Conventional Methods for Biomass Analysis

Transcription

1 Conventional Methods for Biomass Analysis Daniel Hayes DIBANET Summer School 14/12/10 1

2

3 Important Chemical Characteristics: C6 Sugars: Glucose, Galactose, Mannose C5 Sugars: Arabinose, Xylose Lignin content (acid soluble and insoluble) Extractives Ash. Elemental analysis. With the development of the theoretical and, later, practically validated kinetic equations accurate analytical data should enable the prediction of LvA, furfural, FA yields from a variety of LA and EU feedstocks. The amount of solid residue for downstream processing (pyrolysis) can also be predicted. Hence, determinations as to the value/feasiblity of feedstocks in the DIBANET process chain can be made.

4 Challenges in Processing Lignocellulosics

5 Miscanthus A C4 grass that is native of Asia Various hybrids, typically Miscanthus x giganteus grown Grows quickly, harvested annually. Commercial operations require the planting of Miscanthus rhizomes harvested from standing stock. No real pest problems. Very competitive against weeds. Low nutrient requirement nutrients translocated to rhizomes following senescence. Requires abundant rainfall and mild winters. Irish conditions are appropriate for reasonable yields of up to 20 dry tonnes per hectare.

6

7

8 Over 500 samples of Miscanthus collected: Whole plant Leaves (dead/live) Sheaths (dead/live) Internodes (var heights) Nodes (var heights) Flowers Over 250 other samples collected (e.g. agricultural wastes, energy crops, municipal wastes)

9 Sample Preparation

10 Chemical Analysis Procedure

11 DIBANET Itemised Analysis Procedures UL has developed and written a series of step-bystep analytical procedures for use in the DIBANET project by all partners conducting wet-chemical analysis of biomass feedstocks and NIRS calibration (UL, CTC, UNICAMP). Samples have also been sent from UL to the other partners to cross-check the results of analysis. The target is that methods and data are transferrable and NIRS spectra and reference values can be combined and compared on the same basis.

12 1. Extractives Content Extractives - extraneous components that may be separated from the insoluble cell wall material by their solubility in water or neutral organic solvents. Of relevance to biorefining due to the effects they have on the dynamics of stored biomass piles and the accuracy of analysis procedures. Many different extractives, many of which are speciesspecific. Major categories include monosaccharides, polysaccharides, volatile oils, terpenes, fatty acids and their esters, waxes, polyhydric alcohols, alkaloids and aromatic compounds. 12

13 Extractives cont d. Many extractives have roles in the metabolic processes of a plant. The primary metabolites are inter-convertible biogenic intermediates and include monosaccharides, amino acids, simple fats and various carboxylic acids. The more complex secondary metabolites tend to be irreversibly formed. These include starch, sitosterol, simple terpenoids, chlorophyll, phenylpropanoids, the common flavenoids and simple tannins. Given the transient nature of many extractives, quantities vary greatly depending on the characteristics of the producing tissue and the influence of the environment. Since photosynthesis takes place in the leaves, extractives contents tend to be highest there. 13

14 Extractives Removal and Analysis As the composition of extractives varies between species and within different components of the same species over time, no solvent is equally applicable to all biomass and all biomass components. Different types of neutral solvents are needed for different types of extractives: water-soluble carbohydrates, tannins, and inorganic salts can be liberated from the biomass with hot or cold water. organic solvents such as ethanol, acetone or dichloromethane are needed for the extraction of other extractives, such as resin acids, fat, and terpenes. Removal of all extractives may require sequence of treatments. 14

15 Extractive Removal Employed Quantifying all extractives separately is a costly/lengthy process. Primary concern in our analysis is removal of majority of extractives (but not their separate quantification) and, particularly, those that may interfere with acid hydrolysis and quantification of hemicellulosic/cellulosic sugars. We use the ASTM standard test method for the determination of extractives in biomass feedsotcks. One solvent (95% ethanol) used, reducing time and chemical costs. 15

16 Ethanol-Soluble Extractives Traditional method Soxhlet extraction: Ethanol is heated to reflux and passes through the sample. Extractable components of the sample will dissolve in the solvent. When the Soxhlet chamber is almost full it is emptied automatically by a side arm which allows the solvent to return to the distillation flask. This is termed a cycle. With subsequent cycles the extractives become concentrated in the distillation flask. Under ASTM method, at least 160ml of 95% ethanol is used and the system is heated under reflux for 24 hrs cycles. Rotary evaporator used to remove ethanol and quantify extractives. Extractives also quantified as dry mass loss in solid during process. 16

17 Carbolea Extractives Analysis Use Dionex ASE-200 (Accelerated Solvent Extractor) Operating pressure psi. Extractions automated, take approx 30 minutes (not 24 hours!) Uses less ethanol Up to 24 extractions/samples can be done in a sequence. 17

18 Use of ASE ml cell filled with biomass sample (moisture known). Collection vial placed to collect extract. Conditions: Pressure 1500 psi Temperature 100 o C Initial Heat 5 mins Static Cycle 7 mins Flush Volume 150% # Static Cycles 3 Purge - 120s (N2) Solvent evaporated from collection vial for direct determination of extractives weight. 18

19 Advantages of ASE-200 By using conventional solvents at elevated temperatures and pressures, the efficiency of extraction is increased. The increased temperature accelerates the extraction kinetics while the elevated pressure keeps the solvent below its boiling point. Allows us to extract up to 12 cells a working day. In the last 2 years have analysed 570 samples (490 in duplicate, 80 in triplicate) Would have taken 5+ years using one soxhlet extractor every working day!!! 19

20

21 Hydrolysis of Biomass 21

22 Conventional Gravimetric Methods The use of detergents in feed analysis was introduced by Van Soest (1963a). It is now the most commonly used procedure for the determination of ruminant-feed forage fibre quality. However, there are problems associated with its use for the accurate determination of polysaccharide quantities. The acid detergent fibre (ADF) method is a detergent procedure The sample is heat-treated with 0.5 M sulphuric acid containing cetyltrimethyl-ammonium bromide. The residue that remains is called the ADF residue and is said to be comprised mostly of cellulose and lignin. The lignin content (ADL) of this can be determined by permanganate oxidation with the remaining residue assumed as being cellulose. i.e. cellulose ~ ADF - ADL The neutral detergent fibre (NDF) method involves extraction using a hot solution of sodium lauryl sulphate. This is said to be an approximation for the cell wall content. Hence, ADF NDF = noncellulosic polysaccharides (i.e hemicellulose). 22

23 Detergent Methods: Problems These are Indirect Methods!! Reliability is dependent on the ash and bound protein contents of the feedstock, these are present in the ADF and NDF fractions and could result in an overestimation of cellulose. Hemicellulose content tends to be overestimated. This is due to the variable responses of the polysaccharides of differing structure to the extraction conditions (Wiselogel et al., 1996). E.g. (Theander and Aman, 1980) found that the ADF fraction contained 7-14% hemicellulose and 1-4% crude protein, besides cellulose and lignin, with the cellulose residue also containing 8-13% hemicellulose and 2-7% lignin. The NDF fraction also contained 1 to 6% crude protein, as well as the expected hemicellulose, cellulose and lignin. 23

24 Inaccuracies with Detergent Methods Theander and Westerlund (1993) - while the detergent method was almost exact for the cellulose content of smooth bromegrass, it overestimated cellulose in RCG, Miscanthus and both alfalfa fractions. The correlation factors necessary were 0.84, 0.81, 0.93, 0.89 and 1.02 for alfalfa stem and leaf, RCG, smooth bromegrass and Miscanthus, respectively. The hemicellulose content significantly overestimated. Hemicellulose Cellulose Lignin 24

25 Summary on Detergent Methods Therefore, comparing the detergent-derived polysaccharide values between different species is highly problematic. Indeed, even given a calibration factor for a specific species, polysaccharide determinations may still be inaccurate under certain conditions (e.g. different time of harvest, different environmental conditions etc). For waste feedstocks, detergent methods are likely to be significantly more inaccurate than for many plant species, given the highly heterogeneous character of some wastes (e.g. municipal solid waste and human/animal effluent). In such instances it is likely that more non-carbohydrate and non-lignin components will remain in the residues. In summary, while the detergent methods are reasonably quick and simple, they are not sufficiently accurate or reliable for sugar analysis nor do they provide necessary information on the monosaccharide residues contained in the polysaccharides. 25

26 2. Acid Hydrolysis of Extracted Biomass Follows the Uppsala method, which through a multi-lab and multi-feedstock round robin, was found to be suitable for both woody and herbaceous feedstocks. Involves acid hydrolysis in order to break apart the polysaccharides to their constituent free monosaccharide units which can then be quantified with chromatography. Particle size important for hydrolysis (between 180 and 850 mm). Sample needs to have extractives/starches removed. Take 300mg of extracted sample (moisture content known), add 3ml of 72% H 2 SO 4 and mix, keep for 1 hr and 30 o C mixing thoroughly at intervals. 26

27 2. Acid Hydrolysis cont d. After 1 hour add 84ml water to dilute solution to 4% acid. Seal pressure tube and put in autoclave at 121 o C for 1 hour to complete the hydrolysis and liberation of sugars Autoclaving step associated with loss of free sugars (degradation to HMF, furfural, levulinic acid etc.). So place sugar solutions of known free sugar (glucose, xylose, arabinose etc.) content in autoclave and determine the loss of sugars correction factor for samples. 27

28 Outputs of Hydrolysis Filter contents of pressure tube. Hydrolysate contains: Liberated sugars. Sugar degradation products. Acid soluble lignin. Acid soluble ash. Other components of polysaccharides (uronic acids, acetyl groups). Solid residue contains: Klason Lignin Acid insoluble ash. Weigh residue (Acid Insoluble Residue, AIR) then ash it. Klason lignin = AIR (Ash from AIR) 28

29 Klason Lignin The majority of lignin is retained upon acid hydrolysis of polysaccharides. Non-ligneous elements - such as condensed proteins, suberins and cutins - may be present in the residue. Possible for sugar derivatives to be incorporated into the residue. Some monosaccharides may be degraded by the acid to insoluble materials by the Maillard reaction. Lignin content may be particularly vulnerable to overestimation when analysing weathered feedstocks. Carbohydrate decomposition products are known to undergo browning reactions with amino compounds, forming partially water-insoluble polymeric compounds (humic-type compounds). These would also be analysed 29 as Klason lignins.).

30 Klason Lignin cont d. The results of acid-insoluble lignin analysis are affected by incomplete hydrolysis of biomass. Unless the sample is hydrolyzed completely, the results will be biased high. The acid treatment results in a partial change in the lignin structure and properties (condensation reactions). There are methods that attempt to extract the lignin with minimal effect on the polysaccharides. For example, lignin can be removed from depectinated material by treatment with sodium chlorite-acetic acid at 70 o C for 2 to 4 hours. That which remains should be the holocellulose. However, the procedure is not perfect. The permanganate method involves oxidation of the ADF fraction with potassium permanganate however this 30 method often underestimates total lignin content.

31 Lignin Analysis - Comment Care should therefore be taken in comparing lignin concentrations derived from the different methods, especially between species as the disparities between the methods vary. For the DIBANET project, the most accurate analysis is needed for polysaccharide sugars. Detailed lignin characterisation not necessary. Klason Lignin is considered a suitable indicator for lignin content for oujr needs particularly given that both DIBANET and Uppsala method use acid treatment residue quantities from analysis could be good indicator of those expected from non-polysaccharide components of biomass in DIBANET hydrolysis. 31

32 Analysis of Sugars in Hydrolysate Most interest in (for DIBANET process yields): Glucose - - Arabinose Galactose - - Xylose Mannose - 5 carbon sugars 6 carbon sugars Rhamnose - Fucose (internal standard) - 6 carbon deoxy sugars Can we accurately analyse for all these with one method? 32

33 Analysis of Hydrolysate Chromatography The sample in chromatographic techniques is dissolved in a mobile phase (e.g. gas, liquid). The mobile phase is then forced through an immobile, immiscible stationary phase. Different components have differing solubilities in each phase; a component which is quite soluble in the stationary phase will take longer to travel through it than a component which is not very soluble in the stationary phase but very soluble in the mobile phase. Main chromatographic methods for monosaccharide analysis are gas-liquid chromatography (GLC, GC), highperformance-liquid chromatography (HPLC) and Ion Chromatography. 33

34 Gas Chromatography - GC These use columns, narrow tubes packed with the stationary phase, through which the mobile phase is forced. Through elution the sample is transported by continuous addition of the mobile phase. In GC the absorbing, stationary, medium is a liquid of low-volatility, called the liquid phase. This is dispersed over the surface of an inert solid support. The sample is vaporised and carried through the column by a chemically inert carrier gas (e.g. helium). The components of the mixture are distributed between the gaseous and liquid phases and so travel more slowly than the carrier gas. These phase differences allow separations of different components. 34

35 GC of Carbohydrates After elution from the column a detector converts the concentration or mass of substance contained in the exit gas stream into an equivalent signal, producing the chromatogram. Typically a flame ionisation detector (FID) is used since it responds to carbohydrate related molecules over an extremely wide linear range HOWEVER!!! For GC analysis, volatile derivatives of carbohydrates need to be prepared in order to separate and quantify monosaccharides. Usually the alditol acetate derivatives are prepared. HOH 2 C O HO HO OH OH -D-glucopyranose CHO HC OH HO CH HC OH HC OH H 2 C OH NaBH 4 Reduction CH 2 OH HC OH HO CH HC OH HC OH H 2 C OH AcOAc Acetylation CH 2 OAc HC OAc AcO CH HC OAc HC OAc H 2 C OAc Glucitol This is a time consuming process and allows increased possibilities for laboratory errors. 35

36 HPLC HPLC allows, in some instances, the analysis of monosaccharides after hydrolysis without the necessary volatile derivatisation steps of GC. The mobile phase is liquid and the components of the sample solution migrate according to the non-covalent interactions of the compound with the column and affinities for the mobile phase. Refractive index detector typically used the deflection of a light beam changes according to the composition of the solution flowing through it. Due to limitations of detector and columns, perfect separation and full resolution of all the sugars of interest to DIBANET in one run is not achievable with conventional HPLC. 36

37 Ion Chromatography Separates ions and polar molecules. Analyte molecules from the sample are retained on the column due to coulombic (ionic) interactions. Separation depends on the reversible adsorption of charged solute molecules to immobilised ion exchange groups of opposite charge. IC can be termed HPAEC (anion exchange chromatography) when anions are to be resolved and analysed and HPCEC (cation exchange chromatography) when cations are analysed. The stationary phase have ionic functional groups R-X (X is a cation X + in HPAEC and an anion X - in HPCEC) 37

38 Ion Chromatography (2) The charged groups of the insoluble matrix (i.e. X) are associated with mobile counterions. These can be reversibly exchanged with other ions of the same charge without altering the matrix. The type of charged group determines the type and strength of the ion exchanger and the total number of charged groups and their availability determines the capacity. Tertiary amines are used in HPAEC and sulphonate groups in HPCEC. Mobile phases of constant (isocratic) or varying (gradient) ionic strength are used to push the analytes along the column. 38

39 Columns - Anion R 3 N + OH - Cl - Resin Bead R 3 N + R 3 N + OH - OH - SO 4 2- OH - R 3 N + OH - OH - OH -

40 OH - R 3 N + OHŌH - OH - Resin Bead R 3 N + OH - OH - OH - R 3 N + OH - OH - OH - OH - R 3 N + OH - OH - OH - OH - OH - OH -

41 COLUMN CELL DETECTOR 41

42 IC Analysis of Carbohydrates Carbohydrates are weak acids with pk a s above 11 The use of sodium hydroxide as an eluent promotes ionisation of the carbohydrates to their anionic form IC utilises small differences in the pk a s of the hydroxyl groups of carbohydrates allowing the separation of such similar molecules as galactose, mannose and glucose. 42

43 IC for Carbohydrates (2) The lower the alkalinity of the mobile phase, the higher is the selectivity for separation. For our analysis it is necessary to use only deionised water for the elution of sugars = Fucose (ISTD) 2 = Arabinose 3 = Galactose 4 = Rhamnose 5 = Glucose 6 = Xylose 7 = Mannose 43

44 IC for Carbohydrates (3) A mobile phase of water requires a regeneration step to restore full ion-exchange capacity to the column. Strongly retained molecules (e.g. uronic acids) also need to be pushed from the column. Concentrated NaOH used after elution of the monosaccharides of interest. We also introduce sodium acetate to the mobile phase at this stage. Acetate binds to the active sites on the column, reducing chromotographic run times. This shift is proportional for all the analytes except rhamnose, where it is less pronounced. Hence, we adjust the acetate loading level to achieve elution of rhamnose between galactose and glucose. 44

45 Solid Phase Extraction Lignin and other hydrophobic components may negatively affect chromatographic performance. Off-line SPE is time consuming. We use a valve in the IC to direct flow (post injector valve and pre-column) for the first minute to position A (through an NG1 guard) and then to position B which bypasses this guard. Allows injected sample to pass through NG1 and ensure against non-reversible binding of hydrophobics to divinylbenzene backbone of analytical column. Periodic cleaning of NG1 column is necessary (acetonitrile).

46 Some methods advocate neutralisation of acid in hydrolysate and removal of sulphate prior to injection. Our samples are diluted 5X with an internal standard solution (fucose) giving approx. 0.8% acid concentration. We find that a sulphate loading of this level can improve peak sharpness and reduce retention times. Our conditions allow linearity in detection for all sugars in quantities relevant to biomass samples. Conditions: Sample Preparation Flow rate 1.1 ml/min, 18 o C 0-16 min H 2 O, 30 sec ramp to 240mM NaC 2 H 3 O 2 in 400mM NaOH, held for 2 min then 30 sec ramp to H20 Equilibration with H 2 O for 15 minutes Post column addition of base (0.3M NaOH, 0.3ml/min) required for electrochemical detection of carbohydrates.

47 Electrochemical Detector Carbohydrates detected at high sensitivity with an electrochemical detector (PAD). The gold electrode surface catalyses the oxidation of polar aliphatic compounds in media with ph > 12. Pulsed amperometry detects only those compounds that are oxidisable at the detection voltage employed. Neutral or cationic components elute in, or close to, the void volume of the column so, even if oxidsable, they do not interfere with the analysis of the carbohydrates. 47

48 Pulsed Amperometric Detection (1) Carbohydrates are detected by measuring the electrical current generated by their oxidation. But the products of this oxidation also poison the surface of the electrode meaning it must be cleaned between measurements. This is done by raising the potential to a level sufficient to oxidise the gold surface, causing desorption of the carbohydrate oxidation products. The electrode potential is then lowered to reduce the electrode surface back to gold. 48

49 Pulsed Amperometric Detection (2) E1 = carb. oxidation E2 = gold electr. oxidation E3 = gold oxide reduction t1, t2, t3 = duration of potentials E1, E2, E3 The step from one potential to the next produces a charging current that is not part of the analyte oxidation current. Analyte oxidation current is measured after a delay that allows the charging current to decay. 49

50 Pulsed Amperometric Detection (3) The carbohydrate oxidation current is measured by integrating the cell current after the delay. Current integrated over time is charge, so the detector response is measured in coulombs. Or the average current during the integration period can be reported. (units are amperes). Optimal potentials can be determined by electrochemical experiments such as cyclic voltammetry, in which the applied potentials are slowly scanned back and forth between positive and negative potential limits. 50

51 PAD Waveform Used We use Waveform A has a negative potential for electrode cleaning results in less electrode wear and long term stable results. Requires only 500ms data at 2Hz 51

52 Miscanthus Pig Slurry 250 Tetrapak

53 Wood Peat

54 IC Sequences Standard (containing known amounts of all the sugars) injected every 5 injections to determine relative response factors (RRFs) compared with fucose (internal standard). Each sample has a known quantity of internal standard, giving a response factor. This is multiplied by the most recent RRFs giving a unique response factor for all sugars for the injected sample. Separation is not perfect specific standards needed for each biomass type (e.g. grass, wood, paper, peat etc.) Sequence can be left to run over weekend (approx 100 injections).

55 IC - Conductivity Detector We also use IC for the analysis of other anions. These include products of the DIABNET process (e.g. levulinic acid, formic acid) and their intermediates (e.g. hydroxymethylfurfral). For some analytes a conductivity detector is used. This applies a pulsed voltage to two electrodes in the conductivity cell to obtain a conductivity which is measured in Siemens (S). Conductivity is dependant on the mobility of ions in the applied electric field, on the electrical charge and their concentration. 55

56 Conductivity Suppression Before Suppression F - Cl - SO 4 2- After Suppression F - Cl - SO 4 2- IC eluents used for ion chromatography contain many ions and so have high, noisy background conductance. Suppression reduces the background conductivity of the eluent. Suppressors consist of a microporous membrane. These membranes have exchange sites for ions having a charge opposite to that of the analyte. Results in the sample components being converted to their acid forms, which have higher conductance than their salt forms. The suppressor can function in three modes chemical, external water and auto regeneration.

57 Chemical Suppression - Anion Na 2 SO 4 H 2 S0 4 SO 4 2- Na + H + 2- Na + SO Na H NaOH Na + OH - OH - H + OH - H 2 0 NaCl Na + Cl - Cl - H + Cl - HCl Na H H + SO 4 2- Na + Na + SO 4 2- Na 2 SO 4 H 2 S0 4 The micromembrane has a negatively charged ion exchange site. These sites exchange the positively charged counter ions (e.g. Na +, K + ) for hydrogen ions from the regenerant. The hydrogen forms water with hydroxide leaving only the analyte anions to be detected

58 Chemical Suppression Anion SO 2-4 H + H + Na 2 SO 4 H 2 S0 4 Cation Exchange Membrane NaOH 2NaCl Na + OH - Na + OH - Na + Cl - Na + Cl - 2H 2 0 2HCl Cation Exchange Membrane H + SO 2-4 Na 2 SO 4 H 2 S0 4 H +

59 Autosuppression - Anion NaOH H 2 0 OH - Na + H + OH Na H NaOH Na + OH - OH - H + OH - NaCl Na + Cl - Cl - Cl - H H HCl Na H H + OH - Na + OH - NaOH Here the hydroxyl ions are being supplied by the eluent itself. After the eluent has been through the cell, a potential is applied to it causing dissociation. These ions exchange with the counterions in the eluent stream. H 2 0

60 KOH Concentration (mm) Conductivity (ms) Detection of LvA and FA 1 = Levulinic Acid 2 = Formic Acid Chromatogram Time (mins) 40 Eluent Hydroxide Conc Time (mins) 60

61 K + OH - EG40 -Anion KOH K + K + K + OH- K + OH - OH - OH - K + OH - Cation Exchange Membrane H 2 0

62 UV-Visible Analysis Some analytes (e.g. furfural, HMF) require ultraviolet (UV) detectors. UV radiation between 200 and 400 nm is strong enough to cause loosely held electrons to change locations. These electrons can be either the non-bonding electrons (n-electrons) of aldehydes or ketones, or they can be the -electrons of conjugated -systems. 62

63 UV Analysis of Acid Soluble Lignin A small portion of lignin is solubilized during the hydrolysis procedure. Traditionally the UV absorbance of the hydrolysate is measured at 205nm and an absorptivity (extinction coefficient) value of 110 L/g-cm used to calculate the amount of acid-soluble lignin present in the hydrolyzate. 205nm is chosen because it corresponds to a minimum in the UV spectrum of furfural and HMF, and is in a region where lignin components (e.g. vanillin, p-coumaric acid, ferulic acid) absorb strongly. Beer s Law is assumed to hold (the absorbance of an analyte is proportional to its concentration and the path length through the sample (transmission spectroscopy)). 63

64 Calculation of ASL % ASL = UVabsorbance Volume of Hydrolysate (Dilution Factor) Absorptivity (Dry weigt of sample) 100 If the ASL content is known (through other means) then the absorptivity constant can be determined for each biomass type. Absorbance data can be taken at different wavelengths according to particulars of the hydrolysate matrix. We have a diode-array UV-Vis detector so can collect all wavelength simultaneously ( nm) UV Spectra of Biomass Hydrolysate Components between 180 and 400 nm 64

65 Uronic Acids Uronic acids can be measured colorimetrically in an aliquot from the acid-hydrolysis procedure, using galacturonic acid as the calibration standard. 3,5-dimethylphenol offers certain advantages in the colorimetric determination of uronic acids in plant materials since it is less affected by hexoses present. Procedure: Hydrolysate mixed with boric acid-sodium chloride solution. Then add 18 M H 2 SO 4 and place the tubes in a 70 o C water bath for 40 minutes. Cool tubes to room temperature then add 200 ml dimethylphenol solution. Sample absorbance, A, should then be measured at 400 and 450 nm against the blank solution minutes after the addition of dimethylphenol. 65

66 Uronic Acids (2) The absorbance at 400 nm should be subtracted from that at 450 nm to correct for interference of hexoses. A calibration curve is prepared from solutions of various concentrations of galacturonic acid monohydrate. The content (%) of uronic acid residues, UA, in a sample, given as polysaccharide residues can be calculated by: Where: S UA = W u x F u x F c S = weight (dry matter, mg) of original sample; W u = weight (mg) of galacturonic acid monohydrate/100 ml hydrolysate, obtained from calibration curve; F u = factor for recalculation of galacturonic acid to polysaccharide residues (0.915 x = 0.830); and F c = compensation factor to adjust for greater degradation of free galacturonic acid as opposed to that of polygalacturonate, under conditions 66 of uronic acid calibration (F c = 0.81 is typically used).

67 Observations Approx 700 samples analysed (in duplicate) for lignocellulosic compositions. Total carbohydrates range from 2% (piglet manure) to 96% (tetrapak cartons). Approx. 350 miscanthus samples analysed (various anatomical parts). Lignin highest in lower parts of stem and nodes. As we go up the stem lignin decreases, cellulose decreases, hemicellulosic sugars increase and extractives increase. Green leaves have the lowest cellulose contents but higher amounts of extractives and ash. Extractives content higher in young plantations (first year of growth) and early in harvest window. 67

68 Summary Analysis of lignocellulosic components in biomass can be complex and imperfect. There needs to be a balance between analytical accuracy and time/cost involved. What are the most important analytes will determine what methods are used. In DIBANET our yields are dependant on the constituent units of the polysaccharides (cellulose, hemicellulose) so we focus on the accurate analysis of these and use gravimetric methods for the other components (lignin, extractives, etc.) Faster/cheaper methods of analysis would be very welcome!!! 68