -0.41). The negative correlation between acetylation and glucose could be explained by acetyl’s intrinsic hydrophobicity and steric bulk in the cell wall. The acetyl group is hydrophobic, whereas cellulose is moderately hydrophilic. Repulsive forces between acetyl groups and glucan chains in the apoplast could decrease the rate of cellulose deposition, which would decrease the degree of polymerization of cellulose, and the overall cellulose content in the secondary cell wall (Fujita et al., 2012). Disruption of hydrogen bonding by acetyl groups would also explain the weak, negative correlation between cellulose crystallinity and acetylation (R = -0.18, Table 2).  Acetyl content is positively associated with lignin in the cell wall (R = 0.28). This can be explained by molecular dynamics, which dictate that it is energetically favourable for acetyl groups to interact with lignin via hydrophobic effects. Moreover, lignin phenylpropane subunits are polymerized in muro, following the deposition of acetylated polysaccharides. The hydrophobicity of the apoplast, established largely by acetyl groups, could increase the recruitment of hydrophobic lignin monomers to the secondary cell wall. A recent study modelled the docking of acetylated xylan onto cellulose microfibrils using mass spectrometry, NMR and a molecular dynamics simulation (Busse-­‐‑Wicher et al. 2014). The study concluded that the main chain of xylan interacts with the hydrophilic face of the cellulose microfibril, positioning acetyl groups away from the cellulose microfibril-xylose interaction. Xylan is thus a ‘compatibilizer’ between the hydrophilic surfaces of cellulose and the hydrophobic lignin matrix (Busse-­‐‑Wicher et al. 2014). Based on the molecular dynamics simulation performed by Busse-Wicher et al. (2014), acetyl groups at the periphery of the paracrystalline region surrounding cellulose microfibrils should interact with lignin, strengthening the cell wall. It follows that wood that has 78  been chemically acetylated has improved mechanical strength, indicated by higher moduli of rupture and elasticity (Youngquist et al., 1986).  Lignin appears to affect the proportion of acetyl groups released during pretreatment (R = -0.59, Table 7). The correlations highlight the importance of acetyl groups not only in cell wall mechanics, but also in altering the apoplastic environment to enhance the deposition of polymers with similar polarities. 4.1.3    Spacial and temporal variability in acetylation Acetyl content in green wood is not necessarily stable over time, and varies depending on the location in the tree from which the wood was sampled (Appendix Table 5, Appendix Table 9). Since acetyl groups vary depending on the location within a tree, wood should be sampled at multiple locations within a tree in order to obtain an accurate value for acetyl content (Krilov & Lasander, 1988). The spacial variation of acetylation could be due to knots or tension wood, or the differential emission of acetic acid vapours from green wood (Fujii, 1982; Arni et al., 1965). In contrast, acetyl content was not affected by drying in a 105ºC oven or acetone extraction (Appendix Table 3, Appendix Table 4). This suggests that acetyl groups in wood are stable in hot, dry conditions. 4.2 Chemical Transformations of P. trichocarpa Wood During Pretreatment 4.2.1 Small-scale assay versus industrial-scale pretreatment In this study, a small-scale dilute-acid-pretreatment method was employed and optimized. Pressure (200 psi) and temperature (180°C) in the columns were comparable to those in 79  mainstream dilute acid pretreatment (Appendix Figure 10, Esteghlalian et al., 1997). Precise weighing and pipetting gave results that were accurate and repeatable using a mass as small as 2.5 × 10-1 g (Appendix Table 7). Other studies have pretreated samples weighing between 2.6 × 10-3 g and 5.0 × 105 g, with results deemed repeatable (Esteghlalian et al., 1997; Studer et al., 2011; Trajano et al., 2015; Lindedam et al., 2014).  Physical characteristics of biomass, such as surface area, are a major factor determining repeatability of pretreatment on the small-scale. Repeatability of pretreatment of wheat straw using three independently-engineered, microscale pretreatment systems was tested in one published study (Lindedam et al., 2014). The study concluded that differences in carbohydrate yield between the systems were due to differences in the physical properties of the wheat. Additionally, when softwood chips of varying sizes and moisture contents were pretreated under the same regime, carbohydrate yield changed substantially (Cullis et al., 2004). In this study, poplar wood was ground to 40-mesh before pretreatment to facilitate weighing of 30 mg. There are numerous differences in the physical properties of poplar wood flour compared to those of industrially utilized wood chips which could affect the outcome of pretreatment. For example, in this study the correlation between moisture content and pretreatment yield was not significant (|R| < 0.2, Table 7). However, in studies using wood chips, moisture content played a key role in pretreatment yield (Cullis et al., 2004). 4.2.2 Mode of heat transfer in pretreatment assay This pretreatment assay differs from other dilute acid pretreatments in its mode of heat transfer to the biomass-liquid slurry. Many pretreatments use high-pressure steam as the mode of heat transfer (Grous et al., 1986; Esteghlalian et al., 1997). Due to the high heat capacity of 80  water, heating biomass to 180°C with steam takes only seconds. In steam pretreatment, conduction and convection are the mechanisms of heat transfer employed. Alternately, some dilute acid pretreatments use a sand bath to transfer heat, where conduction is the major mode of heat transfer and heat-up time is in the order of minutes (Trajano et al., 2015).  The heating of reaction vessels in study differed from hot steam in that convection did not contribute substantially to heat transfer. Moreover, conductive heat transfer was largely inefficient, because only a small portion of the reaction vessel was in direct contacted with the oven. Preincubation at 60°C mitigated heat-up time; even so, it took 80 minutes to reach the target temperature (Appendix Figure 10). Since heat transfer occurred along a temperature gradient, this pretreatment was non-isothermal. Residence time at each temperature was not assigned; therefore, determining the severity index of non-isothermal pretreatment would be complicated. Nevertheless, pretreatment was consistent and repeatable in this study (Appendix Table 7). Following time, temperature and acid concentration gradients, we tested sugar release and degradation. Components released following a range of severities elucidate characteristics of biomass deconstruction. For example, hemicellulose is first to dissolve, followed by the degradation of its constituent sugars. Next, a small fraction of cellulose—which likely represents the amorphous region—dissolves. Acid-soluble lignin dissolves in the first stages of pretreatment, leaving the acid-insoluble lignin and crystalline cellulose in the residue. HPLC column ‘bombs’ provide a reliable platform upon which pretreatment recalcitrance of poplar wood can be assessed. 81  4.2.3 Chemical transformations by varying treatment severity Acetic acid enhances sugar release during pretreatment Previous work investigated the effect of 0.1, 1.0 or 3.0 % acetic acid (w/w) on the pretreatment efficiency of Japanese beech wood residues using a hot-compressed water pretreatment system at temperatures ranging from 210 to 290°C (Phaiboonsilpa & Saka, 2011). The authors showed that by adding 1% AcOH, the treatment could be performed at temperatures 10–20°C lower than acid-free conditions, with similar product yields. Similarly, treating southern red oak in 5% acetic acid allowed for the more rapid removal of xylan from wood (Conner & Lorenz, 1986). Moreover, it has been shown that acetic acid inputs of less than 0.9% (w/w) did not change pre-hydrolysis results of Eucalyptus globulus wood compared to the autohydrolysis control (Gütsch et al., 2012). Therefore, we hypothesized that pretreating poplar wood in moderate concentrations of acetic acid would enhance hemicellulose dissolution. For this study, acetic acid concentrations between 0 and 9% (v/v) were chosen. Figure 10 shows that acetic acid increases glucose and xylose release. The incremental increase in furfural in the presence of increasing acetic acid concentrations is due to catalysis of xylose dehydration reactions by acetic acid. This experiment provided a proof-of concept for the genotype-varying experiments discussed in 4.2.4.  Patterns of xylose and glucose release in another study treating mixed hardwood chips with 2 – 10% (v/v) acetic acid generally agreed with the present study (Tunc et al., 2014). Other components were measured, such as lignin removal and the removal of the remaining hemicellulose-derived sugars. 5% (w/w) lignin removal was facilitated at concentrations of 4% (v/v) acetic acid or above. In the study performed by Tunc et al., uronic acids in solution reached 82  a maximum of 2% acetic acid (v/v), with decarboxylation occurring at higher concentrations. Arabinose, galactose and mannose dissolved at 4% (v/v) acetic acid.        Temperature selection We pretreated wood samples in hot compressed liquid water to investigate dissolution over different severity gradients facilitated by temperature. Dilute acid pretreatment is generally carried out between 140 and 200°C (Esteghlalian et al., 1997; Lloyd & Wyman, 2005). On the other hand, autohydrolysis or hot water pretreatment generally occurs between 200 and 230°C (Mosier et al., 2005). We selected temperatures between 160 and 200°C and measured products following hot water pretreatment (Figure 11). 180°C was chosen as the optimal temperature because the sample pretreated at 180°C released the highest proportion of oligomeric xylose with the least degradation. Moreover, oligomeric glucose release at this temperature was second-to-highest. Acetic acid in free, dissolved form equated to roughly one-third total acetyl groups. Next, a mineral acid catalyst was added to enhance the pretreatment at 180°C.        Selection of dilute acid pretreatment regime   A series of dilute acid pretreatments was assayed on poplar wood. Sulphuric acid was the mineral acid catalyst, being the most widely used in dilute acid pretreatment (Kootstra et al., 2009; Yang & Wyman, 2008). The addition of catalyst improved sugar release relative to the autohydrolysis control with fewer degradation products formed. Figure 11b shows degradation without the presence of acid catalyst, while Figure 14 shows degradation with acid catalyst. The total amount of xylose in solution is comparable after pretreatment at 160°C and 200°C (Figure 11a). However, at 200°C there is also an appreciable amount of furfural in solution (Figure 11b). 83  Therefore, on a per unit sugar release basis, dilute acid pretreatment yields fewer degradation products than autohydrolysis. Having determined that addition of sulphuric acid reduces sugar loss to degradation, we investigated whether acetic and sulphuric acids can have synergistic effects in deconstructing the cell wall. Twelve pretreatments using varying concentrations of sulphuric acid were tested (Figure 12, Table 3). The goal was to obtain pretreatments ranging in severity from mild to harsh. Severity index, which is used to measure pretreatment severity6, could not be determined because pretreatment was non-isothermal. Twelve regimes did not represent a continuous increase in severity. Rather, different acid concentrations and times gave distinct pretreatment severities, arbitrarily numbered 1 to 12.  The most optimal pretreatment regime was selected based on the release and degradation of cell wall components. In order of abundance, pretreatment products are oligomeric xylose, monomeric xylose, acetic acid, oligomeric glucose, and monomeric glucose. We also looked at degradation products furfural and 5-hydroxymethyl-2-furaldehyde, which were not present in significant concentrations until regime 10 (Figure 14). In the spectrum of mild to harsh, two pretreatments stood out as having close-to-optimal sugar and acetyl release. These were regimes 7 and 8, representing 0.3 and 0.6% acid, respectively, for 30 minutes. Under both conditions, over 25% of the wood dissolved into solution (Table 3), and sugar degradation in both cases was minimal (Figure 14). There was 50% more monomeric xylose release following regime 8 than 7, and oligomeric xylose or glucose release did not differ substantially between regimes 7 and 8. Acetic acid release was two-fold higher in regime 8 compared to regime 7. Finally, the wood                                                 6𝑅? = 𝑡?? ™?™ . ™ , where R0 is the severity factor, T is temperature in °C, and t is time in minutes (Garrote et al., 1999). 84  residue was markedly darker in colour following regime 8 compared with 7 (Figure 12). Since the goal of this study was to reveal differences in sugar release between different wood samples, we chose milder pretreatment conditions. A previous study found mild pretreatment to be effective in comparing pretreatment yield in leading short rotation coppice willow genotypes (Ray et al., 2012).  The concentrations of xylose and glucose are lower following regimes 10-12 than in regime 9 (Figure 13a). This is due to the loss of pentoses and hexoses in the form of furfural and 5-hydroxymethyl-2-furaldehyde, respectively. The quantity of these furans in the pretreatment liquor following 12 pretreatment regimes is shown in Figure 14. Levels of both furans were near-zero up to regime 9. At regime 10, furfural and HMF increased to 9 mg/g and 0.4 mg/g, respectively. Following regime 12, furfural had increased four-fold. In a similar study, it was shown that formic and levulinic acids evolve concurrently with furfural (Li et al., 2014). In this study, no levulinic acid was detected, and formic acid was not quantified due to co-eluting peaks. We also sought to trace the location of acetyl groups after each pretreatment. To the best of our knowledge, acetic acid is not formed from sources other than O-acetyl groups; neither is it consumed in reactions during pretreatment. After any given pretreatment, acetyl groups from the initial substrate can be found either on the wood, on dissolved xylooligosaccharides, or as free acetic acid (Table 4). A handful of studies have quantified the fate of the acetyl group from wood-bound xylan to dissolved oligosaccharides and, finally, cleavage to acetic acid over temperature and time gradients (Borrega et al., 2011; Kuitunen et al., 2013; Tunc et al., 2014). This experiment is unique because it monitored the fate of the acetyl group varying catalyst concentration. In Figure 15, acetylation of wood residue, dissolved xylose and free acetic acid in solution are shown for five of the twelve pretreatment regimes. Low pretreatment severities did 85  not remove acetyl from wood residue. Increasing severity improved acetyl-to-acetic acid conversion. As severity increased, O-acetyl-4-O-methylglucuronoxylan dissolved to acetylated xylooligosaccharides (Figure 13, Figure 15). Xylooligosaccharides were sequentially hydrolyzed to xylose monomers and acetic acid. The data suggest that acetyl groups can also be directly hydrolyzed from O-acetyl-4-O-methylglucuronoxylan on wood, as certain wood residues contained xylose but minimal acetyl groups (Table 4, Appendix Table 8). This same pattern of acetyl release was observed in autohydrolysis experiments on eucalyptus (Leschinsky et al., 2009). Increasing severity enhanced hydrolysis of acetyl groups to acetic acid. Pseudo-lignin can be formed during dilute acid pretreatment when carbohydrate and lignin degradation products combine. Since the measured lignin did not significantly change across treatments (Appendix Figure 12), we conclude that pseudo-lignin was not formed. In a study of beechwood xylan, pseudo-lignin formed at low-severity dilute acid pretreatment was also negligible (Kumar et al., 2013).  4.2.4 Chemical transformation of different genotypes Zero-acetyl negative control Wood samples varied in acetyl content from 3.6% to 6.7% (w/w). It would have been tempting to strip acetyl groups from wood, thereby producing a zero-acetyl negative control. This could provide a picture of carbohydrate dissolution in the absence of acetyl groups. Previously, Douglas-fir pulp was modified to include 6.2% and 0.0% acetyl groups. Saponification of the pulp with 1% NaOH did not affect overall lignin or cell wall carbohydrate content (Pan et al., 2006). Moreover, selectively de-acetylating poplar wood using KOH did not result in lignin or xylan backbone loss (Chang & Holtzapple, 2000). These studies show that, at 86  low concentrations of base, the integrity of the cell wall is not compromised and would provide a suitable zero-acetyl negative control. However, at a sufficient KOH loading to remove all acetyl groups, 14% lignin loss was reported in aspen wood (Kong, Engler et al. 1992). Uronic acid side groups were also removed by treatment with alkali (Chang & Holtzapple, 2000). In the present study, which treated samples with 0.2M NaOH, the structural integrity of the wood was compromised, as 11.3% lignin (w/w) and 14.5% xylan (w/w) were removed (Appendix Table 2). In order to remove acetyl groups without disturbing xylan or lignin, NaOH concentration would have to be again optimized. Given that swelling and hydrolysis occurs in wood under alkaline conditions (Fengel & Wegener 1983), we surmised that the zero-acetyl control would not be representative of the original wood substrate.  Another way to learn about the structural and fuctional consequences of altered cell wall acetyl content is by studying cell wall biosynthesis mutants or ecotypes. For example, a degree of acetylation of glucuronoxylan ranging from 31 to 65% could be found in Arabidopsis glycosyltransferase mutants (Chong et al., 2014).  Natural ecotypes also exhibit a range of cell wall acetylation levels, providing a new platform from which functional consequences of acetylation can be studied. Xyloglucan acetylation varied four-fold among 125 Arabidopsis ecotypes studied, yet no reduction in cell wall strength was reported (Gille et al., 2011).  In this study, there was enough endogenous variation in wood acetyl content of Populus trichocarpa ecotypes to give differential concentrations of acetic acid in the pretreatment liquid (Figure 16). Controlling for as many substrate components as possible, acetyl content was manipulated non-intrusively.  87 Assessing sugar release by genotype: selection of genotypes by proportional versus absolute sugar release In this research we examined sugar release both as proportional to the total in the substrate, and also as an absolute value. The implications for using each are outlined below. The proportional sugar release is useful where individuals with different starting chemistry are compared. For example, it is meaningful to state that an individual with more lignin releases proportionally less xylose than an individual with less lignin (R = -0.54, Table 7). Stating that an individual with more lignin releases a lower quantity of xylose (R = -0.31, Appendix Table 10) is less meaningful because initial xylose content is a confounding factor (VanderWeele & Shpitser, 2013). However, absolute values for sugar release should be considered where the industrial value of a particular feedstock is being assessed. In the latter case, the large variation in sugar release in Appendix Table 10 would reflect differences both in the amounts of cell wall polysaccharides in various genotypes, and in the quantities released during pretreatment. The majority of this section will discuss proportional sugar release. In cases of exceptional results and for an overall view of pretreatment outcomes, it is useful to refer to Table 6 and Appendix Table 10. Lignin impedes the release of acetyl groups as acetic acid It is well known that lignin impedes bioconversion during pretreatment and enzymatic hydrolysis (Studer et al., 2011). This study found that release of both xylose and glucose were inversely and significantly correlated with total lignin (R2 = 0.24 and 0.28, respectively; Figure 17). Next, acetyl groups in the raw material would theoretically result in more acetic acid being liberated during pretreatment. However, in this study, there was no correlation between acetyl 88  content of the wood and acetic acid released during pretreatment (R = -0.08, Table 7). As expected, acetic acid is strongly correlated with sugar dissolution (R > 0.91, Table 7). This suggests that there is a factor limiting the release of acetyl groups to acetic acid.  Another study showed that a substantial proportion of acetyl groups remain attached to aspen wood following dilute acid pretreatment (Grohmann et al., 1989). In this study, approximately one-third of acetyl groups were retained on a given wood residue sample (Figure 15, regime 7). The proportion of acetyl groups retained on the wood residue related not to xylose, but rather lignin content (R = 0.51, Table 7). This suggests that lignin impedes acetic acid formation, resulting in a lower sugar release.  To reduce entropy of the aqueous cell wall environment, acetyl groups interact with lignin hydrophobically, enhancing recalcitrance of highly acetylated xylan to prehydrolysis. Lignin is already known to hinder enzymatic hydrolysis by non-productively binding to enzymes (Chen, 2007; Studer et al., 2011). Lignin degradation products are also generally inhibitory to fermentative microorganisms (Palmqvist, 2000). Results from the present study suggest that lignin also hinders sugar release during pretreatment. It follows that targeted modification of lignin through genetic engineering provides a novel opportunity to increase acetyl-to-acetic acid hydrolysis (Wilkerson et al., 2014), with the effect of improving pretreatment sugar yield. Sugar release by genotype: the numbers This study identified ‘plus’ candidates of P. trichocarpa for bioconversion based on pretreatment sugar release. The genotype STHB21-4 (TO-17-6, AJ ID 20) yielded the highest absolute xylose (148 mg/g) and glucose (21.8 mg/g) following pretreatment, and is a promising candidate for future bioconversion experiments. Other individuals also yielded in the 85th 89  percentile for xylose release (AJ ID 8, 16, 12, 18, 5 and 2, Table 6) and could be tested for silvicultural traits in a breeding program. However, it should be noted that sugar yield following pretreatment does not necessarily correlate with sugar yield following the enzymatic hydrolysis of that same sample (Chen & Dixon, 2007). The genotype with the lowest repeatable xylose release is VNDL27-3 (TO-37-20), and could also be used in the future as an experimental control.  Oligomer-to-monomer ratios were determined for each genotype and are included in Table 6. All candidate genotypes met the required criteria of having an oligomer-to-monomer ratio of greater than ten. A relatively large xylose oligomer-to-monomer ratio was obtained for all genotypes. A higher oligomer-to-monomer ratio is desirable, as more carbohydrate can be recovered in this form (Galbe & Zacchi, 2007; Pu et al., 2013). Acetylation renders xylan fast- or slow-reacting During pretreatment, xylan removal proceeds in two steps, an initial fast phase and a subsequent slow phase. Xylan is composed of two fractions, each of which reacts according to a different kinetic law (Garrote et al., 2001; Shen & Wyman, 2011). Hydrolysis of the first two-thirds of xylan follows first-order reaction kinetics. The remaining third of the xylan is hydrolyzed at a slower rate, following pseudo-first order reaction kinetics (Maloney et al., 1985). Reaction kinetics of these experiments were not assessed per se. However, xylan removal was observed along increasing severity indices, allowing its sequential deconstruction to be monitored. Following regime 7, xylan was divided between wood residue, dissolved xylooligosaccharides and monomeric xylose in a ratio of 1: 0.36: 0.95. Moreover, acetyl groups were divided among wood residue, dissolved xylooligosaccharides and as free acetic acid in a 90  ratio of approximately 1: 0.62: 0.75. With these values we could separate fast- and slow-reacting xylan and calculate respective degrees of acetylation. A disproportionate amount of acetyl groups remained bound to the wood residue compared to that associated with dissolved xylose (Figure 16).  From the literature, fast-reacting xylan is removed first during pretreatment, followed by slow-reacting xylan (Carrasco & Roy, 1992). Slow-reacting xylan was still bound to wood residue after pretreatment, while fast-reacting xylan was dissolved in monomeric form. We used molar ratios to calculate the degree of acetylation of xylose in the solid and dissolved phases following pretreatment. Overall, the degree of acetylation of the starting wood is 0.6, consistent with other studies on poplar (Gröndahl et al., 2003). The xylan that was un-reacted following pretreatment had a degree of acetylation of 0.73. The xylan that had completely hydrolyzed to xylose and acetic acid during pretreatment had a decree of acetylation of 0.35. It follows that slow-reacting xylan has a higher proportion of acetyl groups than fast-reacting xylan.  Regression analysis showed that, in wood samples with higher lignin contents, less acetic acid was hydrolyzed (R = -0.6). This could be facilitated by hydrophobic effects acting at the lignin-acetyl (hemicelluloses) interface.  Results from this study differ from those of Testova et al. (2011), who found that the liquid fraction contained more acetyl groups than the solid fraction of pretreated birch wood. Several experimental parameters could explain the discrepancy in results between these two studies. First, the latter study relied on autohydrolysis rather than a mineral acid catalyst. Second, the study by Testova et al. used wood chips rather than ground wood. Finally, the severity of pretreatment was much higher compared to that in the present study. 91  Many studies, including the present study, have shown that acetic acid catalyzes the breakdown of hemicellulose during pretreatment (Leschinsky et al., 2009; Nabarlatz et al., 2007; Pu et al., 2013; Samuel et al., 2013; Testova et al., 2011; Tunc et al., 2014). However, a direct link between O-acetyl groups and acetic acid hydrolyzed has not been shown. In this study, we propose that the hydrolysis of acetyl groups to acetic acid during pretreatment is indirect and mediated by lignin. This is supported by correlations analyses between pretreatment outcomes of 19 wood samples differing in lignin and acetyl contents. Hydrophobic interactions between O-acetyl groups and lignin explain why less-acetylated xylan is more easily removed from the lignin matrix (Ramsden et al., 1997). Molecular dynamics dictate that hydrophobic effects occur at the interface between water molecules and hemicellulosic acetyl groups and lignin. The present study proposes that whether xylan is fast- or slow-reacting depends on its degree of acetylation. Xylan with a degree of acetylation of 0.73 has a higher propensity to interact with lignin and is therefore slower to react during pretreatment. Conversely, xylan with a degree of acetylation of 0.35 forms fewer hydrophobic interactions with lignin and is more easily removed during pretreatment. Future work should investigate whether lignin plays a role in the extraction of xylooligomers from biomass at neutral or alkaline pH. That autohydrolysis is at least partially mediated by lignin is a novel finding. Moreover, structural aspects of fast- and slow-reacting xylan are elucidated in the present study. These two key findings provide a starting point to study the deconstruction of a wide range of lignocellulosic biomass feedstocks during pretreatment.    92  Chapter 5 : Conclusions & Future Research 5.1 Summary This work presents a simple method of assessing acetyl content and evaluates the effect of endogenous acetyl groups on pretreatment outcome. We found that wood from 200 P. trichocarpa trees averaged 5% (w/w) acetyl groups. Individuals ranged from 3.6% to 6.7% (w/w) in acetyl content. The survey of the black cottonwood growth trial for acetylation indicate that P. trichocarpa trees with a nearly 2-fold variation in acetyl content can be selected with minimal trade-offs in biomass, height, or other growth traits. Acetylation was positively correlated with Klason lignin content (R = 0.28) and negatively correlated with glucose content (R = -0.41). These findings demonstrate that cell wall assembly is intricately linked to acetylation of hemicelluloses, possibly due to hydrophobic effects acting in the aqueous cell wall environment. Acetyl content of green wood seemed to vary with sampling location within the tree, and with prolonged (> 1 year) periods of storage in the cold room. Therefore, a recommendation for future studies on acetylation is to dry wood immediately after felling and sample wood from multiple locations in an individual tree to obtain a representative value for acetyl content.  The second part of this work provides insight into the deconstruction of P. trichocarpa during pretreatment. We monitored xylan removal as pretreatment progressed, noting the transition of the acetyl group from wood to dissolved xylooligosaccharides and, finally, to free acetic acid. This suggests that the catalytic activity of acetic acid can be controlled by altering the various pretreatment parameters. Moreover, since O-acetyl-4-O-methylglucuronoxylan is the 93  major hemicellulose present in the secondary xylem of most dicot species, this knowledge can be applied to other lignocellulosic bio-ethanol feedstocks, such as shrub willow and eucalyptus.  Nineteen different P. trichocarpa genotypes were processed using the same pretreatment regime. The main factors correlating with sugar release were acetic acid and total lignin content. Acetyl content of the initial substrate relates neither to acetic acid formed nor recoverable sugars following pretreatment, showing that acetyl content does not directly affect autohydrolysis. Instead, acetic acid inversely correlates with total lignin content (R = -0.59). This suggests that lignin hinders pretreatment progress by binding to acetyl groups on xylan. Lignin should therefore be selected against among candidate P. trichocarpa individuals, to ensure high pretreatment sugar yield. The relatively mild pretreatment conditions partitioned slow- and fast-reacting xylan preferentially in solid and liquid phases, respectively, establishing that the two hemicelluloses are structurally distinct. Fast-reacting xylan has a degree of acetylation of 0.35, while slow-reacting xylan has a degree of acetylation of 0.73. Understanding structural aspects of xylans released at different stages in pretreatment will lead to their improved recovery in a desirable form. Choosing bioethanol feedstocks that release more acetic acid during pretreatment would contribute to the 42 mg/L (with additives) or 30mg/L (no additives) allowable acetic acid in gasoline according to the Canadian General Standards Board (2011). Moreover, acetic acid in concentrations of up to 100 mmol/l in pretreatment slurries could increase the final bioethanol yield (Larsson et al., 1999). At low levels, acetic acid enhances sugar release, is allowable in gasoline and does not impede downstream steps in bioconversion.  94  5.2 Future Work 5.2.1 Substituent effects: glucuronic acid In xylan, one in eight residues have been proposed to be substituted with either glucuronic or 4-O-methylglucuronic acid. Substituents on sugar rings can have both electrostatic and inductive effects on the adjacent glycosidic bond (Sinnott, 2007). Electrostatic (or field) effects are the dominant substituent effect and arise from the action of a dipole through space (Jensen & Bols, 2006). The slight displacement of electrons in a sigma bond toward the substituent is termed an inductive effect. Electrostatic and inductive effects stabilize the bond adjacent to the substituent, impeding protonation and hydrolysis (Fengel & Wegener, 1983). For example, the β-(1→4)-glycosidic bond in aldobiouronic acid reacts 18 times slower than its unsubstituted counterpart (Whistler & Richards, 1958).In the case of xylan, the substituent glucuronic acid stabilizes the adjacent glycosidic bond. β-(1→4)-bonds of unsubstituted xylose residues are cleaved by acid hydrolyis, while aldobiuronic acid remains intact (Tenkanen et al., 1995). The conventional Klason method for the determination of structural carbohydrates relies on acid hydrolysis.  However, glucuronic acid, in the form of aldobiuronic acid, is not detected (Bertaud, 2002). Previously, it was shown that only one fifth of methylglucuronic acid residues were hydrolyzed following sulphuric acid hydrolysis of pretreated birchwood extract (Saddler et al., 1993). This has important ramifications for xylose analysis, since it can result in a mass loss of up to 3% (w/w xylose).   One way to accurately quantify glucuronyl substituents is to quantify the partially hydrolyzed oligomers, mainly glucuronoxylose and glucuronoxylobiose and their 4-O-methyl esters, following the Klason hydrolysis. This requires a 4-O-methylglucuronoxylose standard, 95  which is not commercially available. As a proof-of-concept, we hydrolyzed aldotriouronic acid and ran it on an organic acids column together with xylose, hydrolyzed birch xylan, and un-hydrolyzed aldotriouronic acid. Future research should optimize the HPLC program required to detect aldotriuronic acid. Alternately, uronic acids can be accurately quantified using acid methanolysis (Sundberg et al., 1996). Once the glucuronic acid substituent can be quantified, its effect during pretreatment should be assessed. One study found that 4% of xylose released from Quercus falcata wood was in the form of xylose-uronic acids dimers, and this represented 40% of total uronic acid groups (Springer, 1985). Another study traced the hydrolysis and subsequent decarboxylation of uronic acids in pretreatment (Willför et al., 2009). If glucuronic acid remains intact, it will act as a stronger acid than acetic acid during pretreatment. The presence of glucuronic acid in the pretreatment liquor would further catalyze hydrolysis of xylan during pretreatment. On the other hand, extensive hydrolysis results in decarboxylation of uronic acid to glucose (Leschinsky et al., 2009). The catalytic behaviour of glucuronic acid during its cleavage from xylan during pretreatment has not been reported in the literature, and would be a valuable starting point for future research.  5.2.2 Localization of O-acetyl groups O-acetyl decorates either the C-2, C-3 or both C-2 and C-3 of xylose residues in xylan. Appendix Figure 8 shows that NMR can be used for semi-quantitative identification of O-acetyl substituents on hemicelluloses. The polysaccharide anomeric region of the spectrum indicates that xylan is the prevalent hemicellulose in poplar wood. The relative size of the dark green contours indicates that xylose is preferentially acetylated on carbon 2 of xylose. The small 96  contour representing 2-O-Ac-β-D-Manp2 indicates that a small amount of mannans are also present. Further experiments using high- and low-acetylated specimens could elucidate whether position, type of hemicellulose, and abundance of acetyl groups are related.  Finally, the HSQC analysis showed that a proportion of carbons 2 and 3 in xylan are not O-acetylated, a structural feature confirmed by Busse-Wicher et al. (2014). Further experiments using high- and low-acetylated specimens could elucidate whether position and abundance of acetyl groups are related. 5.2.3 Genetic basis for O-acetylation This research has clearly established a method whereby total cell wall acetylation can be determined accurately, quickly and cost-effectively. The method can be used to test acetylation in cell walls from diverse plant taxa. Recently, there has been considerable interest in elucidating the genetic machinery responsible for adding O-acetyl groups to hemicellulose in the scientific community (Chong et al., 2014; S. Gille & Pauly, 2012; Manabe et al., 2013; Pawar et al., 2013; Schultink et al., 2015). Gene discoveries have laid the groundwork for determining molecular mechanisms underlying O-acetylation. Given the industrial importance of this cell wall characteristic, as outlined in this research, the genetic basis for acetylation warrants future study. 97  References   Albersheim, P., Darvill, A., Roberts, K., Sederoff, R., & Staehelin, A. (2010). Plant cell walls: From chemistry to biology. Garland Science, Taylor and Francis Group, LLC.  Al-Haddad, J. M., Kang, K. Y., Mansfield, S. D., & Telewski, F. W. (2013). Chemical responses to modified lignin composition in tension wood of hybrid poplar (Populus tremula x Populus alba). Tree Physiology, 33(4), 365-373. doi:10.1093/treephys/tpt017; 10.1093/treephys/tpt017  Arni, P., Cochrane, G., & Gray, J. (1965). The emission of corrosive vapours by wood. I. Survey of the acid-­‐‑release properties of certain freshly felled hardwoods and softwoods. Journal of Applied Chemistry, 15(7), 305-313.  Bals, B., Rogers, C., Jin, M., Balan, V., & Dale, B. (2010). Evaluation of ammonia fibre expansion (AFEX) pretreatment for enzymatic hydrolysis of switchgrass harvested in different seasons and locations. Biotechnol Biofuels, 3(1), 1.  Bendtsen, B., & Senft, J. (1986). Mechanical and anatomical properties in individual growth rings of plantation-grown eastern cottonwood and loblolly pine. Wood and Fiber Science, 18(1), 23-38.  Borrega, M., Nieminen, K., & Sixta, H. (2011). Degradation kinetics of the main carbohydrates in birch wood during hot water extraction in a batch reactor at elevated temperatures. Bioresource Technology, 102(22), 10724-10732.  Bouveng, H. (1961). Phenylisocyanate derivatives of carbohydrates. Acta Chem.Scand, 15(1).  Brownell, H. H., & Saddler, J. N. (1987). Steam pretreatment of lignocellulosic material for enhanced enzymatic hydrolysis. Biotechnology and Bioengineering, 29(2), 228-235.  Browning, B. L. (1967). Methods of wood chemistry. volumes I & II. Methods of Wood Chemistry. John Wiley and Sons: New York. Bura, R., Chandra, R., & Saddler, J. (2009). Influence of xylan on the enzymatic hydrolysis of steam-­‐‑pretreated corn stover and hybrid poplar. Biotechnology Progress, 25(2), 315-322.  Busse-­‐‑Wicher, M., Gomes, T. C., Tryfona, T., Nikolovski, N., Stott, K., Grantham, N. J., . . . Dupree, P. (2014). The pattern of xylan acetylation suggests xylan may interact with cellulose microfibrils as a two-­‐‑fold helical screw in the secondary plant cell wall of Arabidopsis thaliana. The Plant Journal,  Carrasco, F., & Roy, C. (1992). Kinetic study of dilute-acid prehydrolysis of xylan-containing biomass. Wood Science and Technology, 26(3), 189-208.  98  Cave, I. D., and J. C. F. Walker (1994). Stiffness of wood in fast-grown plantation softwoods: the influence of microfibril angle. Forest products journal 44(5): 43. Chang, V. S., & Holtzapple, M. T. (2000). Fundamental factors affecting biomass enzymatic reactivity. Twenty-First Symposium on Biotechnology for Fuels and Chemicals, 5-37.  Chen, F., & Dixon, R. A. (2007). Lignin modification improves fermentable sugar yields for biofuel production. Nature Biotechnology, 25(7), 759-761.  Chen, L., Du, Y., Tian, Z., & Sun, L. (2005). Effect of the degree of deacetylation and the substitution of carboxymethyl chitosan on its aggregation behavior. Journal of Polymer Science Part B: Polymer Physics, 43(3), 296-305.  Chen, X., Shekiro, J., Franden, M. A., Wang, W., Zhang, M., Kuhn, E., Tucker, M. P. (2012). The impacts of deacetylation prior to dilute acid pretreatment on the bioethanol process. Biotechnol Biofuels, 5(8)  Chong, S. L., Virkki, L., Maaheimo, H., Juvonen, M., Derba-Maceluch, Roach, M., Koutaniemi, S., & Tenkanen, M. (2014). O-acetylation of glucuronoxylan in Arabidopsis thaliana wild type and its change in xylan biosynthesis mutants. Glycobiology, 24(6), 494-506. doi:10.1093/glycob/cwu017; 10.1093/glycob/cwu017  Choon, K. K., & Roffael, E. (1990). The Acidity of Five Hardwood Species. Holzforschung, 44, 53-58  Clair, B., Almeras, T., Pilate, G., Jullien, D., Sugiyama, J., & Riekel, C. (2011). Maturation stress generation in poplar tension wood studied by synchrotron radiation microdiffraction. Plant Physiology, 155(1), 562-570. doi:10.1104/pp.110.167270; 10.1104/pp.110.167270  Conner, A. H. (1984). Kinetic modeling of hardwood prehydrolysis. Part I. Xylan removal by water prehydrolysis. Wood and Fiber Science, 16(2), 268-277.  Conner, Anthony H., and Linda F. Lorenz (1986). Kinetic modeling of hardwood prehydrolysis. Part III. Water and dilute acetic acid prehydrolysis of southern red oak. Wood Fiber Sci 18 (2): 248-263. Cullis, I. F., Saddler, J. N., & Mansfield, S. D. (2004). Effect of initial moisture content and chip size on the bioconversion efficiency of softwood lignocellulosics. Biotechnology and Bioengineering, 85(4), 413-421.  Danon, B., Marcotullio, G., & de Jong, W. (2014). Mechanistic and kinetic aspects of pentose dehydration towards furfural in aqueous media employing homogeneous catalysis. Green Chemistry, 16, 39. doi:10.1039/C3GC41351A  99  DeMartini, J. D., & Wyman, C. E. (2011). Changes in composition and sugar release across the annual rings of Populus wood and implications on recalcitrance. Bioresource Technology, 102(2), 1352-1358.  Ebringerova, A., & Heinze, T. (2000). Xylan and xylan derivatives–biopolymers with valuable properties, I. Naturally occurring xylans structures, isolation procedures and properties. Macromolecular Rapid Communications, 21(9), 542-556.  Esteghlalian, A., Hashimoto, A. G., Fenske, J. J., & Penner, M. H. (1997). Modeling and optimization of the dilute-sulphuric-acid pretreatment of corn stover, poplar and switchgrass. Bioresource Technology, 59(2), 129-136.  Fengel, D., & Wegener, G. (1983). Wood: Chemistry, ultrastructure, reactions Walter de Gruyter.  Fujita, M., Lechner, B., Barton, D. A., Overall, R. L., & Wasteneys, G. O. (2012). The missing link: Do cortical microtubules define plasma membrane nanodomains that modulate cellulose biosynthesis? Protoplasma, 249(1), 59-67.  Galbe, M., & Zacchi, G. (2007). Pretreatment of lignocellulosic materials for efficient bioethanol production. Adv Biochem Engin/Biotechnol. 108: 41-65. Garrote, G., Dominguez, H., & Parajo, J. (1999). Hydrothermal processing of lignocellulosic materials. European Journal of Wood and Wood Products, 57(3), 191-202.  Garrote, G., Dominguez, H., & Parajo, J. (2001). Study on the deacetylation of hemicelluloses during the hydrothermal processing of eucalyptus wood. Holz Als Roh-Und Werkstoff, 59(1-2), 53-59.  Geraldes, A., Farzaneh, N., Grassa, C. J., McKown, A. D., Guy, R. D., Mansfield, S. D., & Cronk, Q. C. (2014). Landscape genomics of Populus trichocarpa: the role of hybridization, limited gene flow, and natural selection in shaping patterns of population structure. Evolution, 68(11), 3260-3280.  Gille, S., de Souza, A., Xiong, G., Benz, M., Cheng, K., Schultink, A., & Pauly, M. (2011). O-acetylation of arabidopsis hemicellulose xyloglucan requires AXY4 or AXY4L, proteins with a TBL and DUF231 domain. The Plant Cell Online, 23(11), 4041-4053.  Gille, S., & Pauly, M. (2012). O-acetylation of plant cell wall polysaccharides. Frontiers in Plant Science, 3, 12. doi:10.3389/fpls.2012.00012; 10.3389/fpls.2012.00012  Grohmann, K., Mitchell, D., Himmel, M., Dale, B., & Schroeder, H. (1989). The role of ester groups in resistance of plant cell wall polysaccharides to enzymatic hydrolysis. Applied Biochemistry and Biotechnology, 20(1), 45-61.  100  Gröndahl, M., Teleman, A., & Gatenholm, P. (2003). Effect of acetylation on the material properties of glucuronoxylan from aspen wood. Carbohydrate Polymers, 52(4), 359-366.  Grous, W. R., Converse, A. O., & Grethlein, H. E. (1986). Effect of steam explosion pretreatment on pore size and enzymatic hydrolysis of poplar. Enzyme and Microbial Technology, 8(5), 274-280.  Gütsch, J.S., Nousiainen, T., Sixta, H. (2012) Comparative evaluation of autohydrolysis and acid-catalyzed hydrolysis of Eucalyptus globulus wood. Bioresour. Technol. 109:77–85. Harner, N. K., Wen, X., Bajwa, P. K., Austin, G. D., Ho, C., Habash, M. B., & Lee, H. (2015). Genetic improvement of native xylose-fermenting yeasts for ethanol production. Journal of Industrial Microbiology & Biotechnology, 42(1), 1-20.  Isebrands, J., & Parham, R. (1974). Tension wood anatomy of short-rotation Populus spp. before and after kraft pulping. Wood Sci, 6, 256-265.  Isebrands, J. G., & Richardson, J. (2014). Poplars and willows: Trees for society and the environment. Oxfordshire: CABI inter-governmental development and information organisation.  Ishizawa, C., Davis, M., Schell, D., & Johnson, D. (2007). Porosity and its effect on the digestibility of dilute sulphuric acid pretreated corn stover. Journal of Agricultural and Food Chemistry, 55(7), 2575. doi:DOI: 10.1021/jf062131a  Jensen, H. H., & Bols, M. (2006). Stereoelectronic substituent effects. Accounts of Chemical Research, 39(4), 259-265.  Kadla, J. F., & Kubo, S. (2004). Lignin-based polymer blends: Analysis of intermolecular interactions in lignin–synthetic polymer blends. Composites Part A: Applied Science and Manufacturing, 35(3), 395-400.  Kaneda, M., Rensing, K., & Samuels, L. (2010). Secondary cell wall deposition in developing secondary xylem of poplar. Journal of Integrative Plant Biology, 52(2), 234-243.  Keating, J. D., Panganiban, C., & Mansfield, S. D. (2006). Tolerance and adaptation of ethanologenic yeasts to lignocellulosic inhibitory compounds. Biotechnology and Bioengineering, 93(6), 1196-1206.  Kim, T., Kim, J., Sunwoo, C., & Lee, Y. (2002). Delignification aspect of enzymatic hydrolysis in the ARP process. 24th Symposium on Biotechnology for Fuels and Chemicals,  Kootstra, A. M. J., Beeftink, H. H., Scott, E. L., & Sanders, J. P. (2009). Comparison of dilute mineral and organic acid pretreatment for enzymatic hydrolysis of wheat straw. Biochemical Engineering Journal, 46(2), 126-131.  101  Krilov, A., & Lasander, W. (1988). Acidity of heartwood and sapwood in some eucalypt species. Holzforschung-International Journal of the Biology, Chemistry, Physics and Technology of Wood, 42(4), 253-258.  Kuitunen, S., Vuorinen, T., & Alopaeus, V. (2013). The role of donnan effect in kraft cooking liquor impregnation and hot water extraction of wood. Holzforschung, 67(5), 511-521.  Kumar, R., Hu, F., Sannigrahi, P., Jung, S., Ragauskas, A. J., & Wyman, C. E. (2013). Carbohydrate derived-­‐‑pseudo-­‐‑lignin can retard cellulose biological conversion. Biotechnology and Bioengineering, 110(3), 737-753.  Larsson, S., Palmqvist, E., Hahn-Hägerdal, B., Tengborg, C., Stenberg, K., Zacchi, G., & Nilvebrant, N. (1999a). The generation of fermentation inhibitors during dilute acid hydrolysis of softwood. Enzyme and Microbial Technology, 24(3), 151-159.  Larsson, S., Palmqvist, E., Hahn-Hägerdal, B., Tengborg, C., Stenberg, K., Zacchi, G., & Nilvebrant, N. (1999b). The generation of fermentation inhibitors during dilute acid hydrolysis of softwood. Enzyme and Microbial Technology, 24(3), 151-159.  Leschinsky, M., Sixta, H., & Patt, R. (2009). Detailed mass balances of the autohydrolysis of eucalyptus globulus at 170 C. BioResources, 4(2), 687-703.  Li, H., Deng, A., Ren, J., Liu, C., Lu, Q., Zhong, L., & Sun, R. (2014). Catalytic hydrothermal pretreatment of corncob into xylose and furfural via solid acid catalyst. Bioresource Technology, 158, 313-320.  Lindedam, J., Bruun, S., Jørgensen, H., Decker, S. R., Turner, G. B., DeMartini, J. D., & Felby, C. (2014). Evaluation of high throughput screening methods in picking up differences between cultivars of lignocellulosic biomass for ethanol production. Biomass and Bioenergy, 66, 261-267.  Little, E. L., Rayfield, S., & Buehl, O. (1980). The Audubon Society field guide to North American trees: Western region. New York, Knopf, Chanticleer Press Inc.  Lloyd, T. A., & Wyman, C. E. (2005). Combined sugar yields for dilute sulphuric acid pretreatment of corn stover followed by enzymatic hydrolysis of the remaining solids. Bioresource Technology, 96(18), 1967-1977.  Maloney, M. T., Chapman, T. W., & Baker, A. J. (1985). Dilute acid hydrolysis of paper birch: Kinetics studies of xylan and acetyl-­‐‑group hydrolysis. Biotechnology and Bioengineering, 27(3), 355-361.  Manabe, Y., Nafisi, M., Verhertbruggen, Y., Orfila, C., Gille, S., Rautengarten, C., . . . Pauly, M. (2011). Loss-of-function mutation of REDUCED WALL ACETYLATION2 in Arabidopsis leads to reduced cell wall acetylation and increased resistance to Botrytis cinerea. Plant Physiology, 155(3), 1068-1078.  102  Manabe, Y., Verhertbruggen, Y., Gille, S., Harholt, J., Chong, S. L., Pawar, P. M., . . . Scheller, H. V. (2013). Reduced wall acetylation proteins play vital and distinct roles in cell wall O-acetylation in Arabidopsis. Plant Physiology, 163(3), 1107-1117. doi:10.1104/pp.113.225193; 10.1104/pp.113.225193  Mansfield, S. D., Kang, K., & Chapple, C. (2012). Designed for deconstruction–poplar trees altered in cell wall lignification improve the efficacy of bioethanol production. New Phytologist, 194(1), 91-101.  McKown, A. D., Guy, R. D., Azam, M. S., Drewes, E. C., & Quamme, L. K. (2013). Seasonality and phenology alter functional leaf traits. Oecologia, 172(3), 653-665.  McKown, A. D., Klápště, J., Guy, R. D., Geraldes, A., Porth, I., Hannemann, J., & Ehlting, J. (2014). Genome-­‐‑wide association implicates numerous genes underlying ecological trait variation in natural populations of Populus trichocarpa. New Phytologist, 203(2), 535-553.  McMillan, J. D. (1994). Pretreatment of lignocellulosic biomass. In ACS Symposium Series, USA. Mellerowicz, E. J., Baucher, M., Sundberg, B., & Boerjan, W. (2001). Unravelling cell wall formation in the woody dicot stem. Plant cell walls. New York: Springer. 239-274.  Mizrachi, E., Maloney, V. J., Silberbauer, J., Hefer, C. A., Berger, D. K., Mansfield, S. D., & Myburg, A. A. (2014). Investigating the molecular underpinnings underlying morphology and changes in carbon partitioning during tension wood formation in eucalyptus. New Phytologist, 206(4): 1351-1363. Mosier, N., Wyman, C., Dale, B., Elander, R., Lee, Y., Holtzapple, M., & Ladisch, M. (2005). Features of promising technologies for pretreatment of lignocellulosic biomass. Bioresource Technology, 96(6), 673-686.  Nabarlatz, D., Ebringerová, A., & Montané, D. (2007). Autohydrolysis of agricultural by-products for the production of xylo-oligosaccharides. Carbohydrate Polymers, 69(1), 20-28.  Orfila, C., Dal Degan, F., Jørgensen, B., Scheller, H. V., Ray, P. M., & Ulvskov, P. (2012). Expression of mung bean pectin acetyl esterase in potato tubers: Effect on acetylation of cell wall polymers and tuber mechanical properties. Planta, 236(1), 185-196.  Palmqvist, E., & Hahn-Hägerdal, B. (2000). Fermentation of lignocellulosic hydrolysates. II: Inhibitors and mechanisms of inhibition. Bioresource Technology, 74(1), 25-33.  Pan, X., Arato, C., Gilkes, N., Gregg, D., Mabee, W., Pye, K., & Saddler, J. (2005). Biorefining of softwoods using ethanol organosolv pulping: Preliminary evaluation of process streams for manufacture of fuel-­‐‑grade ethanol and co-­‐‑products. Biotechnology and Bioengineering, 90(4), 473-481.  103  Pan, X., Gilkes, N., Kadla, J., Pye, K., Saka, S., Gregg, D., . . . Saddler, J. (2006). Bioconversion of hybrid poplar to ethanol and co-­‐‑products using an organosolv fractionation process: Optimization of process yields. Biotechnology and Bioengineering, 94(5), 851-861.  Pan, X., Gilkes, N., & Saddler, J. N. (2006). Effect of acetyl groups on enzymatic hydrolysis of cellulosic substrates. Holzforschung, 60(4), 398-401.  Pan, X., Sano, Y. (2005) Fractionation of wheat straw by atmospheric acetic acid process. Bioresour. Technol., 96:1256–1263. Pan, X., Xie, D., Yu, R. W., Lam, D., & Saddler, J. N. (2007). Pretreatment of lodgepole pine killed by mountain pine beetle using the ethanol organosolv process: Fractionation and process optimization. Industrial & Engineering Chemistry Research, 46(8), 2609-2617.  Patil, R., Genco, J. M., & Pendse, H. (2012). Cleavage of acetyl groups from northeast hardwood for acetic acid production. Abstracts of papers of the American Chemical Society, 244.  Pauly, M., Gille, S., Liu, L., Mansoori, N., de Souza, A., Schultink, A., & Xiong, G. (2013). Hemicellulose biosynthesis. Planta, 238(4), 627-642.  Pawar, P. M., Koutaniemi, S., Tenkanen, M., & Mellerowicz, E. J. (2013). Acetylation of woody lignocellulose: Significance and regulation. Frontiers in Plant Science, 4.  Phaiboonsilpa, N., & Saka, S. (2011). Effect of acetic acid addition on chemical conversion of woods as treated by semi-flow hot-compressed water. Holzforschung, 65(5), 667-672.  Pietarinen, S. P., Willför, S. M., Vikström, F. A., & Holmbom, B. R. (2006). Aspen knots, a rich source of flavonoids. Journal of Wood Chemistry and Technology, 26(3), 245-258.  Pogorelko, G., Lionetti, V., Fursova, O., Sundaram, R. M., Qi, M., Whitham, S. A., & Zabotina, O. A. (2013). Arabidopsis and Brachypodium distachyon transgenic plants expressing aspergillus nidulans acetylesterases have decreased degree of polysaccharide acetylation and increased resistance to pathogens. Plant Physiology, 162(1), 9-23. doi:10.1104/pp.113.214460; 10.1104/pp.113.214460  Pojar, J., MacKinnon, A., & Alaback, P. B. (1994). Plants of Coastal British Columbia. Vancouver: Lone Pine Publishing.  Porth, I., Klapšte, J., Skyba, O., Hannemann, J., McKown, A. D., Guy, R. D., & Tuskan, G. A. (2013). Genome-­‐‑wide association mapping for wood characteristics in populus identifies an array of candidate single nucleotide polymorphisms. New Phytologist, 200(3), 710-726.  Porth, I., Klápště, J., Skyba, O., Lai, B. S., Geraldes, A., Muchero, W., & Mansfield, S. D. (2013). Populus trichocarpa cell wall chemistry and ultrastructure trait variation, genetic control and genetic correlations. New Phytologist, 197(3), 777-790.  104  Pu, Y., Hu, F., Huang, F., Davison, B. H., & Ragauskas, A. J. (2013). Assessing the molecular structure basis for biomass recalcitrance during dilute acid and hydrothermal pretreatments. Biotechnol Biofuels, 6(1), 1-13.  Ramsden, M., Blake, F., & Fey, N. (1997). The effect of acetylation on the mechanical properties, hydrophobicity, and dimensional stability of Pinus sylvestris. Wood Science and Technology, 31(2), 97-104.  Ray, M. J., Brereton, N. J., Shield, I., Karp, A., & Murphy, R. J. (2012). Variation in cell wall composition and accessibility in relation to biofuel potential of short rotation coppice willows. BioEnergy Research, 5(3), 685-698.  Rennie, E. A., & Scheller, H. V. (2014). Xylan biosynthesis. Current Opinion in Biotechnology, 26, 100-107.  Runion, G. B., Entry, J. A., Prior, S. A., Mitchell, R. J., & Rogers, H. H. (1999). Tissue chemistry and carbon allocation in seedlings of Pinus palustris subjected to elevated atmospheric CO2 and water stress. Tree Physiology, 19(4:5), 329-335.  Samuel, R., Cao, S., Das, B., Hu, F., Pu, Y., & Ragauskas, A. (2013). Investigation of the fate of poplar lignin during autohydrolysis pretreatment to understand the biomass recalcitrance. RSC Advances, 3(16), 5305-5309. doi:10.1039/c3ra40578h  Sannigrahi, P., Kim, D. H., Jung, S., & Ragauskas, A. (2011). Pseudo-lignin and pretreatment chemistry. Energy & Environmental Science, 4(4), 1306-1310.  Sas Institute (2012). Base sas 9. 3 procedures guide. SAS Institute, Cary, USA.  Scheller, H. V., & Ulvskov, P. (2010). Hemicelluloses. Plant Biology, 61(1), 263.  Schultink, A., Naylor, D., Dama, M., & Pauly, M. (2015). The role of the plant-specific ALTERED XYLOGLUCAN9 protein in Arabidopsis cell wall polysaccharide O-acetylation. Plant Physiology, 167(4), 1271-1283. doi:10.1104/pp.114.256479; 10.1104/pp.114.256479  Shen, J., & Wyman, C. E. (2011). A novel mechanism and kinetic model to explain enhanced xylose yields from dilute sulphuric acid compared to hydrothermal pretreatment of corn stover. Bioresource Technology, 102(19), 9111-9120.  Sheng-zuo, F., & Wen-zhong, Y. (2003). Interclonal and within-tree variation in wood properties of poplar clones. Journal of Forestry Research, 14(4), 263-268.  Sinnott, M. (2007). Carbohydrate chemistry and biochemistry: Structure and mechanism. Royal Society of Chemistry: London, UK.  105  Springer, E. L. (1985). Prehydrolysis of hardwoods with dilute sulphuric acid. Industrial & Engineering Chemistry Product Research and Development, 24(4), 614-623.  Studer, M. H., DeMartini, J. D., Davis, M. F., Sykes, R. W., Davison, B., Keller, M., & Wyman, C. E. (2011). Lignin content in natural Populus variants affects sugar release. Proceedings of the National Academy of Sciences, 108(15), 6300-6305.  Sun, Q., Foston, M., Meng, X., Sawada, D., Pingali, S. V., O'Neill, H. M., . . . Kumar, R. (2014). Effect of lignin content on changes occurring in poplar cellulose ultrastructure during dilute acid pretreatment. Biotechnology for Biofuels, 7(1). doi:10.1186/s13068-014-0150-6; 10.1186/s13068-014-0150-6  Tenkanen, M., Hausalo, T., Siika-Aho, M., Buchert, J., & Viikari, L. (1995). Use of enzymes in combination with anion exchange chromatography in the analysis of carbohydrate composition of kraft pulps. Proc. 8th Int. Symp. Wood and Pulping Chemistry, , 3 189-194.  Testova, L., Chong, S., Tenkanen, M., & Sixta, H. (2011). Autohydrolysis of birch wood. Holzforschung, 65(4), 535-542.  Teymouri, F., Laureano-Perez, L., Alizadeh, H., & Dale, B. E. (2005). Optimization of the ammonia fiber explosion (AFEX) treatment parameters for enzymatic hydrolysis of corn stover. Bioresource Technology, 96(18), 2014-2018.  Trajano, H. L., Pattathil, S., Tomkins, B. A., Tschaplinski, T. J., Hahn, M. G., Van Berkel, G. J., & Wyman, C. E. (2015). Xylan hydrolysis in Populus trichocarpa × P. deltoides and model substrates during hydrothermal pretreatment. Bioresource Technology, 179, 202-210.  Tunc, M. S., Chheda, J., van der Heide, E., Morris, J., & van Heiningen, A. (2014). Pretreatment of hardwood chips via autohydrolysis supported by acetic and formic acid. Holzforschung, 68(4), 401-409.  Tuskan, G. A., Difazio, S., Jansson, S., Bohlmann, J., Grigoriev, I., Hellsten, U., … Salamov, A. (2006). The genome of black cottonwood, Populus trichocarpa (torr. & gray). Science, 313(5793), 1596-1604.  Ucar, M. B., & Ucar, G. (2008). Variation of wood acidity in hard- and softwoods during storage up to one year. Wood Research, 53(4), 105-113.  Utracki, L. A. (2002). Compatibilization of polymer blends. The Canadian Journal of Chemical Engineering, 80(6), 1008-1016.  VanderWeele, T. J., & Shpitser, I. (2013). On the definition of a confounder. Annals of Statistics, 41(1), 196.  106  Vanholme, R., Demedts, B., Morreel, K., Ralph, J., & Boerjan, W. (2010). Lignin biosynthesis and structure. Plant Physiology, 153(3), 895-905. doi:10.1104/pp.110.155119; 10.1104/pp.110.155119  Werpy, T., Petersen, G., Aden, A., Bozell, J., Holladay, J., White, J., and Jones, S. (2004). Top Value Added Chemicals from Biomass.Volume 1-Results of Screening for Potential Candidates from Sugars and Synthesis Gas, No. DOE/GO-102004-1992. Department of Energy, Washington, DC.  Whistler, R. L., & Richards, G. (1958). Uronic acid fragments from slash pine (Pinus elliottii) and their behavior in alkaline Solution. Journal of the American Chemical Society, 80(18), 4888-4891.  Wilkerson, C. G., Mansfield, S. D., Lu, F., Withers, S., Park, J. Y., Karlen, S. D., & Ralph, J. (2014). Monolignol ferulate transferase introduces chemically labile linkages into the lignin backbone. Science, 344(6179), 90-93. doi:10.1126/science.1250161; 10.1126/science.1250161  Willför, S., Pranovich, A., Tamminen, T., Puls, J., Laine, C., Suurnäkki, A., & Hemming, J. (2009). Carbohydrate analysis of plant materials with uronic acid-containing polysaccharides–a comparison between different hydrolysis and subsequent chromatographic analytical techniques. Industrial Crops and Products, 29(2), 571-580.  Wright, S. (1921). Correlation and causation. Journal of Agricultural Research, 20(7), 557-585.  Xie, C., Ying, C. C., Yanchuk, A. D., & Holowachuk, D. L. (2009). Ecotypic mode of regional differentiation caused by restricted gene migration: A case in black cottonwood (Populus trichocarpa) along the Pacific Northwest coast. Canadian Journal of Forest Research, 39(3), 519-525.  Xiong, G., Dama, M., & Pauly, M. (2015). Glucuronic acid moieties on xylan are functionally equivalent to O-acetyl-substituents. Molecular Plant. doi:10.1016/j.molp.2015.02.013; 10.1016/j.molp.2015.02.013  Xiong, G., Cheng, K., & Pauly, M. (2013). Xylan O-acetylation impacts xylem development and enzymatic recalcitrance as indicated by the Arabidopsis mutant tbl29. Molecular Plant, 6(4), 1373. doi:10.1093/mp/sst014  Yang, B., & Wyman, C. E. (2008). Pretreatment: The key to unlocking low-­‐‑cost cellulosic ethanol. Biofuels, Bioproducts and Biorefining, 2(1), 26-40.  Yelle, D. J., Kaparaju, P., Hunt, C. G., Hirth, K., Kim, H., Ralph, J., & Felby, C. (2013). Two-dimensional NMR evidence for cleavage of lignin and xylan substituents in wheat straw through hydrothermal pretreatment and enzymatic hydrolysis. Bioenergy Research, 6(1), 211-221.  107  Yuan, Y., Teng, Q., Zhong, R., & Ye, Z. (2013). The Arabidopsis DUF231 domain-containing protein ESK1 mediates 2-O-and 3-O-acetylation of xylosyl residues in xylan. Plant and Cell Physiology,  Zhang, J., Tang, M., & Viikari, L. (2012). Xylans inhibit enzymatic hydrolysis of lignocellulosic materials by cellulases. Bioresource Technology, 121, 8-12.    108  Appendices109     Appendix Figure 1. Chromatograph of acetic acid, formic acid, 5-hydroxymethyl-2-furaldehyde, and furfural separated on an Aminex HPX-97H polystyrene-divinylbenzene sulfonic acid resin.  a, pretreated wood residue saponified to detect O-acetyl groups as acetic acid; b, liquid phase following pretreatment; c, non-pretreated wood saponified to detect O-acetyl and O-formyl groups as acetic and formic acid (labelled “1”), respectively; (d) liquid phase saponified to detect O-acetyl groups  on dissolved xylooligosaccharides. Blue line shows the sample, with the black line as a reference standard. Butyric acid was the internal standard.  123               110  Appendix Table 1. Inject-to-inject repeatability of Dionex AS50 Autosampler Inject #  standardized acetic acid peak area (mAU*min) Average (mAU*min) standard deviation (mAU*min) 1 5.82 5.82 0.015 2 5.83 3 5.80   111  Appendix Table 2. Effect of saponifying wood meal in 0.1M NaOH on cell wall constituents.  Lignin Xylan Glucan Acetyl  (% w/w) Loss (%) (% w/w) Loss (%) (% w/w) Loss (%)  (% w/w) Loss (%) Control 22.1 11.3 18.6 14.5 45.5 -12.5* 5.2 100 Saponified wood 19.6 15.9 52.0 0.0    *apparent glucan gain due to the removal of other cell wall components    112  Appendix Table 3. Effect of acetone extraction on acetyl content. Two-sample T-test assuming equal variances, under the null hypothesis that acetone extraction does not affect acetyl content.  Extracted Unextracted Mean (% acetyl w/w) 4.8 5.0 Variance 0.098 0.018 Observations 7 8 Pooled Variance 0.055138  Hypothesized Mean Difference 0  df 13  t Stat -1.643  P(T<=t) one-tail 0.062  t Critical one-tail 1.771  P(T<=t) two-tail 0.124  t Critical two-tail 2.160     113  Appendix Table 4. Effect of baking overnight at 105°C for 1 to 5 days on acetyl content of poplar wood. Two different wood samples were tested, one green wood sample which had been stored in the cold room (a), and the other which had been air-dried and stored at room temperature (b). Samples were re-equilibrated to ambient temperature and humidity for 10 hours between each successive baking. RT, room temperature. Times baked (a) Stored green in cold room (b) Air-dried and stored at RT  Acetyl (% w/w) Standard deviation Acetyl (% w/w) Standard deviation 0 3.53 0.11 3.89 0.07 1 3.26  3.77  2 3.41  3.73  3 3.48  3.74  4 3.33  3.70      114  Appendix Table 5. Effect of storage in the cold room (4°C, 60% relative humidity) for two years on acetyl content. Two-sample T-test assuming equal variances, under the null hypothesis that storage in the cold room does not affect acetyl content. RT, room temperature.  Stored in cold room Air-dried and stored at RT Mean (% acetyl w/w) 4.6 5.5 Variance 0.093 0.026 Observations 6 8 Pooled Variance 0.054  Hypothesized Mean Difference 0  df 12  t Stat -7.146  P(T<=t) one-tail 0.00001  t Critical one-tail 1.782  P(T<=t) two-tail 0.00001  t Critical two-tail 2.179    115    Appendix Figure 2. Difference between two technical replicates in acetyl content. A five percent technical replicate repeatability cutoff was chosen, performing a third technical replicate on samples with a difference of greater than 0.26% (w/w) (circled in red). Samples whose technical replicates met the repeatability cutoff were averaged to give acetyl content.   116   Appendix Figure 3. Acetyl content (% w/w) in diverse genera of commercially important wood species and in the plant model species Thale cress.   117   40 45 50 55 60 -140 -130 -120 -110 Latitude Longitude BC Vancouver Is. Haida Gwaii coast 3 4 10 12 13 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Appendix Figure 4. Map of the northwest coast of North America, showing the provenance of trees used in the study. Numbers in the legend correspond to provenances shown in Appendix Table 6. BC, British Columbia. Courtesy of PopCan. 118  Appendix Table 6. Provenance drainages of trees used in the study. Numbers correspond to those in Appendix Figure 4. Several numbered sites encompass trees from more than one drainage.   Provenance Drainage    Provenance Drainage 3 TAK Taku       4 TNZ Tanzilla  26 LIL Lillooet 10 SKN Skeena  27 CHW Chilliwack 12 KTM Kitimat  27 HOP Hope North 15 SHE Shelley  27 HRS Harrison 15 WLO Willow  27 MCM McMillan Is 16 HIX HiXon  27 MTS Matsqui Is 16 KIM Kimball Cr  27 NHT Nahatlatch 16 QBK Baker Cr  27 VND Vender Canal 16 QCT Cottonwood  27 WEL Wellington Bar 16 QFR Fraser  27 YAL Yale 16 QLK Quesnel Lk  28 BLC Black Creek 17 DEN Dean  28 CMB Campbell 18 BEL Bella Coola  28 CNY Courtenay 19 CHK Chuckwalla  28 CSY Cassidy 19 MCH Machmell  28 FNY Fanny Bay 20 KLN Klinaklini  28 LNZ Lantzville 20 MCF U. Kingcome  28 MEM Memekay 21 HOM Homathko  28 SLM Salmon 21 STH Southgate  28 WHT White 22 PHL Phillips  29 CAR Carson 23 TOB Toba  29 LON Longview 24 SKW Skwawka  30 HAL Halsey 25 ELA Elaho  30 JAS Jasper 25 SQM Squamish  30 JEF Jefferson 26 GLC Glacier Lk  30 LAF Lafayette 26 HAR Lillooet  30 NPL North Plains   119   Appendix Figure 5. Biomass accumulation and appearance of tree cookies selected from Populus balsamifera spp. trichocarpa genotypes in the Black Cottonwood Growth Trial at UBC. Panels show least (a,c) and most (b,d) acetylated wood out of nearly 200 individuals examined. Scale bars in (b) represent 0.5 m. In (a), Daniela Anda stands to the left of a small (QBKR16-3) and large (HRSP27-4) genotype.   120   Appendix Figure 6. Frequency histogram showing biological repeatability of acetylation between the 96 clone pairs studied. Each “Number of Clones” unit represents the difference in acetylation between one clone pair.   121    Appendix Figure 7. Expression of six putative O-acetyltransferases in developing xylem of individuals with the most (grey) and the least (black) acetyl groups. FPKM, average fragments per kilobase of coding sequence per million mapped fragments in ten individuals with the most (dark grey) and least (light grey) acetyl groups. Identity of the six genes is as follows: axy4, Potri.008G146100.1; axy4’, Potri.010G095700.1; rwa2, Potri.008G102300.1; rwa2’, Potri.010G148500.1; axy4L, Potri.011G144100.1; esk1, Potri.008G069900. Data mining by C. Hefer. Expression (FPKM) 122   Appendix Figure 8. HSQC 2D NMR spectrum obtained on total cell wall material from poplar wood. Blue and purple contours represent their respective structures to the right of the spectrum.  Dark green contours represent signals from 2- or 3-O-acetylated xylose. Light green contour represents the signal from 2-O-acetylated mannose.  Dark grey contours X2 to X5 represent signals from the respective carbons in xylose. Data courtesy of J. Ralph. 122 123  Appendix Figure 9. Chromatographs of neutral sugars separated on a CarboPac PA1 anion exchange resin. a, total acid-soluble sugars following Klason hydrolysis of untreated wood; b, total acid-soluble sugars following Klason of wood residue following pretreatment; c, dissolved monomeric sugars in the liquid phase following pretreatment; d, total sugars as oligomers and monomers in liquid phase following acid hydrolysis of pretreatment liquor. Fuc, fucose; Ara, arabinose; Rha, rhamnose; Gal, galactose; Glu, glucose; Xyl, xylose; Man, mannose.   123 124  Appendix Table 7. Repeatability of glucose, xylose and acetyl release following dilute acid pretreatment. Data are an average of 17 samples, each with three technical replicates. Table shows average standard deviations of xylose and glucose, in oligomeric and monomeric form. The mean standard deviation for acetyl groups in three forms—on dissolved xylooligosaccharides (XOS), as acetic acid, or bound to xylan on wood residue—and in total are also shown. % RSD, percent relative standard deviation. Samples were preincubated at 60°C for 60 minutes, and pretreated at 190°C for 60 minutes in aqueous 0.3% (w/w) sulphuric acid at a 5% (w/v) solids loading.   Average (mg/g) Standard deviation (mg/g) % RSD Xylose (total) 117 18.8 16 Glucose (total) 15.3 2.9 19 Acetyl 1. on dissolved XOS  17.5 4.0 23 2. acetic acid 11.1 2.3 21 3. on wood residue 24.1 4.8 20 4. total 53.8 3.9 7.2    125     Appendix Figure 10. Non-isothermal pretreatment curve showing pressure versus time in three stainless steel reaction vessels. Oven temperature was set to 180°C. Reaction vessels were pre-incubated at 60°C for 60 minutes prior to pretreatment.   50	  150	  250	  -­‐5	   65	   135	  Column	  1	  Column	  2	  Column	  3	  Pressure (psi) Time (min) 126   Appendix Figure 11. Effect of heat treatment on a sugar standard solution in 4.0% H2SO4. Ara, arabinose; Rha, rhamnose, Gal, galactose; Glu, glucose; Xyl, xylose; Man, mannose.   127    Appendix Figure 12. Insoluble lignin in wood residue following each pretreatment regime.   128  Appendix Table 8. Composition of wood residue in percent extractives-free dry weight following pretreatment regimes. Percent xylan dissolved for each regime is also shown. Samples were preincubated at 60°C for 60 minutes, and pretreated at 190°C for 60 minutes in aqueous 0.3% (w/w) sulphuric acid at a 5% (w/v) solids loading. Regime Time (min) H2SO4  Glucose   Lignin    Xylan      Dissolved Xylan  1 10 0.0 54 20 17 3.0 2 10 0.1 48 20 17 3.0 3 10 0.3 47 19 16 4.0 4 10 0.6 46 20 16 4.0 5 30 0.0 48 19 16 4.0 6 30 0.1 52 19 14 6.0 7 30 0.3 55 22 8.1 12 8 30 0.6 58 24 4.7 15 9 60 0.0 48 20 11 9 10 60 0.1 59 26 2.3 18 11 60 0.3 58 29 0.52 19 12 60 0.6 56 32 0.73 19    129  Appendix Table 9. Acetyl content for identical wood samples analyzed two years apart. Samples analyzed in 2012 were air-dried soon after felling. Samples analyzed in 2014 were from the same log, which had been stored green at 4°C, 60% relative humidity since the 2012 sampling. Individual Acetyl content (% w/w) 2012 sampling 2014 sampling 302 WLOW 15-7/TO-51-28 5.55 6.15 561 HRSO27-2/TO-37-36 5.73 6.23 111 BLCG28-1/TO-27-32 5.74 6.36 182 SLMB28-2/TO-54-37 5.49 6.48 162 QFRS16-3/TO-20-15 5.34 6.24 221 LILB26-3/TO-4-1 5.51 5.87 534 VNDL27-3/TO-37-20 5.22 6.13 094 BLCG28-3/TO-1-34 5.15 6.42 306 CARS29-3/TO-44-1 6.80 6.51 104 KLNG20-2/TOTO-54-29 5.35 6.37 278 HRSP27-4/TO-8-19 4.50 5.50 025 ELAD25-4/TO-2-35 4.41 5.97 736 AMER13-1/TO-31-14 6.50 6.44 658 QBKR16-3/TO-17-22 3.56 5.33 196 SKWD24-4/TO-12-27 4.40 5.68 689 BLCG28-1/TO-13-16 5.40 6.67 282 STHB21-4/TO-17-6 4.37 5.18    130   Appendix Figure 13. Relationship of dissolved wood (% total wood) and acetic acid (‰ total wood) to oligomeric xylose in solution.  R2acetic acid = 0.75; R2dissolved wood = 0.40. The regression for acetic acid is significant at the 95% confidence level. Each marker represents the average value of three technical replicates. 131  Lig_Ins Lig_S Mois Ara Rha Gal Glc Xyl Man Acet Form % Glu Xyl Man Acet Gal Glc	   Xyl Man Acet Ara Rha Gal Glc Xyl Man AA FF HMF FALig_Tot 0.59 0.25 0.45 0.30 0.06 -­‐0.34 -­‐0.48 0.14 0.18 0.52 -­‐0.27 -­‐0.25 -­‐0.62 0.36 0.48 0.40 0.16 -­‐0.35 -­‐0.31 0.03 -­‐0.19 0.21 -­‐0.23 -­‐0.47 -­‐0.42 -­‐0.26 -­‐0.08 -­‐0.24 -­‐0.03 -­‐0.12 -­‐0.26Lig_Ins 0.50 -­‐0.05 0.27 -­‐0.30 -­‐0.22 -­‐0.32 -­‐0.15 0.32 0.23 -­‐0.13 0.17 -­‐0.38 0.23 0.38 0.32 0.16 -­‐0.29 -­‐0.37 -­‐0.07 -­‐0.23 0.17 -­‐0.25 -­‐0.24 -­‐0.34 -­‐0.33 -­‐0.07 -­‐0.29 0.00 -­‐0.10 -­‐0.24Lig_Sol 0.12 -­‐0.22 -­‐0.52 -­‐0.24 -­‐0.14 -­‐0.01 0.21 0.11 -­‐0.20 0.23 -­‐0.36 0.00 0.18 0.16 0.05 -­‐0.19 -­‐0.22 -­‐0.12 -­‐0.24 -­‐0.03 -­‐0.09 -­‐0.11 -­‐0.26 -­‐0.22 -­‐0.12 -­‐0.19 -­‐0.04 -­‐0.05 -­‐0.17Moisture 0.05 0.26 -­‐0.62 -­‐0.52 0.67 0.33 0.55 -­‐0.26 -­‐0.18 -­‐0.54 0.04 0.06 0.21 0.10 -­‐0.05 0.13 0.26 0.03 0.19 0.02 -­‐0.25 -­‐0.19 0.02 0.02 0.06 -­‐0.18 -­‐0.12 -­‐0.12Arabinose 0.59 0.01 -­‐0.30 0.23 0.07 0.09 -­‐0.22 -­‐0.23 -­‐0.04 0.42 0.22 0.18 -­‐0.08 -­‐0.28 -­‐0.12 -­‐0.07 -­‐0.16 0.25 -­‐0.29 -­‐0.32 -­‐0.22 -­‐0.15 -­‐0.16 -­‐0.16 -­‐0.33 -­‐0.26 0.02Rhamnose -­‐0.11 -­‐0.31 0.47 0.12 0.10 -­‐0.20 -­‐0.21 0.06 0.26 0.01 0.05 -­‐0.20 -­‐0.18 0.04 -­‐0.04 -­‐0.18 -­‐0.12 -­‐0.18 -­‐0.35 -­‐0.12 -­‐0.05 -­‐0.16 -­‐0.02 -­‐0.15 -­‐0.18 0.04Galactose 0.85 -­‐0.71 -­‐0.75 -­‐0.77 0.31 -­‐0.02 0.60 -­‐0.21 -­‐0.36 -­‐0.39 0.13 0.29 0.09 -­‐0.13 0.15 0.11 0.21 0.52 0.35 0.24 0.00 0.07 -­‐0.02 0.01 0.29Glucose -­‐0.61 -­‐0.81 -­‐0.69 0.14 -­‐0.04 0.69 -­‐0.50 -­‐0.48 -­‐0.57 0.10 0.48 0.31 0.05 0.33 0.18 0.51 0.67 0.53 0.45 0.23 0.30 0.00 0.03 0.35Xylose 0.46 0.72 -­‐0.10 -­‐0.15 -­‐0.33 0.14 0.13 0.20 -­‐0.15 -­‐0.18 0.17 0.19 0.04 0.02 -­‐0.09 -­‐0.39 -­‐0.23 -­‐0.01 -­‐0.02 0.07 -­‐0.15 -­‐0.07 -­‐0.18Mannose 0.52 -­‐0.06 0.22 -­‐0.46 0.31 0.42 0.39 -­‐0.12 -­‐0.30 -­‐0.24 0.09 -­‐0.29 -­‐0.30 -­‐0.41 -­‐0.50 -­‐0.37 -­‐0.40 -­‐0.04 -­‐0.19 0.05 0.03 -­‐0.18Acetyl -­‐0.22 -­‐0.02 -­‐0.55 0.30 0.40 0.40 -­‐0.03 -­‐0.32 -­‐0.07 0.20 -­‐0.05 0.12 -­‐0.22 -­‐0.54 -­‐0.40 -­‐0.19 0.00 -­‐0.08 -­‐0.04 -­‐0.02 -­‐0.38Formyl 0.10 0.13 -­‐0.11 -­‐0.16 -­‐0.02 0.08 0.11 0.06 -­‐0.11 0.22 -­‐0.16 -­‐0.09 0.27 0.10 0.05 -­‐0.09 -­‐0.05 0.09 0.14 -­‐0.03Wood_Residue_% -­‐0.62 0.67 0.37 0.81 -­‐0.25 -­‐0.80 -­‐0.88 -­‐0.79 -­‐0.61 -­‐0.17 -­‐0.78 -­‐0.66 -­‐0.84 -­‐0.92 -­‐0.83 -­‐0.92 0.07 0.14 -­‐0.78Glucose -­‐0.44 -­‐0.34 -­‐0.71 -­‐0.19 0.57 0.50 0.30 0.33 0.03 0.53 0.57 0.70 0.59 0.49 0.54 -­‐0.03 -­‐0.02 0.59Xylose 0.84 0.82 -­‐0.23 -­‐0.85 -­‐0.80 -­‐0.61 -­‐0.61 -­‐0.36 -­‐0.84 -­‐0.87 -­‐0.78 -­‐0.81 -­‐0.68 -­‐0.78 -­‐0.13 -­‐0.04 -­‐0.52Mannose 0.55 -­‐0.30 -­‐0.60 -­‐0.62 -­‐0.31 -­‐0.40 -­‐0.27 -­‐0.57 -­‐0.69 -­‐0.53 -­‐0.58 -­‐0.31 -­‐0.51 -­‐0.03 -­‐0.04 -­‐0.28Acetyl -­‐0.13 -­‐0.87 -­‐0.84 -­‐0.67 -­‐0.57 -­‐0.27 -­‐0.84 -­‐0.82 -­‐0.87 -­‐0.87 -­‐0.74 -­‐0.87 -­‐0.03 0.01 -­‐0.66Galactose 0.22 0.25 0.29 0.22 0.36 0.11 0.24 -­‐0.02 0.13 0.02 0.13 0.05 0.05 -­‐0.25Glucose 0.86 0.78 0.63 0.37 0.88 0.88 0.94 0.91 0.82 0.91 0.12 0.13 0.70Xylose 0.83 0.67 0.47 0.84 0.74 0.79 0.93 0.77 0.92 -­‐0.06 -­‐0.03 0.53Mannose 0.50 0.49 0.72 0.52 0.66 0.78 0.86 0.87 0.02 0.00 0.53Acetyl 0.47 0.62 0.63 0.58 0.67 0.52 0.60 0.06 0.14 0.19Arabinose 0.48 0.38 0.26 0.44 0.36 0.33 -­‐0.15 -­‐0.06 0.18Rhamnose 0.83 0.87 0.92 0.84 0.89 0.07 0.06 0.67Galactose 0.86 0.81 0.63 0.74 0.09 0.12 0.58Glucose 0.91 0.84 0.89 0.13 0.12 0.80Xylose 0.86 0.95 -­‐0.01 -­‐0.02 0.68Mannose 0.91 0.06 0.00 0.74Acetic_acid 0.05 0.03 0.69Furfural 0.85 0.005-­‐hydroxymethyl-­‐2-­‐furaldehyde -­‐0.01MonomericOligomericUntreated	  WoodWood	  ResidueUntreated	  Wood MonomericOligomericWood	  Residue  Appendix Table 10. Associations between sugars and other components in the solid and liquid phases following pretreatment.Correlations analysis used absolute values for each component, rather than proportions as in Table 7. Bolded values denote significance at a 95% confidence level. Values are an average of three technical replicates for 19 wood samples. 131     


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items