"Applied Science, Faculty of"@en . "Chemical and Biological Engineering, Department of"@en . "DSpace"@en . "UBCV"@en . "Lee, Jun Sian"@en . "2015-01-07T16:06:46Z"@en . "2015"@en . "Master of Applied Science - MASc"@en . "University of British Columbia"@en . "The export of wood pellets from Canada to Europe has been increasing steadily in recent years (roughly 1.8 million ton in 2013). Due to distances involved, wood pellets remain in transit and storage for months before their final consumption. The net calorific value determines the price of wood pellets purchased in Europe. There have been concerns about the changes of net calorific values over time. In this study, the effects of storage time, storage configuration, storage temperature, and wood pellet quality on the net calorific value of wood pellets for a period of 6 months were investigated. Storage configurations were \u00E2\u0080\u009Copen\u00E2\u0080\u009D or \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D and storage temperatures were 25\u00C2\u00B0C, 35\u00C2\u00B0C and 45\u00C2\u00B0C. Two types of wood pellets used: white (10% bark) and mixed (40% bark). The results in the \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage scenario indicated that storage time had a positive effect on the net calorific value, where the net calorific value increased by 1 to 2% over the storage period. In the open storage scenario, the moisture content had the most significant impact on the net calorific value. A multivariable linear regression and analysis of variance performed verified the graphical results. It was postulated that the higher energy potential compounds \u00E2\u0080\u0093 low molecular weight aldehyde and ketone or off-gasses such as carbon monoxide, methane and hydrogen \u00E2\u0080\u0093 produced during pellet storage, caused the increase in net calorific values."@en . "https://circle.library.ubc.ca/rest/handle/2429/51791?expand=metadata"@en . " CALORIFIC VALUE OF WOOD PELLETS by Jun Sian Lee B.A.Sc., The University of British Columbia, 2012 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemical and Biological Engineering) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) January 2015 \u00C2\u00A9 Jun Sian Lee, 2015 ii Abstract The export of wood pellets from Canada to Europe has been increasing steadily in recent years (roughly 1.8 million ton in 2013). Due to distances involved, wood pellets remain in transit and storage for months before their final consumption. The net calorific value determines the price of wood pellets purchased in Europe. There have been concerns about the changes of net calorific values over time. In this study, the effects of storage time, storage configuration, storage temperature, and wood pellet quality on the net calorific value of wood pellets for a period of 6 months were investigated. Storage configurations were \u00E2\u0080\u009Copen\u00E2\u0080\u009D or \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D and storage temperatures were 25\u00C2\u00B0C, 35\u00C2\u00B0C and 45\u00C2\u00B0C. Two types of wood pellets used: white (10% bark) and mixed (40% bark). The results in the \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage scenario indicated that storage time had a positive effect on the net calorific value, where the net calorific value increased by 1 to 2% over the storage period. In the open storage scenario, the moisture content had the most significant impact on the net calorific value. A multivariable linear regression and analysis of variance performed verified the graphical results. It was postulated that the higher energy potential compounds \u00E2\u0080\u0093 low molecular weight aldehyde and ketone or off-gasses such as carbon monoxide, methane and hydrogen \u00E2\u0080\u0093 produced during pellet storage, caused the increase in net calorific values. iii Preface This master\u00E2\u0080\u0099s thesis is divided into five chapters. The author, Jun Sian Lee, has done the experimental design, performing experiments, material characterization, literature review, data analysis, and thesis preparation under the supervision of Professor Shahab Sokhansanj and Professor Anthony K. Lau in the Chemical and Biological Engineering Department at the University of British Columbia. The material in Chapter 3 is prepared for publication under the following citation: 1. Lee, Jun Sian; Sokhansanj, Shahab; Lau, Anthony; Lim, C. Jum, Bi, Xiaotao, Bassett, Vaughan and Melin, Staffan, \u00E2\u0080\u009CThe effects of long term storage on the net calorific value of wood pellets\u00E2\u0080\u009D. In Review. Table of Contents Abstract ........................................................................................................................................... ii Preface............................................................................................................................................ iii Table of Contents ........................................................................................................................... iv List of Tables ................................................................................................................................. vi List of Figures .............................................................................................................................. viii List of Symbols .............................................................................................................................. ix Acknowledgements ......................................................................................................................... x Dedication ...................................................................................................................................... xi Chapter 1 Introduction ............................................................................................................. 12 1.1 Current state of wood pellet industry ............................................................................. 12 1.2 Motivation and thesis statement ..................................................................................... 15 1.3 Objectives ....................................................................................................................... 16 Chapter 2 Background ............................................................................................................. 17 2.1 Definition and uses of the term, calorific value ............................................................. 17 2.1.1 Acronym of calorific values in this thesis ............................................................... 20 2.1.2 Conversion of NCV from GCV .............................................................................. 21 2.2 Factors that define the calorific values of wood pellets ................................................. 21 2.2.1 Moisture content and ash content ........................................................................... 23 2.2.2 Elemental composition............................................................................................ 24 2.2.3 Chemical composition ............................................................................................ 26 2.3 Effect of storage on calorific values of woody biomass ................................................ 28 2.3.1 Chemical and biological degradation of woody biomass during storage ............... 28 2.3.2 Configurations of storage ........................................................................................ 29 2.3.3 Storage temperatures ............................................................................................... 29 Chapter 3 The Change of Calorific Values of Wood Pellets during Storage .......................... 31 3.1 Introduction .................................................................................................................... 31 3.2 Methods and materials ................................................................................................... 31 3.2.1 Experimental design................................................................................................ 31 3.2.2 Procedure and apparatus ......................................................................................... 33 v 3.2.3 Inherent uncertainties of bomb calorimeter ............................................................ 37 3.3 Results ............................................................................................................................ 37 3.3.1 The observed changes in moisture content and ash content and their effects on calorific value........................................................................................................................ 38 3.3.2 The carbon content after storage ............................................................................. 40 3.3.3 ANOVA analysis to determine the effects of pellet type, storage configuration and storage temperature on NCV ................................................................................................ 42 3.4 Conclusion ...................................................................................................................... 43 Chapter 4 Interpreting and Modelling the Change of Calorific Values .................................. 49 4.1 Introduction .................................................................................................................... 49 4.2 Methods .......................................................................................................................... 49 4.2.1 Multivariable linear regression model (MLR) ........................................................ 49 4.2.2 Single variable linear model ................................................................................... 50 4.3 Result .............................................................................................................................. 51 4.3.1 Comparing the individual cases of MLR ................................................................ 51 4.3.2 The application of the single variable equation ...................................................... 54 4.4 Conclusion ...................................................................................................................... 58 Chapter 5 Conclusions and Future Work ................................................................................ 62 5.1 Conclusions .................................................................................................................... 62 5.2 Future work .................................................................................................................... 65 References ..................................................................................................................................... 66 Appendix A: Supporting Information ........................................................................................... 70 A.1 Additional tables of data ................................................................................................ 70 A.2 Derivation of constant pressure NCV from calorimetric GCV .................................... 101 A.3 Principle of bomb calorimeter ...................................................................................... 105 A.4 Classification of wood pellets ...................................................................................... 107 vi List of Tables Table 2-1: Calorific values (HHV in MJ/kg) of oven dry tree components of softwoods (Singh and Kostecky 1986) ...................................................................................................................... 22 Table 2-2: Calorific values (HHV in MJ/kg) of oven dry tree components of hardwoods ((Singh and Kostecky 1986) ...................................................................................................................... 22 Table 2-3: Calorific value, moisture, ash and carbon content of pellets at a range of bark content (Filbakk et al. 2011) ...................................................................................................................... 23 Table 2-4 Gross Calorific Values of the chemical components of woody biomass (White, 1984 and Rh\u00C3\u00A9n, 2004) ............................................................................................................................ 26 Table 3-1: Pellet types, storage configuration and storage temperatures ..................................... 32 Table 3-2: One-way analysis of variance on the NCV of two wood pellet types ......................... 42 Table 3-3: Two-way analysis of variance on NCV of two storage configuration (\u00E2\u0080\u009COpen\u00E2\u0080\u009D or \u00E2\u0080\u009CClose\u00E2\u0080\u009D and three storage temperatures (25\u00C2\u00B0C, 35\u00C2\u00B0C, 45\u00C2\u00B0C) ....................................................... 43 Table 4-1: Summary of the squared correlation coefficients and model coefficients for all three stages of the multivariable linear regression (MLR) models ........................................................ 60 Table 4-2: Summary of the R2 values of all nine cases analyzed using multivariable linear regression. The table ranks the R2 from the highest to the lowest. ............................................... 61 Table 4-3: Two single-variable linear models for open and closed storage correlate total duration of storage to obtain 3% increase from initial calorific value in days, to the storage temperature in \u00C2\u00B0C for each storage configuration. The data from both white and mixed pellets are used to form this linear model. .................................................................................................................. 61 Table 5-1: Comparing the initial calorific values of white pellets and mixed pellets .................. 65 Table A-1 Effective heat capacity result from 10 calibration runs ............................................... 70 Table A-2: Carbon content and NCV data used to calculate correlation coefficient. ................... 71 Table A-3 Comparison between measurements done by SGS Delta analytical lab and in UBC for samples that are stored for 42 days (1.5 months).......................................................................... 72 Table A-4: Comparison between the measurement method I employed and SGS Delta Analytical Laboratory used. ........................................................................................................................... 73 Table A-5: Comparison between measurements done by SGS Delta analytical lab and in UBC for samples that are stored for 84 days (3 months) ....................................................................... 75 vii Table A-6: The raw data that are used in this thesis. NCV is calculated from measured GCV using equation 3-1 using average hydrogen content. .................................................................... 76 Table A-7: Example of the integration of calorific value over storage period using the coefficients in Table 4-3 \u00E2\u0080\u009COpen\u00E2\u0080\u009D storage configuration. The daily temperatures were obtained from an unnamed wood pellet voyage. This table is only for illustration. ................................... 95 Table A-8: Example of the integration of calorific value over storage period using the coefficients in Table 4-3 \u00E2\u0080\u009CClosed\u00E2\u0080\u009D storage configuration. The daily temperatures were obtained from an unnamed wood pellet voyage. This table is only for illustration. ................................... 98 Table A-9: The quality parameters as required by EN-plus certification standard (A1, A2 and B), Industrial Wood Pellet Buyer (IWPB) standard (I1, I2 and I3) and corresponding testing standards. .................................................................................................................................... 107 viii List of Figures Figure 1-1: North America Pellet Export Prices. Pellet prices are in Canadian dollars (IEA Bioenergy, 2011)........................................................................................................................... 14 Figure 1-2: FETM Biomass Co-firing Index and the price of industrial wood pellets cif ARA (Murray, 2012). For explanation of \u00E2\u0080\u009Ccif ARA\u00E2\u0080\u009D, please see List of Symbols. .............................. 15 Figure 3-1: Open (left) and closed (right) storage configurations. ............................................... 33 Figure 3-2: A typical bomb calorimeter (adapted from Thomson-Brooks/Cole (2003)) ............. 34 Figure 3-3: Wood pellets moisture content on as-received basis (% w.b.) over 180-day of storage for (a) white \u00E2\u0080\u009Copen\u00E2\u0080\u009D, (b) white \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D, (c) mixed \u00E2\u0080\u009Copen\u00E2\u0080\u009D, and (d) mixed \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D. ................. 45 Figure 3-4: Wood pellet ash content on dry basis (% d.b.) over 180-day of storage for (a) white \u00E2\u0080\u009Copen\u00E2\u0080\u009D, (b) white \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D, (c) mixed \u00E2\u0080\u009Copen\u00E2\u0080\u009D, and (d) mixed \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D. ...................................... 46 Figure 3-5: Wood pellets net calorific values over 180-day of storage for (a) white \u00E2\u0080\u009Copen\u00E2\u0080\u009D, (b) white \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D, (c) mixed \u00E2\u0080\u009Copen\u00E2\u0080\u009D, and (d) mixed \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D. ......................................................... 47 Figure 3-6: Wood pellets carbon content over 80-day of storage for (a) white \u00E2\u0080\u009Copen\u00E2\u0080\u009D, (b) white \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D, (c) mixed \u00E2\u0080\u009Copen\u00E2\u0080\u009D, and (d) mixed \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D. ................................................................... 48 Figure 4-1: The cumulative sum of fraction of days approaching the 1.03 times (3% increase in) initial calorific value with respect to the duration of storage (days), where (a) gives the data for open storage and (b) gives the data for closed storage. ................................................................ 57 Figure A-1: The steps to convert high heating value \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089 (dry basis) to net calorific value \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089 (wet basis) ................................................................................................................................... 104 ix List of Symbols AC = ash content ANOVA = analysis of variance CHN = carbon, hydrogen, nitrogen contents cif ARA price of pellets = the price of pellets received at the port of Amsterdam, Rotterdam and Antwerp area traded on a cost, insurance and freight basis. d.b. or db = dry mass basis EMC = equilibrium moisture content EU = European Union GCV = gross calorific value (MJ/kg), always in wet basis HHV = high heating value (MJ/kg), always in dry basis LHV = low heating value (MJ/kg), always in dry basis MC = moisture content (% wt) MLR = multivariable linear regression MMT = million metric tonnes NCV = net calorific value (MJ/kg), always in wet basis w.b. or wb = wet mass basis x Acknowledgements I wish to express my deepest gratitude toward Dr. Shahab Sokhansanj and Dr. Anthony Lau, who have introduced me into the fast-growing industry of wood pellets. Only with their insights and guidance, this work can be completed. I also thank Mr. Vaughan Bassett and Mr. Harold Berkholtz for their practical insights of the wood pellet industry. Through them, I was exposed to a working wood pellet plant in British Columbia. I am extremely grateful for the mentorship provided by Dr. Fahimeh Yazdanpanah and the laboratory assistance given by Ms. Monirehsadat Hedayati as well as the help provided by all the members of the UBC Biomass and Bioenergy Research Group. Lastly, a special and big thank you to my parents, who have supported me morally and financially through my education in UBC. xi Dedication I wish to dedicate this master thesis to my girlfriend, Nikki Luk Chuen Tse who is always here for me every step of the way.12 Chapter 1 Introduction 1.1 Current state of wood pellet industry In recent years, the wood pellet has emerged as a reliable feedstock for production of heat and power, especially in the European Union (EU). According to a recent report (Flach et al. 2013), the EU consumed 14.3 million metric tonnes (MMT) of pellets in 2012. The consumption of wood pellets in EU is expected to increase by more than 3 times to 50-80 MMT by year 2020. At present, Canada is among the major wood pellet exporters; roughly 1.3 MMT of wood pellets was exported to mainly EU in 2012 (Natural Resources Canada 2013). Majority of the Canadian wood pellet production originated at the province of British Columbia (BC), where 65% of the 3 MMT total wood pellet production capacity is located (IEA Bioenergy 2010). According to a United States Department of Agriculture (USDA) report, the main consumers of BC wood pellets are large co-firing power generation plants in United Kingdom, Netherlands, and Belgium (Flach 2012). The BC wood pellets are shipped through an approximately 16,500-kilometre route from ports in British Columbia to industrial pellet exchange ports in Netherlands. Before wood pellets are transported, they are commonly hold in flat bottom bin silos. Each bin held roughly 4500 MT of wood pellets. The bin silos are mounted with centrifugal fans for ventilations. The fans were automatically shut down when the ambient relative humidity exceed 90% (Larsson et al 2012). Air conveyor are usually used to load wood pellets onto ocean vessel holds. The holds are immediately sealed after the loading operations. Wood pellets are stored in hatched holds (cargo spaces) of roughly 5500 m3, each holding roughly 4000 tonnes of wood pellets (Melin et al. 2008). Modern open hatch box-shaped bulk carriers may have anywhere from a single hold up to 12 cargo holds. During the ocean voyage, the atmospheric temperature 13 in the hold is about 5 degree Celsius lower than the ocean water temperature (Melin et al. 2008). Finally, before discharge operation, the stairways are ventilated to remove any carbon monoxide generated by the reaction between wood pellets and atmospheric oxygen during the ocean voyage. The raise in demand of wood pellets for residential use over the recent years resulted in an increase in the pellet\u00E2\u0080\u0099s average retail price every year. In first quarter of year 2013, the average retail price of wood pellets in Europe was \u00E2\u0082\u00AC240 per metric tonne (MT), which was 20% higher compared to $200 per MT in the same quarter in year 2012 (Krajnc et al. 2013). Unlike commodity trades whereby price is set on unit-weight basis, the fuel pellet market price is primarily based on the calorific value of wood pellets. The competitiveness of wood pellet depends on its energy price advantage compared to other fuels. For example, in the United States, the cost of using wood pellet is viewed as the cost of energy, which is measured in dollars per million British thermal units (IEA Bioenergy, 2011). IEA Bioenergy (2011) stated that pellets purchased at the average $150 per MT and burned in a typical pellet stove cost about $14.00 per million Btu (or about $15.00 per gigajoule). This is less than the cost of electric heating (roughly $30 per million Btu or $32 per GJ) and competitive with average energy costs of some other fuels, but less competitive than natural gas prices, which is priced around $5 per million BTU or $5.5 per GJ. In Canada, pellet export contracts are in Euros (IEA Bioenergy, 2011). IEA Bioenergy (2011) saw the decline in pellet export price in Canada as result of the financial crisis in several EU countries in year 2010. In that year, the average export contract price decreased from $154/MT to $125/MT, as shown in Figure 1-1. At the same period, the European pellet price (cif 14 ARA) decreased from \u00E2\u0082\u00AC27/MWh to \u00E2\u0082\u00AC25/MWh, or a decrease from \u00E2\u0082\u00AC7.50/GJ to \u00E2\u0082\u00AC7.00/GJ. This decreasing price is shown in Figure 1-2 (Murray 2012). Figure 1-1: North America Pellet Export Prices. Pellet prices are in Canadian dollars (IEA Bioenergy, 2011). To provide a clear comparison between the energy cost of using wood pellet in co-firing versus using coal, Grbovic (2010) provided an example of index developed by Hawkins Wright Ltd. Biomass Co-firing Index (BCI) to track the competitiveness of co-firing biomass, relative to burning coal in a typical electricity generating plant, which represents the price that plants are willing to pay for biomass. Here, calorific value is using to calculate the energy cost from using wood pellets. For example, On April 30th, 2009, BCI was $14/MWh or $4/GJ cif ARA and was comprised of: the spot price of coal, the price of emissions allowances necessary to cover the CO2 emitted by coal combustion, and the combustion efficiency adjustment (Hawkins Wright 2009). The subtraction of this amount from the price of wood pellets on the same month ($36.52/MWh or $10.14/GJ, assuming calorific value of 4.72MWh/MT or 17 GJ/MT of wood pellets) gives a BCI spread of $22.5/MWh or $6.25/GJ, which has to be compensated by policy instruments (i.e. subsidies, tax credits, etc.) in order for biomass to be competitive with coal 15 (Grbovic 2010). Figure 1-2 illustrates the change of the BCI spread from January 2010 to October 2012. Figure 1-2: FETM Biomass Co-firing Index and the price of industrial wood pellets cif ARA (Murray, 2012). For explanation of \u00E2\u0080\u009Ccif ARA\u00E2\u0080\u009D, please see List of Symbols. 1.2 Motivation and thesis statement Wood pellets are transported by rail from interior BC to Vancouver, BC, and stored in silos before unloading to ocean vessels. Due to the long distance travelled by wood pellets, changes in fuel quality, calorific value in particular, are expected to occur due to auto-oxidation or biological reactions and moisture adsorption from the atmosphere (Melin et al 2008, Yazdanpanah et al. 2014). However, contradicting reports from the pellet industry concerning the changes in fuel quality were obtained. In an unpublished report, Melin and Lee (2013) mentioned about 1 % decrease in the calorific value over 2 months of ocean cargo transportation. However, Basset (2013) stated that the calorific value increased by 1% to 2% over the similar period of ocean transportation. In view of the uncertainties in the changes in calorific value of the wood pellets during ocean transportation, this thesis research study was conducted to quantify the changes in calorific 16 value in a systematic manner and delineate the possible reasons that causes such changes. From this research, the author believes that a carefully designed \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage system could potentially increase the calorific value of wood pellets. This increase in calorific value was potentially caused by re-adsorption of higher energy potential compounds, such as low molecular-weight phenolic, ketones and aldehydes, or off-gasses, such as carbon monoxide (CO), methane (CH4) and hydrogen (H2). 1.3 Objectives The objectives of this thesis are 1. to investigate the change of the calorific value of wood pellet through a six-month storage experiment, and 2. to establish the relationship between the calorific value of wood pellets and the storage environment and the key physical properties of the pellets. 17 Chapter 2 Background 2.1 Definition and uses of the term, calorific value Historically, heating values or calorific values are used in combustion engineering to calculate the thermal efficiency of steam boilers. These steam boilers are most commonly run on coal, natural gas or heavy oils. According to the Dictionary of Mechanical Engineering (2014), the calorific value of a fuel (or heat of combustion or heating value or heat value) is defined as \u00E2\u0080\u009Cthe energy released per unit mass of fuel in complete combustion with oxygen.\u00E2\u0080\u009D For solid and liquid fuels, calorific value can be expressed in Btu/lb (English units) or MJ/kg (GJ/tonne) (SI units), and for gaseous fuels, in Btu/ft3 or GJ/m3. The values are determined by burning the fuel in a bomb calorimeter, which is defined as \u00E2\u0080\u009Ca device used to determine the calorific value of a fuel, where a small mass of the fuel is burned in oxygen at constant volume in small stainless steel pressure vessel (the bomb) immersed in a bath of water (calorimeter).\u00E2\u0080\u009D The temperature rise of the water bath, resulting from the fuel burning in the bomb calorimeter, is used to calculate the calorific value. The calorific value is then referred to as the \u00E2\u0080\u009Cgross calorific value (GCV)\u00E2\u0080\u009D or the \u00E2\u0080\u009Chigher heating value (HHV).\u00E2\u0080\u009D The working principle of a bomb calorimeter is given in Appendix A.3. Not to be mistaken with the calorific value of food ingredients, the term \u00E2\u0080\u009Ccalorific value\u00E2\u0080\u009D in this thesis strictly refers to the energy content of a fuel. It has long been known that coal, natural gas, petroleum and wood fuels contain hydrogen as one of the constituents (Babcock and Wilcox Company 1955; Baker 1982). Water (H2O) is formed as a product of combustion when the hydrogen reacts with oxygen in air. This water may remain in the vapor state or it may be condensed to the liquid state, giving a substantial difference in the heat value. Hence, two values are determined to take into account this 18 difference: the gross calorific value corresponds to \u00E2\u0080\u009Cthe water in the combustion products being in the liquid phase\u00E2\u0080\u009D and the net calorific value (NCV), or lower heating value (LHV) to the vapour phase. According to the Dictionary of Mechanical Engineering (2014), the equation below relates the two values HHV and LHV, \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089 = \u00F0\u009D\u0090\u00BF\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089 +\u00F0\u009D\u0091\u009A\u00F0\u009D\u0090\u00BB2\u00F0\u009D\u0091\u0082\u00E2\u0084\u008E\u00F0\u009D\u0091\u0093\u00F0\u009D\u0091\u0094/\u00F0\u009D\u0091\u009A\u00F0\u009D\u0090\u00B9\u00F0\u009D\u0091\u0088\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0090\u00BF (2-1) where \u00E2\u0084\u008E\u00F0\u009D\u0091\u0093\u00F0\u009D\u0091\u0094 is the latent heat of vaporization of water, \u00F0\u009D\u0091\u009A\u00F0\u009D\u0090\u00BB2\u00F0\u009D\u0091\u0082 is the mass of water produced from combustion of hydrogen in a dry fuel, and \u00F0\u009D\u0091\u009A\u00F0\u009D\u0090\u00B9\u00F0\u009D\u0091\u0088\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0090\u00BF is the mass of fuel in dry basis. However, Briggs (1994), defined HHV in \u00E2\u0080\u009CForest Products Measurements and Conversion Factors,\u00E2\u0080\u009D as a laboratory measurement of the stored chemical energy for \u00E2\u0080\u009Coven-dry wood (zero-per-cent moisture content)\u00E2\u0080\u009D, which is equivalent to the GCV on dry basis. Briggs further defined the gross heating value (GHV) as a separate quantity that \u00E2\u0080\u009Crepresents the available potential heat at given moisture content and is equal to HHV only if the wood is oven-dry\u00E2\u0080\u009D. Thus, GHV is equivalent to the GCV on wet basis (w.b.). GHV is related to HHV through the following equation: \u00F0\u009D\u0090\u00BA\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089 = \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089 (1 \u00E2\u0088\u0092\u00F0\u009D\u0091\u0080\u00F0\u009D\u0090\u00B6100) (2-2) where \u00F0\u009D\u0091\u0080\u00F0\u009D\u0090\u00B6 is the moisture content, in percent (w.b.) Through the use of the term \u00E2\u0080\u009CHHV\u00E2\u0080\u009D on oven dry basis (d.b.) and \u00E2\u0080\u009CGHV\u00E2\u0080\u009D on wet basis, the heating values are effectively separated, avoiding any potential confusion. Briggs (1994) defined lower heating value (LHV) and net heating value (NHV) in a similar manner. Thus, LHV and NHV are defined as \u00E2\u0080\u009Cthe net heat available in a fuel after the various heat losses associated with wood moisture content and water formed during combustion are subtracted from 19 HHV or from GHV, respectively\u00E2\u0080\u009D. From these definitions, it can be concluded that GCV is calorimetrically determined from the bomb calorimeter at constant volume condition where water vapor, from both the wood moisture content and the product of combustion, is condensed at the end of an actual calorimetric measurement. Yet, neither the Dictionary of Mechanical Engineering (2014) nor Briggs (1994) mentioned the thermodynamic conditions from which the LHV or NHV are determined, i.e. whether these values are determined under constant volume or constant pressure conditions. Citing to EN 14918 \u00E2\u0080\u0093 Solid Biofuels \u00E2\u0080\u0093 Determination of Calorific Values, biofuels are burned at \u00E2\u0080\u009Cconstant (atmospheric) pressure\u00E2\u0080\u009D and the water is either not condensed (removed as vapour with the flue gases) or condensed. Since wood pellet is considered a type of biofuel, NCV of wood pellets at constant pressure is used as the operative heat of combustion. NCV at constant pressure is calculated by subtracting the losses of latent heat of vaporization associated wood moisture content and water formed during combustion from gross calorific value. For NCV, the water is assumed to remain in the vapor phase. Depending on the type of fuel, the GCV is generally 3-8% higher than the NCV, for the same weight basis and the thermodynamic state assumptions. On average, the GCV (19 GJ/MT) of wood pellet is about 5% higher than its NCV (18 GJ/MT). According to Babcock and Wilcox Company (1995), in the United States, the practice in boiler combustion calculations is to use the GCV, since these values are available directly from calorimeter determinations. Also, GCV serves as the basis on which fuel is bought and sold. In Europe, NCV values are generally used for purchase of fuel due to historical reasons. In the practical use, Be\u00C3\u00A9r (2005) stated that the NCV has a greater importance than the GCV, especially in internal combustion engine, where the latent heat of water vapour is not recovered. Hence, NCV is used in the rating of engines. The GCV, on another hand, is used to determine the 20 thermal efficiency of combustion plants, where the water vapour is condensed and therefore the latent heat of vaporization of water vapour is recovered as usable energy. In the wood pellet industry, NCV on an \u00E2\u0080\u009Cas-received\u00E2\u0080\u009D or wet basis is most commonly used, which follows the convention in Europe (European Pellet Council 2013; Pellet Fuel Institute 2011). The European Union is the largest consumer of wood pellets, consuming 14 million metric tonnes (MMT) of pellets in year 2012 (Flach et al 2013). To serve the European market, the non-European wood pellet producers in United States and Canada adopted the NCV convention and other European quality standards (EN-plus) in wood pellet commodity trade. As given by the EN-plus Certification Standard, the minimum NCV value of wood pellets at constant pressure and \u00E2\u0080\u009Cas-received or wet basis\u00E2\u0080\u009D shall be no less than 16.5 GJ/MT (EN-plus A1 for residential uses), no less than 16.3 GJ/MT (EN-plus A2 for institutional uses) and no less than 16.0 GJ/MT (EN B for industrial uses). The EN-plus standard is given in Appendix A-4. As stated in EU Biofuels Annual 2013 prepared by USDA, net calorific values are also used to determine the amount of wood pellets required to satisfy the wood pellet demand, given a regulatory requirement. Often, the Renewable Energy Action Plans of the European Union member states are based on the percentage of renewable energy in a certain sector of energy usage. For example, in Netherlands, based on NCV of 17.8 GJ/MT or 4.9 MWh/MT wood pellets, 2.8 million metric tonnes (MT) of wood pellets are needed to meet its 10% target of 500 million GJ biomass co-firing. 2.1.1 Acronym of calorific values in this thesis In this thesis, the acronym HHV shall refer to the oven dry or moisture-free higher heating value, while GCV will be used exclusively for wet basis or as-received gross calorific 21 value. The acronym LHV shall refer to the oven dry or moisture-free lower heating value, while NCV will be used exclusively for wet basis or as-received net calorific value. 2.1.2 Conversion of NCV from GCV As taken from European testing standard EN 14918 \u00E2\u0080\u0093 Solid Biofuels \u00E2\u0080\u0093 Determination of Calorific Values, equation 2-3 shows the LHV (d.b.) at constant volume, expressed in J/g dry mass, as a function of HHV, hydrogen \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB, oxygen \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 and nitrogen \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081 contents (d.b.), thus \u00F0\u009D\u0090\u00BF\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089(\u00F0\u009D\u0091\u0091. \u00F0\u009D\u0091\u008F. ) = \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089(\u00F0\u009D\u0091\u0091. \u00F0\u009D\u0091\u008F. ) \u00E2\u0088\u0092 212 \u00C3\u0097 \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB \u00E2\u0088\u0092 0.8 \u00C3\u0097 (\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 + \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081) (2-3) The NCV (w.b.) and at constant pressure [J/g wet mass] can be expressed in terms of GCV (w.b.) and at constant volume, moisture content \u00F0\u009D\u0091\u0080 (w.b.) and the three elemental composition through equation 2-4. \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089(\u00F0\u009D\u0091\u00A4. \u00F0\u009D\u0091\u008F. ) = \u00F0\u009D\u0090\u00BA\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089(\u00F0\u009D\u0091\u00A4. \u00F0\u009D\u0091\u008F. ) \u00E2\u0088\u0092 [212 \u00C3\u0097 \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB + 0.8 \u00C3\u0097 (\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 + \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081)] \u00C3\u0097100 \u00E2\u0088\u0092\u00F0\u009D\u0091\u0080100 \u00E2\u0088\u0092 24.5\u00F0\u009D\u0091\u0080 (2-4) Equations 2-3 and 2-4 include the work done on the system during the conversion from constant volume to constant pressure. However, clear derivation and explanation are not given in the European standard. The American standard ASTM D5865 - Standard Test Method for Gross Calorific Value of Coal and Coke gives a detailed derivations of equations 2-3 and 2-4. The derivation of NCV at constant pressure from HHV at constant volume involves understanding the work done when fuel is burned at constant pressure. The detail derivation is given in Appendix A.2. 2.2 Factors that define the calorific values of wood pellets The calorific values of wood pellets depend on many factors. From forestry science point of view, the wood pellets made from conifers (more resinous softwood species) have higher calorific values than deciduous trees (less resinous hardwood species). Taking examples from 22 Tables 2-1 and 2-2, on average, wood pellets made from jack pine (softwood tree) have a HHV of 20.8 GJ/dry MT, while wood pellets made from aspen (hardwood tree) have a HHV of 19.3 GJ/dry MT. It is also evident that different tree components, such as stump, stem, treetop, bark, foliage, and branches have different calorific values. Hence, pellets made from stem wood or other tree components must have different calorific values. Table 2-1: Calorific values (HHV in MJ/kg) of oven dry tree components of softwoods (Singh and Kostecky 1986) Species Stump Stem Treetop Bark Foliage Branches Mean White Spruce 19.8 19.0 21.5 19.8 20.5 21.1 20.3 Black Spruce 19.2 18.9 21.6 19.5 20.9 20.7 20.1 Jack Pine 19.9 19.4 21.2 21.3 21.4 21.4 20.8 Eastern White cedar 19.4 20.0 19.5 18.7 21.4 18.7 19.6 Tamarack 19.9 18.8 21.3 19.5 20.1 21.5 20.2 Balsam fir 19.6 18.7 21.4 18.5 21.5 20.6 20.1 Mean 19.6 19.1 21.1 19.6 21.0 20.7 20.2 Standard Deviation 0.3 0.5 0.8 1.0 0.6 1.0 Note that MJ/kg and GJ/MT are equivalent units, i.e. 1 MJ/kg = 1 GJ/MT Table 2-2: Calorific values (HHV in MJ/kg) of oven dry tree components of hardwoods ((Singh and Kostecky 1986) Species Stump Stem Treetop Bark Foliage Branches Mean Aspen 18.7 18.7 20.2 19.5 18.8 19.9 19.3 Balsam poplar 18.4 17.7 20.5 19.5 17.7 19.1 18.8 White birch 18.9 18.5 19.8 20.2 21.1 19.7 19.7 Manitoba maple 18.7 18.7 19.8 18.6 17.2 19.4 18.7 Mean 18.7 18.4 20.1 19.5 18.7 19.5 19.1 Standard Deviation 0.2 0.5 0.3 0.7 1.7 0.3 Note that MJ/kg and GJ/MT are equivalent units, i.e. 1 MJ/kg = 1 GJ/MT Filbakk et al. (2011) studied the effect of mixing certain percentages of Scots pine bark with stem wood chip to produce wood pellets. They attributed the greater calorific value of bark vs. stem wood (by about 5%) (Tables 2-1 and 2-2) to the higher extractives and lignin contents of 23 bark, which led to higher carbon content. Nevertheless, they found that the addition of bark (up to 30% bark content in the mixture) did not increase the calorific value significantly (Table 2-3). Table 2-3: Calorific value, moisture, ash and carbon content of pellets at a range of bark content (Filbakk et al. 2011) Bark Content (% d.b.) Moisture content (% d.b.) GCV (MJ/ kg w.b.) HHV (MJ/kg d.b.) Ash content (% d.b.) Carbon content (% d.b.) 0 6.6 19.3 20.7 0.47 50.9 5 5.5 19.4 20.5 0.45 50.9 10 6.5 19.4 20.7 0.55 50.9 30 5.5 19.5 20.6 0.82 51.0 100 9.5 19.7 21.8 2.5 51.8 The above example suggested that the difference in calorific value between different tree components and tree species can be explained by four major factors: moisture content, ash content, wood chemical composition and elemental composition. 2.2.1 Moisture content and ash content The moisture (or water) and minerals in wood pellets are non-combustible. During wood pellet combustion, the woody materials are oxidized by oxygen, resulting in release of thermal energy. Energy is required to heat up the water up to its boiling temperature and to vaporize it Wood pellets are hygroscopic. They absorb moisture easily when exposed to humid air. The increase in moisture content can be a reason behind the reduction in calorific value in wet basis (w.b.). In literature, solid biofuels that contain high moisture content and/or high ash content have lower NCV (Demirbas 2002; Platace, Adamovics and Gulbe 2013; Toscano et al. 2013). Moreover, only at a sufficiently high moisture content, fungi and other microorganisms can degrade biomass materials and hence alter the chemical and elemental compositions of the 24 biomass. The threshold moisture content for supporting microbial growth on woody biomass is often quoted as 20% (w.b.) in the literature (for instance, Byrne et al. 2005; Deacon 2006). Energy is required to heat up and oxidize minerals in woody materials, forming ash residues. The ash content of solid biofuels is dependent on the biomass species, the tree component (such as stem, branches, or root), and cleanliness of the handling system. Under static storage conditions, ash content of wood pellet is not expected to change. Here, static storage is defined as storage of materials in an environment where the materials are not in any physical motion. After debarking, clean wood has ash content in the range of 0.3-1.0% w.b. The bark, branches and stump usually have higher ash content in the range of 1.0-2.5% w.b., presumably due to contamination during the handling processes (Cassidy and Ashton 2007). The unprocessed wood fuels are referred to as hog fuels. Higher ash content is undesirable, as it not only decreases the calorific value of the fuel but also causes corrosion and slagging in combustion boilers. 2.2.2 Elemental composition The calorific value of biomass is influenced by its elemental composition, particularly the carbon, hydrogen and oxygen contents. Different species of biomass have different elemental composition; hence they have different calorific values (Demirbas 2002). The calorific value of biomass has been widely correlated to the elemental composition (Gaur and Reed 1995; Channiwala and Parikh 2001; Yin 2011). In the book \u00E2\u0080\u009CAn Atlas of Thermal Data for Biomass and Other Fuels\u00E2\u0080\u009D published by USA National Renewable Energy Laboratory, Gaur and Reed (1995) introduced a widely used correlation between the HHV (d.b.) and the elemental composition - carbon (C), hydrogen (H), sulphur (S), oxygen (O), nitrogen (N) and ash (A) contents for biomass fuel. The correlation equation was initially presented by Channiwala (1992) in his PhD thesis and published in year 2001. 25 \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089 = 0.3491\u00F0\u009D\u0090\u00B6 + 1.1783\u00F0\u009D\u0090\u00BB + 0.1005\u00F0\u009D\u0091\u0086 \u00E2\u0088\u0092 0.1034\u00F0\u009D\u0091\u0082 \u00E2\u0088\u0092 0.0151\u00F0\u009D\u0091\u0081 \u00E2\u0088\u0092 0.0211\u00F0\u009D\u0090\u00B4 (2-5) Where HHV is the high heating value in dry basis (d.b.), expressed in MJ/kg and C, H, O, N, S and A are carbon, hydrogen, oxygen nitrogen, sulphur and ash contents of materials respectively, expressed in mass percentages on dry basis. The coefficients of Eqn (2-19) imply that carbon, hydrogen and sulfur contents have positive correlation with HHV (d.b.) whereas nitrogen, oxygen and ash contents have negative correlations with HHV (d.b.). Casal et al. (2010) observed a reduction in the carbon content and hence a reduction in the calorific value in a wood chip storage experiment. Melin et al. (2008) observed oxygen depletion when linseed oil was placed in a sealed bag for a period of time. They postulated that the solid biofuel might absorb oxygen from the storage environment, thereby increasing in the oxygen content of the fuel. As mentioned above, Eqn. (2-19) indicates that an increase in oxygen content of the fuel will decrease its calorific value. Eqn. (2-19) was adopted by Obernberger and Thek (2010) in \u00E2\u0080\u009CThe Pellet Handbook\u00E2\u0080\u009D. They checked the validity of the equation by calculating the HHV of 40 samples of commercially produced wood pellets in Europe and showed that on average the measured HHVs are1.8% lower than the calculated values. One of the latest attempts to correlate the HHV (d.b.) of various biomass fuels to their elemental composition was performed by Yin (2011). Equation shows the correlation equation. \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089 = 0.2949\u00F0\u009D\u0090\u00B6 + 0.8250\u00F0\u009D\u0090\u00BB (2-6) The equation showed a high degree of fitness with the author\u00E2\u0080\u0099s experimental data with adjusted R2 value of 0.97. 26 2.2.3 Chemical composition The calorific value of biomass is influenced by its chemical composition (Demirbas 2002). The four main chemical components of wood are cellulose, lignin, hemicellulose and extractives. The weight percentage of each chemical component varies depending on the biomass species. In a typical tree stem, cellulose is the most abundant chemical, making up to 45% of the tree, followed by hemicellulose (20% to 30%), lignin (25% to 30%) and extractive (3% to 10%). Due to their similar combustion properties, hemicellulose and cellulose are often grouped together and referred to as holocellulose, which is the total polysaccharide fraction of wood or other biomass species. Extractives are low-molecular-weight volatile phenolic substances including flavanoids, stilbenes, lignans, tannins and quinones, which exist in every form of plants (Rowell, 2012). Table 2-4 lists the estimated GCVs on dry basis for the various chemical components of wood. Extractives have the highest GCV, followed by lignin, cellulose and hemicellulose. Table 2-4 Gross Calorific Values of the chemical components of woody biomass (White, 1984 and Rh\u00C3\u00A9n, 2004) Components Gross Calorific Value in dry basis (MJ/kg) Holocellulose (Cellulose and Hemicellulose) 18 to 19 Lignin 24 to 27 Extractives 32 to 38 Due to the higher HHV of extractives, an increase in extractives content had a positive effect to the calorific value as reported by Demirbas (2002). Furthermore, G\u00C3\u00BCnther et al. (2012) summarized previous studies on the effect of lignin and extractive content on calorific value. The authors stated that extractives, lignin, cellulose, and hemicellulose have NCVs of 25.1, 27.0, 17.3, 27 and 16.2 MJ/kg, respectively. Hence, higher lignin and extractive contents result in higher calorific values. G\u00C3\u00BCnther et al. (2012) provided an example by comparing birch and Santos rosewood. Birch has a lignin content of 19.3 to 27.4%, and extractives content of 1.7% to 2.5%, while Santos rosewood has a lignin content of approximately 31.2% and extractives content of 15.6%. The author concluded that these differences in extractive and lignin content are reflected in higher figures in calorific values for Santos rosewood than birch. Calorific value is often correlated to holocellulose, lignin and extractive content. Rh\u00C3\u00A9n (2004) correlated the HHV (d.b.) to these chemical constituents with the literature values and arrived at the equation below: \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089 = 32.3 (\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0091\u00A5\u00F0\u009D\u0091\u00A1) + 24.5 (\u00F0\u009D\u0090\u00BF) + 18.6 (\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u0099\u00F0\u009D\u0091\u0099) (2-7) where HHV is the higher heating value of extracted wood in MJ/kg. Cell, L and Ext are the holocellulose, lignin and extractive content in mass percentage in dry basis. From the equation above, the HHV of extractives, lignin and holocellulose can be estimated as 32.3 MJ/kg, 24.5 MJ/kg and 18.6 MJ/kg. Telmo and Lousada (2011) provided a correlation equation (Eq. 2.2.4) in their work of relating the HHV (d.b.) of 17 wood fuels to their lignin and extractive content. \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089 = 14.3366 + 0.1228(\u00F0\u009D\u0090\u00BF) + 0.1353(\u00F0\u009D\u0090\u00B8\u00F0\u009D\u0091\u00A5\u00F0\u009D\u0091\u00A1) (2-8) The correlation coefficient (R value) of this correlation is 0.915. 28 2.3 Effect of storage on calorific values of woody biomass 2.3.1 Chemical and biological degradation of woody biomass during storage The woody biomass degrades biological and chemically during storage. As stated in the previous section, biological degradation occurs at a moisture content greater than 20% (Byrne, and Uzunovic 2005). However, the reduction was also observed for the calorific value in dry basis (d.b.) or moisture-free basis. Lehtikangas (2001) observed a roughly 1% decrease in calorific value (d.b.) for pelletized sawdust. He attributed this reduction to a decrease in the extractives content, simple sugar in particular. In his outdoor storage experiment involving stored spruce and pine, Lehtikangas observed that the lignin content increased with storage duration and associated this increase to the microbial degradation of cellulose and hemicellulose by decay fungi. The decrease in cellulose and hemicellulose increased the proportion of lignin. The microbial degradation of cellulose and hemicellulose is usually caused by the action of brown rot fungi on softwood species (Deacon 2006, Fengel and Wegener 1984), and the observations made by Lehtikangas confirm this phenomenon. Auto-oxidation of the extractives and proteins of woody biomass occur in the presence of atmospheric oxygen. In this thesis, chemical degradation of biomass is defined as degradation of the chemical components of biomass. Biological degradation as well as chemical degradation can create pre-cursors and radicals that generate compounds with higher energy potential such as quinoa, and other phenolic compounds (Brand et al. 2011); while cellulose and hemicellulose can be degraded by oxygen to form carboxylic acid, ketone and aldehydes (Fengel and Wegener 29 1984; Svedberg et al 2004). These low molecular weight compounds give off more heat per unit mass than polymeric lignin and cellulose. 2.3.2 Configurations of storage Storage configuration can be classified into two types: \u00E2\u0080\u009Copen\u00E2\u0080\u009D storage or \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage. \u00E2\u0080\u009COpen\u00E2\u0080\u009D storage allows the contact by the stored materials with atmospheric air, leading to, an increase in the moisture content. Examples of \u00E2\u0080\u009Copen\u00E2\u0080\u009D storage are the storage of forest residues on the forest floor, and storage of wood shavings in windrows or barns. Fuller (1985) found a 90% decrease in resin and terpenes in wood logs when these materials are stored outdoor for over three months. \u00E2\u0080\u009CClosed\u00E2\u0080\u009D storage usually refers to the storage of materials such as wood pellets in silos or containers, with minimum contact by the stored materials with atmospheric air (Obernberger and Thek 2010). \u00E2\u0080\u009CClosed\u00E2\u0080\u009D storage represents one extreme whereby no air is allowed to enter the storage, while open storage represents another extreme whereby materials are fully exposed to atmospheric air. In practice, wood pellets are stored in enclosed silos or covered barns to avoid damage by precipitation and humid air. 2.3.3 Storage temperatures Storage temperature affects the relative humidity of the storage environment, which in turn influences the equilibrium moisture content of the biomass. According to Oxford Dictionary of Construction, Surveying and Civil Engineering (2012), equilibrium moisture content is the point at which the moisture content of a material is in equilibrium at a given temperature and humidity. 30 As a specific example, the moisture content of wood pellets is influenced by the storage temperature and storage configuration. These environmental factors determine the moisture content of solid fuels by the means of movement in the sorption isotherm of wood. Sorption isotherm for wood describes the relationship between moisture content and the equilibrium relative humidity of a wood surface at a specified temperature (Rowell, 2012). Higher storage temperature implies higher moisture evaporation rate and higher oxidative degradation rate. From wood pellet off-gassing studies, it is known that the concentrations of off-gasses, such as carbon monoxide, methane and hydrogen, are higher at elevated atmospheric temperatures (Tumuluru et al 2009; Yazdanpanah et al. 2014). 31 Chapter 3 The Change of Calorific Values of Wood Pellets during Storage 3.1 Introduction In this chapter, the change of the calorific value of wood pellet through a six-month storage experiment is investigated. The experiment is designed to establish the relationship between the calorific value of wood pellets and these three factors: pellet type, storage configuration and storage temperature. In addition, the effect of storage on the moisture, ash and carbon content of wood pellets are measured. 3.2 Methods and materials 3.2.1 Experimental design The experiment design involves a full factorial design of three factors, which are pellet types (two categories), storage configuration (two categories) and storage temperature (three levels). The two categories of pellet types are white wood pellet and mixed wood pellets, which will be described in next paragraph. The two storage configurations are \u00E2\u0080\u009Copen\u00E2\u0080\u009D storage and \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage, which represents two extremes of ventilation. In actual storage practice, the silos or cargo holds are rarely fully sealed or fully exposed to atmospheric air. From the results of this work, the actual storage conditions, which lie between the two extremes (open and closed), can be represented. The three temperature levels are 25\u00C2\u00B0C, 35\u00C2\u00B0C and 45\u00C2\u00B0C. These temperature levels are selected because they represent the storage temperatures during the ocean transportation from British Columbia, through the tropical region of Panama Canal to Europe. In the tropical regions, the average day temperatures are up 33\u00C2\u00B0C (INEC Panama 2014). There, the storage temperature 32 in a cargo hold can reach 45 \u00C2\u00B0C during a sunny day due to heat accumulation. Multiplying the levels together, the total number of treatments is 12. Table 3-1 shows the experimental design. Table 3-1: Pellet types, storage configuration and storage temperatures Pellet type Storage temperature and configuration 25\u00C2\u00B0C 35\u00C2\u00B0C 45\u00C2\u00B0C White pellet Open Open Open Closed Closed Closed Mixed pellet Open Open Open Closed Closed Closed The two types of pellets are received from Pinnacle Renewable Energy (PRE) Inc: 10% bark pellets (\u00E2\u0080\u009Cwhite pellets\u00E2\u0080\u009D) and 40% bark pellets (\u00E2\u0080\u009Cmixed pellets\u00E2\u0080\u009D). The white pellets and mixed wood pellets used in the experiment were manufactured at the PRE Inc. pellet plants in Williams Lake and Burns Lake, BC. The pellets were four days old after production when sampled and shipped to the University of British Columbia (UBC) campus in Vancouver, BC. At the beginning of the experiment, a portion of wood pellets was poured into mason jars, which serve as the \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage configuration (no exposure to air), while another portion was spread onto an aluminum tray which serves as the \u00E2\u0080\u009Copen\u00E2\u0080\u009D storage configuration. The mason jars and trays, filled with 1 kg and 2 kg wood pellets respectively, were subjected to three temperature conditions. One set was placed on a table in the laboratory (25\u00C2\u00B12\u00C2\u00B0C throughout the test period). The other two sets were placed in two ovens with temperature set at 35 and 45oC, respectively. The wood pellets were sampled weekly in the first month, bi-weekly for the next two months, and monthly for the remaining three months for fuel composition (moisture content, ash 33 content, and carbon content) and gross calorific value (GCV) measurements. Figure 3-1 pictures the open and closed storage configurations. Figure 3-1: Open (left) and closed (right) storage configurations. 3.2.2 Procedure and apparatus Ash content and moisture content were measured immediately after sampling based on the standards NREL/TP-510-42622 (Sluiter et al. 2005) and NREL/TP-510-42621 (Sluiter et al. 2008), respectively. The ash content is expressed in dry mass basis (d.b.). The moisture content is expressed in wet mass basis (w.b.). GCV were measured as received according to EN 14918 using an oxygen bomb calorimeter (PARR model 6100). Calorific values were measured within 3 days after sampling. Before the measurements were performed on the samples, the samples were stored in a sealed plastic bag under -1oC condition in a refrigerator. NCV, expressed in MJ/kg were then calculated from GCV in MJ/kg using Eq. 2-4 from Chapter 2 of this thesis. Note that the unit MJ/kg is equivalent to GJ/MT. Figure 3-2 shows a basic oxygen bomb calorimeter. There are two types of commonly used bomb calorimeter: adiabatic and isoperibol. Zielenkiewicz and Margas (2002) described in detail the difference between the adiabatic and isoperibol calorimeters. In adiabatic calorimeter, the temperature of the water in the calorimeter can and the temperature of the bomb (also called oxygen combustion vessel) are maintained at the same temperature to prevent any heat losses 34 (Zielenkiewicz and Margas 2002). The calorific value can then be calculated directly from the temperature rise. For isoperibol calorimeter, the surrounding air jacket (thermostat) is maintained at a constant temperature while the temperature of the bomb and water in the calorimeter can rises as heat is released by the fuel combustion (EN 14918). Correction of the heat loss to the thermostat is then applied to the temperature rise, which in turn is used to determine the calorific value. Figure 3-2: A typical bomb calorimeter (adapted from Thomson-Brooks/Cole (2003)) As it was mentioned in Chapter 2, the calorimetrically measured calorific value is called the gross calorific value. The measurement of gross calorific values using PARR 6100 isoperibol oxygen bomb calorimeter involves the following general steps, which is a modification of EN 14918. 35 1. About 50 grams of wood pellet was sampled as-received and immediately kept into a sealed plastic bag. 2. A single wood pellet was withdrawn from the plastic bag and weighed in a combustion capsule. The wood pellet was selected to have a weight of 0.8\u00C2\u00B10.1 g. 3. The combustion capsule with the wood pellet was placed between two electrodes. 4. Fuse wire was attached to the electrodes and wood pellet was placed between the fuse wire. 5. The combustion vessel was assembled and was charged with 400 psig of pure oxygen (99.5%v/v). 6. 2000\u00C2\u00B10.5 milliliters of distilled or de-ionized water were added into calorimeter can. 7. The bomb was carefully lowered into the calorimeter can. 8. The ignition leads were attached and the calorimeter run was initiated. 9. After the calorimeter run, gross calorific value in as-received basis (or wet basis) was calculated automatically by PARR 6100 calorimeter. The numerical value of GCV was shown on the instrument printout. 10. The average gross calorific value was calculated from three measurements series per sample. NCV (MJ/kg) was calculated from GCV with the following simplified equation, where the oxygen and nitrogen term is removed from equation 2-4 and the unit of calorific value is converted from J/g to MJ/kg by dividing by 1000: \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089 = \u00F0\u009D\u0090\u00BA\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089 \u00E2\u0088\u0092 (0.212 \u00C3\u0097 \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB) \u00C3\u0097 (1 \u00E2\u0088\u0092 0.01 \u00C3\u0097\u00F0\u009D\u0091\u0080) \u00E2\u0088\u0092 0.0245 \u00C3\u0097 \u00F0\u009D\u0091\u0080 (3-1) 36 where M is the total moisture content in wet basis in percentage (%) XH is the hydrogen content in dry weight of the sample in percentage (%) The constant 0.212 in MJ/kg-hydrogen represents the heat of vaporization of water formed during the combustion of hydrogen content in the wood pellet less the work done on the system during the conversion from constant volume to constant pressure, whereas 0.0245 in MJ/kg-moisture represents the heat of vaporization of the moisture within the wood pellet structure. Moisture content \u00F0\u009D\u0091\u0080 used in equation 3-1 was an average of three replications for each sample. Hydrogen content \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB, that was used to calculate the NCV, are averages for each sample population of white pellet and mixed pellets, respectively. Moisture content is determined by the following steps: the sample was dried at a temperature of 105\u00C2\u00B12 \u00C2\u00B0C in forced convection oven for at least 24 hours until constant mass is achieved. The percentage moisture calculated from the loss in mass of the sample. The average moisture content was calculated from three iterations per sample (ANSI/ASAE, 2012). The ash content is determined by calculating the mass of the residue after the sample was heated in air under controlled time steps, sample weight and equipment specifications to a controlled temperature of 575 \u00C2\u00B1 20 \u00C2\u00B0C using a muffle furnace (Thermo Fisher Scientific Lindberg/Blue M Box Furnaces). The average ash content (d.b.) was calculated from three iterations per sample (Sluiter et al., 2005). To compare elemental compositions, in particular carbon content of white wood pellets and mixed wood pellets stored in two storage configurations, the elemental analysis were performed by UBC Mass Spectrometry Centre using EA 1108 Elemental Analyzer and by SGS 37 Delta Analytical Laboratory using LECO 628 Carbon, Hydrogen, Nitrogen Determinator. The ultimate analysis followed the standard procedures in CEN/TS 15104 (2005) and CEN/TS 15289 (2006). The measurements were performed in the laboratory of the Department of Chemical & Biological Engineering, UBC. The properties of stored wood pellets were compared to those of the initial wood pellets in terms of NCV, moisture content (% w.b.), ash content and carbon content. The uncertainties in measurement are expressed as confidence interval at \u00F0\u009D\u009B\u00BC = 0.05. The methods of the laboratories in UBC and SGS are compared in Table A-4. 3.2.3 Inherent uncertainties of bomb calorimeter Calibration of the calorimeter is maintained at percentage of relative standard deviation (%RSD) of lower than 0.4% as recommended by PARR Instrument Company PARR 6100 isoperibol oxygen bomb calorimeter operating manual. The percentage relative standard deviation represents the systematic error introduced by oxygen bomb calorimeter. %\u00F0\u009D\u0091\u0085\u00F0\u009D\u0091\u0086\u00F0\u009D\u0090\u00B7 =\u00F0\u009D\u0091\u0086\u00F0\u009D\u0090\u00B7\u00F0\u009D\u0091\u0080\u00F0\u009D\u0091\u0092\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u009B \u00C3\u0097 100 (3-2) Calibration runs were done on 10 one gram of standard materials, benzoic acid. The means and standard deviations of effective heat capacity \u00F0\u009D\u009C\u0080 of the 10 calibration runs were calculated in Table A-1. The calculation showed that %RSD is 0.2% which is within the precision level required by EN 14918 standard. EN 14918 required the %RSD of more than eight calibration runs to be less than 0.2%. Thus, it is safe to assume that the error introduced by the bomb calorimeter is small and can be ignored. 3.3 Results To determine the variation of the initial values of NCV, ash and moisture content, ten replications of measurements were performed on day 0. However, only three replications were 38 performed on sampling days after the initial tests. Hence, the error bars might not represent the actual situation. 3.3.1 The observed changes in moisture content and ash content and their effects on calorific value The plots in Figure 3-3 display the moisture content (MC) in wet basis (w.b.) over 162-day storage. The initial moisture content are 2.7\u00C2\u00B10.2% for white pellet and 4.9\u00C2\u00B10.8% for mixed pellet. In \u00E2\u0080\u009Copen\u00E2\u0080\u009D storage, the wood pellets quickly achieved their equilibrium MC after 20-day of storage and remained relatively unchanged for the balance of storage period. The equilibrium moisture content is dependent on air humidity and storage temperature. The equilibrium MC thus varies according to storage temperatures. MC of white pellets equilibrated at 5.9\u00C2\u00B10.2% at 25\u00C2\u00B0C, 3.6\u00C2\u00B10.1% at 35\u00C2\u00B0C, and 1.9\u00C2\u00B10.2%. Mixed pellets have equilibrium moisture content of 6.3\u00C2\u00B10.4% at 25\u00C2\u00B0C, 4.0\u00C2\u00B10.1% at 35\u00C2\u00B0C and 2.1\u00C2\u00B10.2% at 45\u00C2\u00B0C. In \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage, the wood pellets have little change in its moisture content compared to initial condition. Moisture content of white pellets arrived at 2.6\u00C2\u00B10.1% regardless of storage temperature. For mixed pellets, their moisture content spreads out slightly, which can be explained by the wider variation in mixed pellet moisture content at day 0. This initial wider variation is shown on Figure 3-3 (d) as a wider error bar on day 0, where minimum and maximum of initial moisture contents of mixed pellets are 3.3% and 6.5%. At 25\u00C2\u00B0C, moisture content stabilized at 5.9\u00C2\u00B10.1%; the equilibrium MCs at 35\u00C2\u00B0C and 45\u00C2\u00B0C are 4.6\u00C2\u00B10.3% and 4.1\u00C2\u00B10.3%, which have little difference from initial MC of mixed pellet of 4.9\u00C2\u00B10.8%. 39 Although pellets are not placed in an open jar in this experiment, pellets, which are placed in an open jar are expected to reach equilibrium moisture content later than those placed on a tray. The plots in Figure 3-4 display the trend of ash content in dry basis (d.b.) over 180-day storage. The initial ash content was 0.5\u00C2\u00B10.1% for white pellet and 1.2\u00C2\u00B10.2% for mixed pellet. It is observed that the measured ash content has fluctuated randomly around the initial ash content regardless of the storage condition. It appears that the degree of randomness for mixed pellets is higher than the white pellet. As mixed pellets is made from a larger proportion of bark than white pellets, this results in a higher degree of inhomogeneity in the raw material of mixed pellets. Bark also contains more ash due to higher mineral content, causing the mixed pellets to have higher ash content than white pellets (Filbakk et al. 2011). Because the ash content remained relatively constant during the experiment, it is useful to explain the changes in NCV. The plots in Figure 3-5 display the trends of NCVs for four conditions: (a) white \u00E2\u0080\u009Copen\u00E2\u0080\u009D, (b) white \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D, (c) mixed \u00E2\u0080\u009Copen\u00E2\u0080\u009D, and (d) mixed \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D over 180-day storage. The initial NCVs are 18.2\u00C2\u00B10.1 MJ kg-1 for white pellets and 17.9\u00C2\u00B10.2 MJ kg-1 for mixed pellets. It is observed that NCV has generally increased over the 6-month storage regardless of the storage conditions, except for \u00E2\u0080\u009Copen\u00E2\u0080\u009D storage at 25\u00C2\u00B0C. The percentage change, %\u00F0\u009D\u0091\u0090\u00E2\u0084\u008E\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u0094\u00F0\u009D\u0091\u0092 in NCV is calculated using the formula below: %\u00F0\u009D\u0091\u0090\u00E2\u0084\u008E\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u0094\u00F0\u009D\u0091\u0092 =\u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u0093\u00F0\u009D\u0091\u0096\u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u0099 \u00E2\u0088\u0092 \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u0096\u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u0096\u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u0096\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u0099\u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u0096\u00F0\u009D\u0091\u009B\u00F0\u009D\u0091\u0096\u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u0096\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u0099\u00C3\u0097 100 (3-1) For white pellets, At \u00E2\u0080\u009Copen\u00E2\u0080\u009D storage, the NCV at 25\u00C2\u00B0C dropped by 2.0%, while NCV at 35\u00C2\u00B0C and at 45\u00C2\u00B0C increased by 0.5% and 2.0% respectively. The drop in NCV of the sample at 40 25\u00C2\u00B0C was observed in parallel to the increase in moisture content in the same sample. The decrease in moisture content of samples at 35\u00C2\u00B0C and at 45\u00C2\u00B0C is observed in conjunction of the increase in NCV. All wood pellets under \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage experienced a 2% increase in average NCVs while the moisture content remained constant. For mixed pellets, wood pellets in \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage have slightly varying NCVs. This variation in NCV is due to the same variation in moisture content. \u00E2\u0080\u009COpen\u00E2\u0080\u009D storage had a 1.0% decrease at 25\u00C2\u00B0C, a 3.0% increase at 35\u00C2\u00B0C and 4.0% increase at 45\u00C2\u00B0C. By comparison, NCV for mixed pellets under \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage had an increase of 1.0% at 25\u00C2\u00B0C, an increase of 2.0% at 35\u00C2\u00B0C, and an increase of 3.0% at 45\u00C2\u00B0C. Samples of white \u00E2\u0080\u009Cclose\u00E2\u0080\u009D and mixed \u00E2\u0080\u009Cclose\u00E2\u0080\u009D at 45\u00C2\u00B0C are sent to SGS Delta Analytical Laboratory for verification of the NCV measurement at day 42 and day 84. These verification data are presented in Tables A-3 in Appendix A.1. The values are similar to the values obtained at UBC with only 0.5% difference for NCV. 3.3.2 The carbon content after storage Samples were sent for CHN composition analysis at four storage period: day 0, day 10, day 40 and day 80. Only pellets stored at 45\u00C2\u00B0C in \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage was measured on day 80. Although hydrogen and nitrogen contents are also measured, their effect on NCV cannot be confirmed due their small measured values. On average, the hydrogen content is 6.12% (d.b.) for white pellets and 6.08% (d.b.) for mixed pellets. The nitrogen content of the wood pellets are below the detection level of 0.05%. Carbon constitutes of around 50% (d.b.) of a wood pellet\u00E2\u0080\u0099s mass. The positive effect of carbon on NCV is more of a certainty. For example, Casal et al. (2010) observed a drop in calorific value in as received basis over 12-month open storage of 41 wood chips and associated the drop to a decrease in carbon content. Hence, only the carbon content is discussed. Over the storage period, carbon contents of wood pellets do not appear to follow any particular trend, as shown in Figures 3-6. For white pellets in open storage, after 10 days in storage, the carbon content appears to increase by 1.5% for pellets stored at 35\u00C2\u00B0C, while carbon content remained the almost same for pellets stored at 25\u00C2\u00B0C and 45\u00C2\u00B0C. The carbon contents of all the white pellets in open storage returned to their initial average value of 50.4% after 35 days in storage. As for white pellets kept in \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage, the carbon content stayed within \u00C2\u00B10.5% from their initial values (49.9% to 50.9%), with the highest value of 50.9% observed in pellets stored in 45\u00C2\u00B0C on day 84 and the lowest value of 49.9% observed in pellets stored in 35\u00C2\u00B0C on day 40. The carbon content of mixed pellets has a general trend of increasing or staying the same in subsequent measurements after day 0. In open storage, the carbon contents of mixed pellets at 35\u00C2\u00B0C and 45\u00C2\u00B0C remained the same after the first 10 days and then increased by about 1.1% on day 40 from 50.4% to 51.5%. The carbon content of pellets stored at 25\u00C2\u00B0C did not change significantly over storage. In \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage, the carbon content of all pellets remained the same on day 10. On day 40, pellets at 25\u00C2\u00B0C had a 0.5% decrease in carbon content from 50.4% to 49.9%; pellets at 35\u00C2\u00B0C had an increase of 1.0% from 50.4% to 51.4%; pellets at 45\u00C2\u00B0C had a slight increase from 50.4% to 50.8% on day 40 and further increased to 51.35% on day 80. Because there is no apparent relationship between NCV and carbon content observed in Figure 3-6, correlation coefficients are calculated to investigate the overall degree of correlation between carbon content and NCV. All the carbon content and NCV are grouped together 42 regardless of their storage configurations and storage temperatures. The correlation coefficient of 0.086 is obtained. This low correlation coefficient indicates the lack of overall relationship between carbon content and NCV. The group of data used to calculate the correlation coefficient is given in Table A-2. 3.3.3 ANOVA analysis to determine the effects of pellet type, storage configuration and storage temperature on NCV One-way ANOVA was used to investigate the effects of pellet type on NCV; Two-way ANOVA was used to investigate the effect of storage configuration and storage temperature on NCV. The one-way analysis of variance (Table 3-2) showed that the NCVs are significantly different between white pellets and mixed pellets, although the numerical difference is small. The initial NCVs of white pellets and mixed pellets are 17.9 MJ/kg and 18.1 MJ/kg respectively. Although the numerical difference is small, the mixed pellets have higher moisture content and higher ash content than white pellet. The higher moisture and ash contents are caused by the higher bark content in mixed pellets. The initial moisture content and ash content of mixed pellets was (4.9% w.b., 1.2% d.b.), whereas white pellets have an initial moisture and ash content of (2.7% w.b., 0.5% d.b.). Table 3-2: One-way analysis of variance on the NCV of two wood pellet types Source of Variation SS df MS F p-value Among pellet type 1.723 1 1.723 16.99 4.4\u00C3\u009710-5** Within Groups 49.282 486 0.101 43 Total 51.006 487 **P-value is significant at a 0.01 level. The two-way analysis of variance (Table 3-3) showed that the NCV of wood pellets significantly differ (P < 0.01) between \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D and \u00E2\u0080\u009Copen\u00E2\u0080\u009D storage and among the three temperature levels: 25\u00C2\u00B0C, 35\u00C2\u00B0C, and 45\u00C2\u00B0C. Table 3-3: Two-way analysis of variance on NCV of two storage configuration (\u00E2\u0080\u009COpen\u00E2\u0080\u009D or \u00E2\u0080\u009CClose\u00E2\u0080\u009D and three storage temperatures (25\u00C2\u00B0C, 35\u00C2\u00B0C, 45\u00C2\u00B0C) Source of Variation SS df MS F P-value Among storage configuration 1.325 1 1.325 34.43 < 0.0001 Among storage temperature 12.986 2 6.492 168.63 < 0.0001 Among storage temperatures within a storage configuration 6.7077 2 3.353 87.112 < 0.0001 Within storage temperatures 12.705 330 0.031 Total 33.724 335 **P-value is significant at a 0.01 level. 3.4 Conclusion It can be concluded that the NCV of the two storage configurations (\u00E2\u0080\u009Copen\u00E2\u0080\u009D and \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D) are significantly different. The difference between the two storage configurations can be explained by the change in moisture content due to exposure to atmospheric air. \u00E2\u0080\u009COpen\u00E2\u0080\u009D storage exposed the wood pellets to atmospheric air. \u00E2\u0080\u009CClosed\u00E2\u0080\u009D storage did not. Exposure to atmospheric air allowed the exchange in water molecules and led the change in moisture content of wood pellets. Higher storage temperature (35 and 45\u00C2\u00B0C) exposes wood pellets to atmospheric air with higher capacity to hold water. This allows the evaporation of water from water vapor. This loss of water decreases the moisture content of pellets and hence increases the NCV. At lower 44 temperature (25\u00C2\u00B0C), the water vapor in the atmospheric air condensed on the wood pellet, increasing the moisture content. At the same time, the relative humidity of atmospheric air at 35 and 45\u00C2\u00B0C are lower (20%RH and 18%RH), compared to 30%RH at 25\u00C2\u00B0C. Rather than storage temperature, the higher ambient relative humidity at 25\u00C2\u00B0C could have caused the difference in moisture content. It is widely known that moisture content has negative effect on calorific value of biomass. Higher moisture content results in lower calorific value. In \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage, however, there is a lack of change in moisture content to justify the increase in NCV. The moisture content of in \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage stayed within \u00C2\u00B10.5% of the initial value for the white pellets and \u00C2\u00B11.0% for mixed pellets. The correlation level between carbon content and net calorific value is very low (R = 0.086). Since the results from carbon content to explain is non-conclusive, it is hypothesized that the formation of higher energy compounds, such as ketone and aldehyde or off-gasses such as methane, carbon monoxide and hydrogen through the fermentation of cellulose and hemicellulose as well as the auto-oxidation of extractive compounds in wood will result in increase on the NCV of the wood pellets (Brand et al., 2011). The result here will be further elaborated in the discussion of the multivariable linear regression analysis. 45 Figure 3-3: Wood pellets moisture content on as-received basis (% w.b.) over 180-day of storage for (a) white \u00E2\u0080\u009Copen\u00E2\u0080\u009D, (b) white \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D, (c) mixed \u00E2\u0080\u009Copen\u00E2\u0080\u009D, and (d) mixed \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D. 46 Figure 3-4: Wood pellet ash content on dry basis (% d.b.) over 180-day of storage for (a) white \u00E2\u0080\u009Copen\u00E2\u0080\u009D, (b) white \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D, (c) mixed \u00E2\u0080\u009Copen\u00E2\u0080\u009D, and (d) mixed \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D. 47 Figure 3-5: Wood pellets net calorific values over 180-day of storage for (a) white \u00E2\u0080\u009Copen\u00E2\u0080\u009D, (b) white \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D, (c) mixed \u00E2\u0080\u009Copen\u00E2\u0080\u009D, and (d) mixed \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D. 48 Figure 3-6: Wood pellets carbon content over 80-day of storage for (a) white \u00E2\u0080\u009Copen\u00E2\u0080\u009D, (b) white \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D, (c) mixed \u00E2\u0080\u009Copen\u00E2\u0080\u009D, and (d) mixed \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D. 49 Chapter 4 Interpreting and Modelling the Change of Calorific Values 4.1 Introduction This chapter show the attempts to develop a multivariable linear regression model (MLR) and a single variable model. The objective of developing a multivariable linear regression model (MLR) is to correlate the change in calorific values with the physical properties of wood pellets and the external storage conditions. On another hand, the single variable model can be used as a tool to determine the storage duration to attain a certain increase in initial calorific value with the temperature of storage. 4.2 Methods 4.2.1 Multivariable linear regression model (MLR) The MLR model is implemented using the function \u00E2\u0080\u009Clm( )\u00E2\u0080\u009D in the R programming language version 3.0.0 (R Development Core Team, 2008). The assumptions that were made to develop this model are: (1) NCV is linearly related to the parameters; and (2) Storage duration is the total storage period experienced by the wood pellets when they are sampled. The data for Figs. 3-3, 3-4, and 3-5 from Chapter 3 were analyzed with MLR analysis (\u00CE\u00B1 = 0.01). The NCV of the wood pellets were correlated with storage duration, storage temperature (three levels: 25, 35 and 45\u00C2\u00B0C), and storage configurations (\u00E2\u0080\u009Copen\u00E2\u0080\u009D or \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D), moisture content (w.b.), ash content (d.b.) and two pellet types (white or mixed). 50 Since confounding variables can affect the results of the multivariable linear regression, the following systematic steps are used: 1. First, all the data from both pellet types and both storage configurations are fitted with a multivariable linear regression to obtain an overall picture of the relationships between the variables 2. Secondly, the data are grouped into two pellet types as well as two storage configurations, which are: a. white and mixed pellets at open storage, b. white and mixed pellets at closed storage, c. white pellets at both open and closed storage, and d. mixed pellets at both open and closed storage. 3. Finally, the data were grouped into individual cases: a. white pellets at open storage, b. white pellets at closed storage, c. mixed pellets at open storage, and d. mixed pellets at closed storage. This grouping analysis resulted in nine cases. 4.2.2 Single variable linear model The steps to arrive at the single variable linear model are as follow: 1. The overall data from both white and mixed pellets are grouped into a. open storage scenario, 51 b. closed storage scenario. 2. The data from each group are split into the three temperature levels, which are 25\u00C2\u00B0C, 35\u00C2\u00B0C and 45\u00C2\u00B0C. 3. Data from each temperature level are fitted to a linear regression with NCV as its response variable and duration of storage as its input variable. 4. The number of days to achieve a certain percent increase in calorific value for individual temperature level is calculated from the slope and intercept of the NCV versus duration of storage linear model. 5. The linear regression is used to fit the three temperature levels to the three total duration of storage in each storage configuration. As a result, the single variable model takes the following form: \u00F0\u009D\u0090\u00B7 = \u00F0\u009D\u0091\u0090 +\u00F0\u009D\u0091\u009A \u00E2\u0088\u0097 \u00F0\u009D\u0091\u0087 (4-1) where \u00F0\u009D\u0090\u00B7 is the total duration of storage in days to obtain the percent increase of initial calorific value, in this case, 3%. \u00F0\u009D\u0091\u0087 is the storage temperature in \u00C2\u00B0C. \u00F0\u009D\u0091\u0090 is the intercept. \u00F0\u009D\u0091\u009A is the slope. 4.3 Result 4.3.1 Comparing the individual cases of MLR The results of the MLR analysis are summarized in Table 4-1 in the end of the chapter. In the first stage, the overall linear regression is fitted to the six variables with NCV as the response variable. The initial values of all variables are taken as constants. Referring to the \u00E2\u0080\u009Cfirst stage of analysis\u00E2\u0080\u009D in Table 4-1, NCV showed a reasonably good correlation (R2 = 0.73) with all the variables (p < 0.0001), except for ash content and the type of pellets. The sign of a coefficient in the MLR equation would indicate that a factor is either positively or negatively 52 correlated to the NCV. Thus, moisture and ash content are negatively correlated with NCV. This decrease in NCV with moisture content and ash contents is understood and is shown earlier in this thesis (Th\u00C3\u00B6rnqvist 1987; Telmo and Lousada 2011; Demirbas 2002). Monti et al. (2008), in particular, stated that 1% increase in ash content results in a 0.2 MJ kg-1 decrease in calorific value. However, a decrease in NCV with exposure to air and increase in NCV with duration of storage are not well explained. The lack of overall correlation between ash content and NCV (p = 0.023) is understandable as the ash content (0.0-2.0%, d.b.) did not change significantly throughout the experiment. In the second stage of linear regression analysis in Table 4-1, when only data (aggregating both white and mixed pellets) relevant to \u00E2\u0080\u009Copen storage\u00E2\u0080\u009D were involved, the R2 value improved significantly to 0.90 from a value of 0.72. However, the effect of storage duration has diminished with p = 0.10. In contrast, when only data relevant to \u00E2\u0080\u009Cclosed storage\u00E2\u0080\u009D were involved, the R2 value became lower (0.62) when compared to R2 of 0.72 for the overall regression. This could be due to the lack of differences in NCV, moisture content and ash content at the three temperature levels under closed storage conditions. While fitting linear regression to data for white pellets alone (aggregating both \u00E2\u0080\u009Copen\u00E2\u0080\u009D and \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D configurations), the R2 value is 0.79. By comparison, when the linear regression was fitted to data for mixed pellets alone (again, aggregating both configurations), the R2 was 0.69. Based on the differences in the R2 values, it is evident that the linear regression model fits the white pellets better than the mixed pellets. The lesser degree of correlation between NCV and the input variables for mixed pellets is consistent with the plots shown in Chapter 3, whereby the observed NCV for the white pellets have a smaller change in value compared to that for the mixed pellets. With p value larger than 0.01, ash content did not have a statistically significant 53 role in influencing the NCV of both mixed and white wood in stage one and stage two of the MLR analysis. The third stage of linear regression analysis involves the fitting the linear regression model to individual cases, without segregating the pellet types or storage configurations. The individual cases are (white pellets, open storage), (white pellets, closed storage), (mixed pellets, open storage) and (mixed pellets, closed storage). Here, the number of observations used for linear regression is 36. Again, shown in Table 4-1, the data from open storage for both white and mixed pellets appear to fit the linear regression model better than the data from closed storage. In open storage, the storage temperature influenced the NCV significantly (p < 0.01). In closed storage, however, the storage temperature, storage configuration and moisture content did not have significant effect on NCV, which is also evident from inspecting Figures 3-5 (b) and (d). From overall relationship, it was shown that the increase of NCV over the duration of storage was in part due to the decrease of moisture content over storage. Graphically, the decrease in moisture contents in \u00E2\u0080\u009Copen\u00E2\u0080\u009D storage in Chapter 3 Fig. 3-2 (a) and (c) caused increase in the NCV in Fig. 3-4 (a) and (c) or vice versa. In \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage, however, a different relationship was observed between moisture content and NCV. The white pellets had little change in moisture content but still showed an increase in calorific value, whereas the mixed pellets showed an increase in NCV together with an increase in moisture content. Hence, it is hypothesized that time-dependent reactions, such as microbial degradation and auto-oxidation of extractives, hemicellulose and cellulose had occurred during the duration of storage. The microbial degradation of extractives, hemicellulose and cellulose resulted in the formation of compounds with higher energetic potential, such as fatty acids, and alcohols (Brand et al. 2011). Auto-oxidation generates methane and hydrogen in 54 the form of off-gassing (Yazdanpanah et al. 2014). These highest energetic compounds might increase the NCV. For the microbial activities to occur on woody materials, a sufficiently high moisture content is required. The minimum required moisture content for mold or fungi to grow has been reported as 20% (w.b.) (Bryne and Uzunoic 2005). Since the highest moisture content of the wood pellet is 8%, fermentation through microbial enzymatic reaction is not likely to happen. With microbial activities ruled out, the auto-oxidation of extractives and lignin compounds through free radicals was the reaction that generated the higher energetic potential compounds. It might be further extended that the longer duration of storage allows more compounds of higher energetic potential to form through auto-oxidation reaction, resulting in the increase of NCV. Further research has to be conducted to confirm the formation of higher energetic potential compounds over storage. The change of NCV is most sensitive to the change in storage temperature and the moisture content. It can be explained partly because the storage temperature is strongly negatively related to the moisture content. Duration of storage has marginal positive effect on NCV whereas ash content has small effect on the NCV of the pellet. 4.3.2 The application of the single variable equation Although the linear regression model for open storage has a decent predictive ability, it involves many variables. A simpler model is important for practical purposes. Hence, a model to relate the storage temperature to the duration of storage corresponding to a 3% increase from initial calorific value was developed. A 3% increase in initial calorific value was taken because the magnitude of 3% is within the reasonable range of increase of calorific value in the observed data. 55 From the exploratory plots in chapter 3 and linear regression analyses done earlier in this Chapter, it is apparent that the storage configuration has the most confounding impact on the calorific value. The data from open storage and closed storage respectively are used to develop this model. The simple model clarifies this effect. As seen in Table 4-3, the slope of the closed storage model is almost 300 times smaller compared to its intercept. This large difference in magnitude between slope and intercept implies that storage temperature effect is not significant in closed storage. On the other hand, the slope of open storage is a significant portion of its intercept, at 2% of its intercept. In fact, for open storage, the critical value of storage temperature \u00F0\u009D\u0091\u0087, where the duration of storage is zero, is 54 \u00C2\u00B0C. At this temperature, a percentage increase in calorific value can be achieved in zero days. This analysis result is obviously faulty and cannot be taken at face value. However, the effect of storage temperature is definitely important in an open storage; higher temperature dries and removes the moisture from the pellets, thereby resulting in higher NCV. Single-variable linear models have large standard error for both the slope of T and the intercept. Hence this equation is not deterministic and should only be used as a reference. The equation can be applied to estimate the approximate number of days for wood pellets to achieve a certain percentage increase of calorific value during storage and transportation. Below are the steps to illustrate the application of this equation. The steps are: 1. Record an average daily temperature (each day is represented with one average temperature) basis for the period of time you desire. 56 2. Use the average daily temperature, \u00F0\u009D\u0091\u0087 to calculate the numbers of days, \u00F0\u009D\u0090\u00B7 from equation 4-1. 3. Calculate the reciprocal of the number of days, 1\u00F0\u009D\u0090\u00B7. 4. Calculate the cumulative sum of 1\u00F0\u009D\u0090\u00B7 for each daily temperature during storage. 5. The final net calorific value \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u0093 is estimated from initial calorific value \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u0096 as below: \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u0093 = \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u0096 \u00C3\u0097 [1 + 0.03 \u00C3\u0097 (\u00E2\u0088\u00911\u00F0\u009D\u0090\u00B7)] (4-2) where \u00E2\u0088\u00911\u00F0\u009D\u0090\u00B7 is the cumulative sum of 1\u00F0\u009D\u0090\u00B7. Figure 4-1 illustrates an example of the application of the two single-variable linear models for two cases of open and closed storages in table 4-3 for 365-days storage of biomass. 57 Figure 4-1: The cumulative sum of fraction of days approaching the 1.03 times (3% increase in) initial calorific value with respect to the duration of storage (days), where (a) gives the data for open storage and (b) gives the data for closed storage. 58 In the Appendices, tables A-7 and A-8 show the record of daily temperatures (step 1) of an unnamed ocean vessel, the determination of the numbers of days required for calorific value to increase to 1.03 times the initial value (step 2), and the reciprocal of number of days (step 3), and the cumulative sum of the reciprocal (step 4). Taking an example from table A-7 (open storage) on day 1, the average daily temperature was 7.3\u00C2\u00B0C. The number of days required for calorific value to increase to 1.03 times its initial value was calculated as 512 days, using equation 4-2. The reciprocal of the number of days was 0.00042 1\u00F0\u009D\u0090\u00B7. This calculation was repeated for 365-days of storage. The cumulative sum of the reciprocal of number of days was calculated for every consecutive day. For open storage, the final cumulative sum \u00E2\u0088\u00911\u00F0\u009D\u0090\u00B7 was 0.220. Thus, applying step 5 using equation 4.2, the final net calorific value \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u0093 after 365-day storage was: \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u0096 \u00C3\u0097 [1 + (0.03 \u00C3\u0097 0.220)] = 1.007 \u00C3\u0097 \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u0096 (4-3) This indicated that the NCV value increases by 0.7% over the 365-days open storage. The same procedure can be applied to closed storage linear model, which gives an increase of 1.7% over 365-days storage. 4.4 Conclusion The analysis using multivariable analysis has led to the conclusion that the NCV of both types of wood pellets at open storage are best predicted using the six variables used in this analysis. The NCV of wood pellets at closed storage cannot be predicted adequately through linear regression using the six variables. The well-fitness of the linear regression is ranked in Table 4-2. 59 On another hand, the simple models, although only marginally accurate, give a further insight about the two storage configurations. It shows that the closed storage, without heating, is a better tool in maintaining the calorific value of wood pellet, than an open heated storage. This reaffirmed the observations from the NCV versus duration of storage figures in Chapter 3. 60 Table 4-1: Summary of the squared correlation coefficients and model coefficients for all three stages of the multivariable linear regression (MLR) models R2 ao a1 a2 a3 a4 a5 a6 First stage of analysis (n = 146; \u00CE\u00B1 = 0.01) Overall relationship - both pellet types (white and mixed) and both storage configurations (open and closed) 0.73 18.59* 0.0012* 0.0104* -0.1025** -0.1366* -0.1117 -0.1119 Second stage of analysis (n = 73; \u00CE\u00B1 = 0.01) white and mixed pellets; open storage 0.90 17.27* 0.0005 0.0374* n/a -0.0397 -0.1582* -0.1694** white and mixed pellets; closed storage 0.62 18.36* 0.0014* 0.0063 n/a -0.0530 -0.1737 0.0893 white pellets; open and closed storage 0.79 18.74* 0.0013** 0.0065 -0.1185** -0.1622* -0.1643 n/a mixed pellets; open and closed storage 0.69 18.01* 0.0009 0.0194* -0.0137 -0.0827* -0.1069 n/a Third stage of analysis (n = 36; \u00CE\u00B1 = 0.01) white pellets; open storage 0.91 17.20* 0.0011** 0.0339* n/a -0.0485 -0.1189 n/a white pellets; closed storage 0.52 18.46* 0.0015** 0.0015 n/a 0.0556 -0.4318** n/a mixed pellets; open storage 0.90 17.43* -0.0004 0.0393* n/a -0.0484 -0.2581** n/a mixed pellets; closed storage 0.48 17.95* 0.0012 0.0129** n/a -0.0332 -0.0952 n/a * The p-value (< 0.0001) indicates statistical significance of the parameter associated with this coefficient at 0.01 level ** The p-value (0.01-0.0001) indicates statistical significance of the parameter associated with this coefficient at 0.01 level NCV = ao + a1X1 + a2X2 + a3X3 + a4X4 + a5X5 + a6X6 where X1, X2, X 3, X4, X5, X6 denote the parameters storage duration, storage temperature, storage configuration, moisture content, ash content and pellet type, respectively 61 Table 4-2: Summary of the R2 values of all nine cases analyzed using multivariable linear regression. The table ranks the R2 from the highest to the lowest. Rank Cases R2 value 1 white at open 0.91 2 both white and mixed at both open 0.90 3 mixed open 0.90 4 white at both open and closed 0.79 5 both white and mixed at both open and closed 0.73 6 mixed both open and closed 0.69 7 both white and mixed at both closed 0.62 8 white closed 0.52 9 mixed closed 0.48 Table 4-3: Two single-variable linear models for open and closed storage correlate total duration of storage to obtain 3% increase from initial calorific value in days, to the storage temperature in \u00C2\u00B0C for each storage configuration. The data from both white and mixed pellets are used to form this linear model. Storage Configuration Intercept, c Standard Error of Intercept Slope, m Standard Error of Slope R2 value Open 2770 1834 -51 51 0.50 Closed 586 2206 2 61 0.00 \u00F0\u009D\u0090\u00B7 = \u00F0\u009D\u0091\u0090 +\u00F0\u009D\u0091\u009A \u00E2\u0088\u0097 \u00F0\u009D\u0091\u0087 where \u00F0\u009D\u0090\u00B7 is the total duration of storage in days to obtain the percent increase of initial calorific value, in this case, 3%. \u00F0\u009D\u0091\u0087 is the storage temperature in \u00C2\u00B0C. \u00F0\u009D\u0091\u0090 is the intercept. \u00F0\u009D\u0091\u009A is the slope. 62 Chapter 5 Conclusions and Future Work 5.1 Conclusions A laboratory-scale storage experiment was performed to determine the change of net calorific value (NCV) of two types of wood pellets (\u00E2\u0080\u009Cwhite\u00E2\u0080\u009D and \u00E2\u0080\u009Cmixed\u00E2\u0080\u009D) over six-month period at two storage configurations (\u00E2\u0080\u009Copen\u00E2\u0080\u009D and \u00E2\u0080\u009Cclosed) with three storage temperature (25, 35, 45\u00C2\u00B0C) and to establish the relationship between the calorific value of wood pellets and the storage environment as well as the key physical properties of the pellets. After six-month storage, NCV of wood pellets that are stored in \u00E2\u0080\u009Cclosed environment\u00E2\u0080\u009D increased by 2.0%, regardless of storage temperatures. And the NCV of wood pellets that are stored in \u00E2\u0080\u009Copen environment\u00E2\u0080\u009D was found to be dependent on temperature (2% decrease, 0.5% increase, and 2.0% increase) at (25, 35, and 45\u00C2\u00B0C) respectively. Since the changes of NCV are within 5% of their initial values, the uncertainties of measurement of NCV must be small to justify the change. The uncertainties or errors of calorific value measurements were determined in Chapter 3 as the percentage of relative standard deviation (%RSD). The inherent error is 0.2%. Therefore, any changes of more than 0.2%, which is the case in this thesis, are likely to be real changes. For open storage, higher storage temperature (35 and 45\u00C2\u00B0C) or low ambient relative humidity (< 30%) increased the NCV of wood pellet as the moisture content decreased through the evaporation. An increase in moisture content is observed at lower storage temperature (25\u00C2\u00B0C) or high ambient relative humidity (>30%). This work reaffirms the well-known knowledge that an increase in the moisture content of woody materials will reduce the NCV. 63 For closed storage, however, NCV increased when the moisture content did not change appreciably. This increase in calorific value was potentially caused by re-adsorption of higher energy potential compounds, such as low molecular-weight phenolic, ketones and aldehydes, or off-gasses, such as carbon monoxide (CO), methane (CH4) and hydrogen (H2). The ash content has varied within \u00C2\u00B10.5% from their initial ash content during the storage experiment. This small variation did not correlate with the changes of NCV over the storage experiment. The carbon content, on another hand, did not showed any conclusive trend. The overall correlation coefficient between NCV and carbon content is 0.086, which indicates a lack of correlation. The storage of wood pellets for more than 3 months is beneficial in terms of the NCV if the wood pellets are stored under a relatively \u00E2\u0080\u009Cclosed environment\u00E2\u0080\u009D condition and at an elevated temperature. The results of this project provide an extra benefit of the common practice to seal the ocean cargo or railcars immediately after loading the wood pellets. Producers and exporters of wood pellets in British Columbia can now be assured that the wood pellets will at least retain its original calorific value (if not higher) when proper procedures are implemented to seal the wood pellet railcars and ocean cargos. To predict the change in NCV, a multivariate linear regression (MLR) model between the NCV and six parameters (storage \u00E2\u0080\u0093 duration, temperature, and configuration; wood pellets - type, moisture content, and ash content) was developed. This MLR showed that storage duration has a significant impact on the NCV of pellets, with overall p-value less than 0.01. The best fit (R2 = 0.90) is obtained for the data group of \u00E2\u0080\u009Copen\u00E2\u0080\u009D storage scenario mainly because the change of NCV can be correlated to moisture content and storage temperature. As for \u00E2\u0080\u009Cclosed\u00E2\u0080\u009D storage, the 64 NCV increased despite the observation that the moisture content remained the same over the experiment period. As a result, its MLR has a lower R2 of 0.62. Furthermore, a simple single-variable linear model was developed to estimate the number of days to achieve 3% increase from the initial NCV as a function of storage temperature. The single-variable linear model can be used in a quick determination of the percent increase in NCV over the transportation of wood pellets. Comparing the initial quality of the two types of pellets, the initial moisture and ash content are 2.7% (w.b.) and 0.5% (d.b.) for white pellets, and 4.9% (w.b.) and 1.2% (d.b.) for mixed pellets. Despite the fact that mixed pellets have higher ash and moisture content than white pellets, they have similar NCVs of 18 MJ/kg. One can thus conclude the higher bark content of mixed pellets had increased their dry ash free calorific value above that of white pellets. Therefore, the resulting NCV of white and mixed pellets are higher. A comparison table of these two types of pellets are shown in Table 5-1. On the economics perspective, the potential economic gain of the increase in calorific value of pellets is large even for 1% increase due to the size of wood pellet production. The year 2013 total production amount in British Columbia is 1 million metric tonnes (MT). Given a net calorific value of 18 GJ/MT and energy price of $10/GJ, the annual revenue of the British Columbia wood pellet industry is $180 million. A 1% increase in calorific value is translated to 1.8 million dollar extra in revenue. From the safety point of view, however, self-heating and off-gassing concerns must also be taken into consideration. Proper ventilation must be performed before unloading the wood pellets to remove the toxic off-gases. The wood pellet pile height should conform to below the 65 self-ignition limit to avoid spontaneous combustion due to self-heating. One suggestion is to partition the ship cargo into smaller compartments to reduce the pile height and hence reduce such risks. 5.2 Future work Future work and recommendations are suggested below to further this study: \u00EF\u0082\u00B7 The identification of the compounds with high energy potential should be done in order to delineate the mechanism that causes the rise in calorific value. \u00EF\u0082\u00B7 Field tests should be performed, before and after a shipment of wood pellets, to confirm the observations made in the laboratory. \u00EF\u0082\u00B7 More iterations of carbon content analysis should be made to get a more thorough understanding of the change in carbon content during storage. \u00EF\u0082\u00B7 The change in the total extractives content (the sum of water and organic solvent extractives contents) over time should also be measured. Table 5-1: Comparing the initial calorific values of white pellets and mixed pellets Pellet Type White Mixed Bark Content % w.b. 10 40 Moisture content % w.b. 2.7 4.9 Ash content % d.b. 0.5 1.2 Net calorific value, NCV MJ/kg w.b. 17.9 18.1 Lower heating value, LHV MJ/kg d.b. 18.5 19.2 Lower heating value, LHV MJ/kg dry ash free (daf) basis 18.6 19.4 66 References Afzal, M., S. Bendane A. S. and Mahmood W. 2010. Storage of Comminuted and Un-comminuted Forest Biomass and Its Effect on Fuel Quality. BioReserouces 5(1): 55-59. ASTM International. 2012. D5865-12: Standard Test Method for Gross Calorific Value of Coal and Coke. Atkins, T. and M. Escudier. 2013. A Dictionary of Mechanical Engineering. Oxford University Press. Babcock and Wilcox Company. 1955. Chapter 9: Principles of Combustion. In Steam, its generation and use, 4-3-4-12. New York, USA: Babcock and Wilcox Company. Baker, A. 1982. Wood fuel properties and fuel products from woods. In Proc. Fuel Wood Management and Utilization Seminar, 14-25. Basset, V, J.S., Lee. 2013. Personal communication during an internship. Be\u00C3\u00A9r, J. 2005. High Efficiency Electric Power Generation; The Environmental Role. Available at: http://mitei.mit.edu/system/files/beer-combustion.pdf. Accessed 3/19 2013. Brand, M. A., Bolzon de Mu\u00C3\u00B1iz, Graciela In\u00C3\u00AAs, W. F. Quirino and J. O. Brito. 2011. Storage as a tool to improve wood fuel quality. Biomass and Bioenergy 35(7): 2581-2588. Briggs, D. 1994. Chapter 9: Energy. In Forest Products Measurements and Conversion Factors, 105-110. ed. L. Charbonneau, Seattle, WA: College of Forest Resources, University of Washington. Byrne, T., and Uzunovic, A. 2005. Addressing marketpalce durability issues with post-mountain pine beetle lodgepole pine - A compilation of three reports. Forintek Canada Corp. (No. 25). Casal, M. D., M. V. Gil, C. Pevida, F. Rubiera and J. J. Pis. 2010. Influence of storage time on the quality and combustion behaviour of pine woodchips. Energy 35(7): 3066-3071. Cassidy, P.; S. Ashton. 2007. Ash Content. Pages 201\u00E2\u0080\u0093202. In: Hubbard, W.; L. Biles; C. Mayfield; S. Ashton (Eds.). 2007. Sustainable Forestry for Bioenergy and Bio-based Products: Trainers Curriculum Notebook. Athens, GA: Southern Forest Research Partnership, Inc. EN 14918. 2014. Solid biofuels - method for the determination of calorific value. Channiwala, S.A. & Parikh, P.P. 2001. A unified correlation for estimating HHV of solid, liquid and gaseous fuels. Fuel (81): 1051-1063. Chico-Santamarta, L., R. J. Godwin, K. Chaney and D. R. White. 2013. On-farm storage of baled and pelletized canola straw. Energy 50(1): 429-437. Gorse, C., Johnston, C., and Pritchard, M. 2012. A Dictionary of Construction, Surveying and Civil Engineering. Oxford University Press. 67 A Dictionary of Construction, Surveying and Civil Engineering Deacon, J. W. 2006. Fungal Biology. Malden, MA: Blackwell Pub. Demirbas, A. 2002. Relationships between Heating Value and Lignin, Moisture, Ash and Extractive Contents of Biomass Fuels. Energy, Exploration & Exploitation 20(1): 105-111. European Biomass Association (AEBIOM). 2012. European bioenergy outlook 2012. European Pellet Council. 2013. EN-plus - handbook for the certification of wood pellets for heating purposes. (No. 2). Fengel, D. and G. Wegener. 1984. Wood: chemistry, ultrastructure, reactions. New York; Berlin: W. de Gruyter. Filbakk, T., R. Jirjis, J. Nurmi and O. H\u00C3\u00B8ib\u00C3\u00B8. 2011. The effect of bark content on quality parameters of Scots pine (Pinus sylvestris L.) pellets. Biomass and Bioenergy 35(8): 3342-3349 Flach, B. 2013. The Market for Wood Pellets in the Benelux. Global Agricultural Information Network NL3001. Flach, B., K. Bendz, R. Krautgartner and S. Lieberz. 2013. Biofuels annual: EU biofuels annual 2013. Global Agricultural Information Network NL3034. Fuller, W. S. 1985. Chip pile storage - a review of practices to avoid deterioration and economic losses. Tappi Journal 68(8): 48-52. Garstamg, J., A. Weekes, R. Poulter and D. Barlett . 2002. Identification and characterisation of factors affecting losses in the large-scale, non-ventilated bulk storage of wood chips and development of best storage practices. 116. Gaur, S. and Reed, T.B. 1995. An Atlas of Thermal Data for Biomass and Other Fuels. NREL/TB-433-7965. National Renewable Energy Laboratory, Golden, Colorado, USA. Gautam, S., R. Pulkki, C. Shahi and M. Leitch. 2012. Fuel quality changes in full tree logging residue during storage in roadside slash piles in Northwestern Ontario. Biomass and Bioenergy 42(0): 43-50. Gillespie, G. D., Everard, C. D., Fagan, C. C., and McDonnell, K. P. 2013. Prediction of quality parameters of biomass pellets from proximate and ultimate analysis. Fuel 111: 771-777. doi:http://dx.doi.org/10.1016/j.fuel.2013.05.002 Grbovic, M. 2010. Export Potential of U.S. - Produced Switchgrass and Wood Pellets for the EU Market. Master\u00E2\u0080\u0099s Thesis. Knoxville, TN: University of Tennessee. G\u00C3\u00BCnther, B., K. Gebauer, R. Barkowski, M. Rosenthal, and C. Bues. 2012. Calorific value of selected wood species and wood products. European Journal of Wood and Wood Products 70(5): 755-757. 68 International Energy Agency (IEA) Bioenergy. 2011. Global wood pellet industry market and trade study. Task 40: Sustainable International Bioenergy trade. Instituto Nacional de Estadistica Censo de la Contraloria General de la Republica de Panama (Panama). 2014. Geography and Climate: Panama in Figures, 2008-12 [Panama en Cifras, Anos 2008-12]: Panama in Figures, 2008-12 [Panama en Cifras, Anos 2008-12]; 2014 IIS 7230-S1.1. Jenkins, J. H. and F. W. Guernsey. 1932. Effect of moisture content and storage on heating value of sawdust. Department of the Interior, Canada -- Forest Service Circular (n 33): 16p. Jirjis, R. 2005. Effects of particle size and pile height on storage and fuel quality of comminuted Salix viminalis. Biomass and Bioenergy 28(2): 193-201. Krajnc, N., T. Jemec and P. Prislan. 2013. Monitoring of wood fuel prices in Slovenia, Austria, Italy, Croatia, Romania, Germany, Spain (incl. Catalonia) and Ireland. Report No. 4. Biomass Trade Centre. Larsson, S., T. Lestander, D. Crompton, S. Melin and S. Sokhansanj. 2012. Temperature patterns in large scale wood pellet silo storage. Applied Energy 92: 322-327. Lehtikangas, P. 2001. Quality properties of pelletised sawdust, logging residues and bark. Biomass and Bioenergy 20(5): 351-360. Melin, S. and J.S. Lee. 2013. Data compilation of fuel quality data of wood pellet before and after ocean transportation from ports in British Columbia to ports in Netherlands. Melin, S., U. Svedberg and J. Samuelsson. 2008. Emissions from Woodpellets during Ocean Transportation (EWDOT) - Research Report. Wood Pellet Association of Canada. Murray, G. 2012. Canadian Perspective on EU and other Export Market. In PFI Annual Conference, Wood Pellet Association of Canada. Natural Resources Canada. 2013. Wood pellet production trends in Canada. http://cfs.nrcan.gc.ca/selective-cuttings/57. (2013/11/12). Obernberger, I. and G. Thek. 2010. Chapter 3: Physio-chemical characterisation of raw materials and pellets. In The pellet handbook: the production and thermal utilisation of biomass pellets, 47-84. ed. Anonymous , London, UK: Earthscan Ltd. Obernberger, I. and G. Thek. 2010. The Pellet Handbook: The production and thermal utilisation of biomass pellets. London, UK: Earthscan. OECD/IEA. 2010. Power Generation from Coal \u00E2\u0080\u0093 Measuring and Reporting Efficiency Performance and CO2 Emissions. Paris, France: International Energy Agency (EIA). 69 Pellet Fuel Institute (PFI). 2011. Pellet fuels institute standard specification for residential/commercial densified fuel. Platace, R., Adamovics, A., and Gulbe, I. 2013. Evaluation of factors influencing calorific value of reed canary grass spring and autumn yield. Engineering for Rural Development. Rh\u00C3\u00A9n, C. 2004. Chemical composition and gross calorific value of the above-ground biomass components of young picea abies. Scandinavian Journal of Forest Research, 19(1): 72-81. doi:10.1080/02827580310019185 Svedberg, U. R. A., H. H\u00C3\u00B6gberg, J. A. G. H\u00C3\u00B6gberg and B. Galle. 2004. Emission of Hexanal and Carbon Monoxide from Storage of Wood Pellets, a Potential Occupational and Domestic Health Hazard. The Annal of Occupational Hygiene 48(4): 339-349. Singh, T. and M.M. Kostecky. 1986. Calorific value variations in components of 10 Canadian tree species. Canadian Journal of Forest Research 16: 1378-1381. Telmo, C. and J. Lousada. 2011. The explained variation by lignin and extractive contents on higher heating value of wood. Biomass and Bioenergy 35(5): 1663-1667. Thomson-Brooks/Cole. 2003. Organic Chemistry Laboratory Notebook. 1st Ed. Boston: Cengage Learning. Toscano, G., Riva, G., Foppa Pedretti, E., Corinaldesi, F., Mengarelli, C., and Duca, D. (2013). Investigation on wood pellet quality and relationship between ash content and the most important chemical elements. Biomass and Bioenergy 56: 317-322. doi:http://dx.doi.org/10.1016/j.biombioe.2013.05.012 Tumuluru, J. S., X. Kuang, S. Sokhansanj, C. J. Lim and X. Bi. 2009. Wood Pellet Properties Produced in British Columbia and Studies on Off-Gas Emissions at Low and High Storage Temperatures. In ASABE Paper No. 097265, St. Joseph, Mich.: ASABE. White, R. H. 1987. Effect of lignin content and extractives on the higher heating value of wood. Wood and Fiber Science 19(4): 446-452. Yazdanpanah, F., S. Sokhansanj, J. Lim, A. Lau, X. Bi, P. Y. Lam and S. Melin. 2014. Potential for flammability of gases emitted from stored wood pellets. The Canadian Journal of Chemical Engineering 92(4): 603-609. Yin, C. 2011. Prediction of higher heating values of biomass from proximate and ultimate analyses. Fuel 90(3): 1128-1132. Zielenkiewicz, W. and E. Margas. 2002. Theory of calorimetry. Boston; Dordrecht: Kluwer Academic Publishers. 70 Appendix A: Supporting Information A.1 Additional tables of data Table A-1 Effective heat capacity result from 10 calibration runs Run Effective heat capacity \u00CE\u00B5 (cal/\u00C2\u00B0C) 1 2417.3 2 2416.7 3 2415.6 4 2411.9 5 2413.2 6 2411.3 7 2417.9 8 2404.6 9 2413.0 10 2419.3 Average 2414.1 SD 4.3 %RSD 0.18 SD is standard deviation of the 10 calibration runs %RSD is percentage relative standard deviation 71 Table A-2: Carbon content and NCV data used to calculate correlation coefficient. Type Storage Configuration Temperature Storage Time (days) C (% d.b.) NCV (w.b.) White Initial NA 0 50.4 18.2 White Open 25 10 50.7 17.5 25 40 50.4 17.9 35 10 51.9 18.2 35 40 50.7 18.3 45 10 50.2 18.6 45 40 50.4 18.4 Closed 25 10 50.7 18.2 25 40 50.1 18.6 35 10 50.7 18.4 35 40 49.9 18.6 45 10 50.8 18.5 45 40 50.7 18.7 45 80 50.9 18.7 Mixed Initial NA 0 50.4 17.9 Mixed Open 25 10 50.5 17.9 25 40 50.5 17.8 35 10 50.1 18.3 35 40 51.3 18.3 45 10 50.2 18.7 45 40 51.8 18.5 Closed 25 10 50.3 17.9 25 40 49.9 18.0 35 10 50.3 18.3 35 40 51.4 18.4 45 10 50.3 18.6 45 40 50.8 18.2 45 80 51.4 18.6 72 Table A-3 Comparison between measurements done by SGS Delta analytical lab and in UBC for samples that are stored for 42 days (1.5 months) Storage Days 42 42 42 42 Sample type White Wood Mixed Wood White Wood Mixed Wood Storage Configuration Close Close Close Close Laboratory SGS Delta SGS Delta UBC BBRG Lab 628 UBC BBRG Lab 628 Date of Test Thursday, February 28, 2013 Thursday, February 28, 2013 Saturday, March 02, 2013 Saturday, March 02, 2013 Moisture Content (% MC) 3.06 6.13 2.78 5.29 Ash Content dry basis (% Ash) 0.60 1.21 0.77 1.57 Hydrogen Content (% dry mass) 6.08 6.12 6.08 6.12 Gross Calorific Value wet basis (MJ/w.b. kg) 19.90 19.63 19.81 19.53 Gross Calorific Value dry basis (MJ/d.b. kg) 20.53 20.91 20.38 20.62 Net Calorific Value wet basis (MJ/ w.b. kg) 18.58 18.26 18.49 18.17 Net Calorific Value dry basis (MJ/d.b. kg) 19.24 19.61 19.09 19.32 73 Table A-4: Comparison between the measurement method I employed and SGS Delta Analytical Laboratory used. Parameters My method SGS Delta Gross Calorific Values (GCV) Test Standard Modified EN 14918:2013 EN 14918 Sample Form Pellet was intact and not grinded before test Pellet was initially grinded to 1mm particle size and mixed. Type of Instrument PARR 6100 Isoperibol Oxygen Bomb Calorimeter Leco AC600 Isoperibol Oxygen Bomb Calorimeter Operating pressure 400 psig (2.7 MPa) 3 MPa Brief description of procedure 1. A wood pellet of ~0.8 gram is sampled. 2. The wood pellet is burned as it is, in high pressure oxygen in the bomb calorimeter. 1. 1 gram of grinded wood is pelletized. 2. The wood pellet is burned in high-pressure oxygen in the bomb calorimeter. Thermochemical Correction Nitric, sulfuric acid as well as fuse wire correction were applied. Nitric, sulfuric acid as well as fuse wire correction were applied. Moisture Content (%MC) Test Standard NREL/TP-510-42621 EN 14775-Oven dry method Sample Form Whole pellet Grinded powder of wood pellet Type of Instrument Convection Oven Convection Oven Oven drying temperature 105\u00C2\u00B13 \u00C2\u00B0C 105\u00C2\u00B12 \u00C2\u00B0C Brief description of Procedure 1. 1-2 gram sample was placed on an aluminium pan. 2. The sample was dried uncovered for 6 to 24 hours until constant mass was observed. 1. ~1 gram sample was placed on a crucible. 2. The sample was dried uncovered until constant mass was observed. 74 Ash Content (%Ash) Test Standard NREL/TP-510-42622 EN 14774-2 - Furnace method Sample Form Whole pellet Grinded powder of wood pellet Type of Instrument Programmable Furnace Programmable Furnace Maximum temperature 575\u00C2\u00B125 \u00C2\u00B0C 550\u00C2\u00B110 \u00C2\u00B0C Brief description of Procedure 1. ~1 gram sample is placed on a crucible. 2. Temperature is raised from 25 \u00C2\u00B0C to 105 \u00C2\u00B0C over 40 minutes then to 250 \u00C2\u00B0C over 15 minutes with heating rate of 10 \u00C2\u00B0C/minute. 3. 250 \u00C2\u00B0C is maintained for 30 minutes. The temperature is further ramped to 575 \u00C2\u00B0C over 16 minutes with heating rate of 20 \u00C2\u00B0C/minute. 4.. 575 \u00C2\u00B0C was maintained for 3 hours before cooling down to 105 \u00C2\u00B0C. 1. ~1 gram sample is placed on a crucible. 2. Temperature is raised from 25 \u00C2\u00B0C to 250 \u00C2\u00B0C over 30-50 minutes with heating rate of 4.5 to 7.5 \u00C2\u00B0C/minute. 3. 250 \u00C2\u00B0C is maintained for 1 hour. The temperature is further ramped to 550 \u00C2\u00B0C over 30 minutes with heating rate of 10 \u00C2\u00B0C/minute. 4. 550 \u00C2\u00B0C was maintained for 2 hours before cooling down to 105 \u00C2\u00B0C. 75 Table A-5: Comparison between measurements done by SGS Delta analytical lab and in UBC for samples that are stored for 84 days (3 months) Storage Days 84 84 84 84 Pellet type White Wood Mixed Wood White Wood Mixed Wood Storage Configuration Close Close Close Close Laboratory SGS Delta SGS Delta UBC BBRG Lab 628 UBC BBRG Lab 628 Date of Test April 11-2013 April 11-2013 April 12-2013 April 12-2013 Moisture Content (% MC) 2.81 5.34 2.59 5.25 Ash Content dry basis (% Ash) 0.62 1.21 0.65 1.25 Hydrogen Content (% dry mass) 5.96 6.08 5.96 6.08 GROSS Calorific Value AS RECEIVED (MJ/kg) 19.95 19.61 20.01 19.66 GROSS Calorific Value DRY basis (MJ/kg) 20.52 20.71 20.55 20.75 NET Calorific Value AS RECEIVED (MJ/kg) 18.62 18.22 18.72 18.31 NET Calorific Value DRY basis (MJ/kg) 19.23 19.46 19.28 19.46 76 Table A-6: The raw data that are used in this thesis. NCV is calculated from measured GCV using equation 3-1 using average hydrogen content. Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 0 White Initial Initial 1st 19.62 18.28 2.93 0.38 50.23 6.14 0.05 0 White Initial Initial 2nd 19.63 18.31 2.43 0.53 50.39 6.11 0.05 0 White Initial Initial 3rd 19.48 18.15 3.19 0.44 50.6 6.08 0.05 0 White Initial Initial 4th 19.53 18.20 2.81 0.35 50.33 6.13 0.05 0 White Initial Initial 5th 19.54 18.20 2.90 0.54 0 White Initial Initial 6th 19.34 18.02 2.53 0.45 0 White Initial Initial 7th 19.64 18.32 2.38 0.46 0 White Initial Initial 8th 19.64 18.30 3.08 0.42 0 White Initial Initial 9th 19.54 18.21 2.86 0.54 0 White Initial Initial 10th 19.51 18.19 2.43 0.47 5 White Open 25 1st 18.94 17.58 5.56 0.47 5 White Open 25 2nd 19.06 17.69 5.63 0.12 5 White Open 25 3rd 19.01 17.65 5.59 0.34 5 White Open 35 1st 19.14 17.80 3.82 0.25 5 White Open 35 2nd 19.23 17.89 3.81 0.44 5 White Open 35 3rd 19.17 17.83 3.80 0.37 5 White Open 45 1st 19.56 18.24 2.08 1.41 5 White Open 45 2nd 19.70 18.38 1.99 1.16 9 White Open 25 1st 18.88 17.51 6.54 0.36 50.70 5.94 0.05 9 White Open 25 2nd 18.88 17.51 6.20 0.42 9 White Open 35 1st 19.73 18.39 3.78 0.57 51.91 6.14 0.05 9 White Open 35 2nd 19.50 18.17 3.42 0.38 9 White Open 45 1st 19.97 18.66 1.27 0.46 50.20 6.08 0.05 9 White Open 45 2nd 19.98 18.67 1.25 0.65 77 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 14 White Open 25 1st 19.14 17.79 5.04 0.48 14 White Open 25 2nd 19.21 17.85 5.16 0.54 14 White Open 25 3rd 19.27 17.90 5.75 0.59 14 White Open 35 1st 19.41 18.07 3.71 0.45 14 White Open 35 2nd 19.53 18.19 3.51 0.60 14 White Open 35 3rd 19.51 18.17 3.68 0.41 14 White Open 45 1st 19.81 18.49 1.55 0.53 14 White Open 45 2nd 19.83 18.52 1.58 0.66 14 White Open 45 3rd 19.88 18.57 1.63 0.62 21 White Open 45 1st 19.78 18.46 1.66 0.69 21 White Open 45 2nd 19.98 18.66 2.22 0.57 21 White Open 45 3rd 19.95 18.63 1.81 0.65 21 White Open 25 1st 19.12 17.75 6.00 0.55 21 White Open 25 2nd 19.06 17.70 5.79 0.41 21 White Open 25 3rd 19.04 17.67 5.78 0.49 21 White Open 35 1st 19.58 18.25 3.45 0.49 21 White Open 35 2nd 19.57 18.23 3.51 0.73 21 White Open 35 3rd 19.52 18.18 3.44 1.13 28 White Open 25 1st 19.32 17.95 6.09 0.48 28 White Open 25 2nd 19.07 17.70 6.00 0.44 28 White Open 25 3rd 19.20 17.83 6.13 0.60 28 White Open 35 1st 19.47 18.12 3.85 0.47 28 White Open 35 2nd 18.91 17.57 3.81 0.52 28 White Open 35 3rd 19.47 18.13 3.80 0.52 28 White Open 45 1st 19.88 18.56 1.89 0.55 78 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 28 White Open 45 2nd 19.89 18.57 1.91 0.54 28 White Open 45 3rd 19.96 18.64 1.88 0.73 35 White Open 25 1st 18.99 17.63 6.08 0.48 50.39 6.00 0.05 35 White Open 25 2nd 19.19 17.83 6.08 0.47 35 White Open 25 3rd 19.18 17.81 6.14 0.26 35 White Open 35 1st 19.54 18.20 3.74 0.59 50.70 6.01 0.05 35 White Open 35 2nd 19.69 18.34 3.78 0.49 35 White Open 35 3rd 19.68 18.34 3.79 0.45 35 White Open 45 1st 20.15 18.83 1.82 0.42 50.42 6.11 0.05 35 White Open 45 2nd 20.01 18.69 1.73 0.55 35 White Open 45 3rd 20.15 18.84 1.70 1.02 49 White Open 25 1st 19.22 17.85 6.05 0.36 50.14 6.20 0.05 49 White Open 25 2nd 19.23 17.87 6.07 0.35 49 White Open 25 3rd 19.26 17.89 6.17 0.55 49 White Open 35 1st 19.67 18.33 3.70 0.48 49.87 6.15 0.05 49 White Open 35 2nd 19.76 18.42 3.88 0.47 49 White Open 35 3rd 19.67 18.33 3.71 0.49 49 White Open 45 1st 19.73 18.40 2.94 0.57 50.72 6.17 0.05 49 White Open 45 2nd 19.70 18.37 2.86 0.46 49 White Open 45 3rd 19.86 18.53 2.89 1.05 63 White Open 25 1st 19.20 17.83 5.89 0.44 63 White Open 25 2nd 19.21 17.84 5.90 0.60 63 White Open 25 3rd 19.13 17.76 5.83 0.43 63 White Open 35 1st 19.73 18.39 3.61 0.44 63 White Open 35 2nd 19.61 18.28 3.40 0.45 79 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 63 White Open 35 3rd 19.75 18.42 3.40 0.41 63 White Open 45 1st 20.06 18.74 1.76 0.52 63 White Open 45 2nd 19.96 18.65 1.54 0.79 63 White Open 45 3rd 20.06 18.74 1.50 0.69 77 White Open 25 1st 19.26 17.89 5.85 0.36 77 White Open 25 2nd 19.16 17.79 5.83 0.43 77 White Open 25 3rd 19.26 17.90 5.99 0.50 77 White Open 35 1st 19.61 18.28 3.26 0.52 77 White Open 35 2nd 19.69 18.35 3.37 0.68 77 White Open 35 3rd 19.67 18.34 3.41 0.52 77 White Open 45 1st 20.03 18.72 1.59 0.61 77 White Open 45 2nd 20.04 18.73 1.50 0.45 77 White Open 45 3rd 20.03 18.71 1.53 0.54 91 White Open 25 1st 19.31 17.94 6.08 0.45 91 White Open 25 2nd 19.27 17.90 5.92 0.61 91 White Open 25 3rd 19.26 17.90 6.00 0.59 91 White Open 35 1st 19.62 18.28 3.68 0.47 91 White Open 35 2nd 19.68 18.34 3.52 0.53 91 White Open 35 3rd 19.66 18.32 3.45 0.52 91 White Open 45 1st 19.85 18.53 1.84 0.62 91 White Open 45 2nd 20.03 18.71 1.82 0.52 91 White Open 45 3rd 19.98 18.66 1.82 0.77 133 White Open 25 1st 19.06 17.68 7.27 0.43 133 White Open 25 2nd 18.88 17.50 7.30 0.74 133 White Open 25 3rd 19.11 17.73 7.28 1.09 80 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 133 White Open 35 1st 19.62 18.27 4.45 0.45 133 White Open 35 2nd 19.57 18.22 4.42 1.09 133 White Open 35 3rd 19.61 18.26 4.25 0.87 133 White Open 45 1st 20.11 18.78 2.43 0.51 133 White Open 45 2nd 20.05 18.73 2.53 0.52 133 White Open 45 3rd 19.88 18.55 2.75 0.52 162 White Open 25 1st 19.21 17.85 5.22 0.43 162 White Open 25 2nd 19.31 17.95 5.45 0.64 162 White Open 25 3rd 19.16 17.81 5.16 1.00 162 White Open 35 1st 19.68 18.35 3.27 0.45 162 White Open 35 2nd 19.53 18.20 3.25 1.00 162 White Open 35 3rd 19.58 18.25 3.23 0.83 162 White Open 45 1st 20.03 18.71 1.31 0.51 162 White Open 45 2nd 19.88 18.57 1.35 0.49 162 White Open 45 3rd 19.88 18.57 1.33 0.55 197 White Open 25 1st 18.71 17.35 5.07 0.47 197 White Open 25 2nd 18.90 17.54 5.48 0.48 197 White Open 25 3rd 19.01 17.65 5.47 0.46 197 White Open 35 1st 19.26 17.92 3.25 0.68 197 White Open 35 2nd 19.32 17.99 3.23 0.72 197 White Open 35 3rd 19.42 18.08 3.50 0.64 197 White Open 45 1st 19.66 18.35 1.35 0.50 197 White Open 45 2nd 19.70 18.39 1.33 0.56 197 White Open 45 3rd 19.78 18.47 1.51 0.42 This row is left blank 81 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 5 White Closed 25 1st 19.65 18.32 2.76 0.42 5 White Closed 25 2nd 19.55 18.22 2.67 0.38 5 White Closed 25 3rd 19.60 19.53 2.72 0.40 5 White Closed 35 1st 19.52 18.19 2.92 0.51 5 White Closed 35 2nd 19.51 18.18 2.62 0.63 5 White Closed 35 3rd 19.50 19.43 2.72 0.54 5 White Closed 45 1st 19.23 17.90 2.85 1.65 5 White Closed 45 2nd 19.32 17.99 2.67 0.59 9 White Closed 25 1st 19.44 18.12 1.91 0.34 50.69 6 0.05 9 White Closed 25 2nd 19.51 18.18 3.07 1.04 9 White Closed 35 1st 19.78 18.45 2.48 0.55 50.65 6.19 0.05 9 White Closed 35 2nd 19.71 18.39 2.41 0.45 9 White Closed 45 1st 19.90 18.57 2.10 0.40 50.78 6.37 0.05 9 White Closed 45 2nd 19.81 18.49 2.11 0.65 14 White Closed 25 1st 19.73 18.40 2.65 0.56 14 White Closed 25 2nd 19.71 18.38 2.56 0.73 14 White Closed 25 3rd 19.74 18.41 2.72 0.50 14 White Closed 35 1st 19.59 18.26 2.62 0.61 14 White Closed 35 2nd 19.73 18.40 2.68 0.68 14 White Closed 35 3rd 19.80 18.47 2.59 0.45 14 White Closed 45 1st 19.70 18.37 2.72 0.53 14 White Closed 45 2nd 19.66 18.33 2.70 0.50 14 White Closed 45 3rd 19.71 18.38 2.68 0.78 21 White Closed 25 1st 19.88 18.55 2.57 0.46 21 White Closed 25 2nd 19.95 18.62 2.49 0.54 82 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 21 White Closed 25 3rd 19.85 18.52 2.62 0.54 21 White Closed 35 1st 19.98 18.65 2.95 0.55 21 White Closed 35 2nd 19.87 18.54 2.96 0.64 21 White Closed 35 3rd 19.67 18.34 2.91 0.52 21 White Closed 45 1st 19.88 18.55 2.75 0.58 21 White Closed 45 2nd 19.78 18.46 2.75 0.52 21 White Closed 45 3rd 19.76 18.43 2.71 0.85 28 White Closed 25 1st 19.99 18.66 2.64 0.69 28 White Closed 25 2nd 19.85 18.52 2.66 0.65 28 White Closed 25 3rd 19.83 18.50 2.62 0.49 28 White Closed 35 1st 19.86 18.54 2.71 0.52 28 White Closed 35 2nd 19.70 18.37 2.71 0.56 28 White Closed 35 3rd 19.80 18.47 2.68 0.57 28 White Closed 45 1st 19.88 18.55 2.61 0.59 28 White Closed 45 2nd 19.64 18.31 2.69 0.54 28 White Closed 45 3rd 19.97 18.64 2.69 0.79 42 White Closed 25 1st 19.86 18.54 2.51 0.51 42 White Closed 25 2nd 19.94 18.62 2.50 0.48 42 White Closed 25 3rd 19.91 18.58 2.60 0.47 42 White Closed 35 1st 19.97 18.64 2.89 0.62 42 White Closed 35 2nd 19.88 18.55 3.03 0.53 42 White Closed 35 3rd 19.92 18.59 2.81 0.56 42 White Closed 45 1st 20.00 18.67 2.82 0.59 42 White Closed 45 2nd 19.90 18.57 2.77 0.67 42 White Closed 45 3rd 20.10 18.77 2.76 0.98 83 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 56 White Closed 25 1st 20.03 18.71 2.42 0.54 56 White Closed 25 2nd 19.99 18.67 2.52 0.61 56 White Closed 25 3rd 19.99 18.66 2.35 0.44 56 White Closed 35 1st 20.07 18.74 2.75 0.40 56 White Closed 35 2nd 19.97 18.64 2.77 0.42 56 White Closed 35 3rd 19.87 18.54 2.77 0.52 56 White Closed 45 1st 19.83 18.50 2.62 0.48 56 White Closed 45 2nd 19.90 18.58 2.41 0.53 56 White Closed 45 3rd 20.00 18.67 2.51 0.73 70 White Closed 25 1st 19.92 18.60 2.40 0.55 70 White Closed 25 2nd 19.97 18.65 2.47 0.48 70 White Closed 25 3rd 19.99 18.67 2.41 0.58 70 White Closed 35 1st 19.81 18.48 2.83 0.47 70 White Closed 35 2nd 19.87 18.54 2.83 0.48 70 White Closed 35 3rd 20.17 18.84 2.71 0.49 70 White Closed 45 1st 19.94 18.62 2.40 0.55 70 White Closed 45 2nd 20.03 18.71 2.44 0.59 70 White Closed 45 3rd 19.92 18.59 2.34 0.80 84 White Closed 25 1st 20.05 18.72 2.78 0.50 84 White Closed 25 2nd 19.92 18.59 2.74 0.51 84 White Closed 25 3rd 19.96 18.63 2.60 0.66 84 White Closed 35 1st 19.65 18.32 2.80 0.44 84 White Closed 35 2nd 19.90 18.57 2.78 0.50 84 White Closed 35 3rd 19.76 18.43 2.74 0.57 84 White Closed 45 1st 19.85 18.53 2.51 0.57 50.9 5.96 0.05 0.02 84 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 84 White Closed 45 2nd 19.90 18.58 2.52 0.51 84 White Closed 45 3rd 19.97 18.65 2.53 0.69 118 White Closed 25 1st 20.15 18.82 2.66 0.61 118 White Closed 25 2nd 20.09 18.76 2.70 0.59 118 White Closed 25 3rd 20.02 18.69 2.66 0.55 118 White Closed 35 1st 20.07 18.73 3.17 0.41 118 White Closed 35 2nd 19.94 18.61 3.10 0.46 118 White Closed 35 3rd 19.99 18.66 3.15 0.42 118 White Closed 45 1st 20.00 18.67 2.85 0.38 118 White Closed 45 2nd 20.07 18.74 2.98 0.59 118 White Closed 45 3rd 20.03 18.70 2.90 0.55 157 White Closed 25 1st 19.97 18.64 3.03 0.85 157 White Closed 25 2nd 19.97 18.64 3.03 0.41 157 White Closed 25 3rd 20.27 18.93 3.07 0.37 157 White Closed 35 1st 19.91 18.58 3.02 0.28 157 White Closed 35 2nd 19.97 18.64 2.90 0.42 157 White Closed 35 3rd 19.99 18.66 2.94 0.61 157 White Closed 45 1st 20.06 18.73 2.78 0.39 157 White Closed 45 2nd 20.01 18.68 2.75 0.59 157 White Closed 45 3rd 19.96 18.63 2.77 0.49 192 White Closed 25 1st 19.93 18.61 2.03 0.32 192 White Closed 25 2nd 19.82 18.49 2.35 0.61 192 White Closed 25 3rd 19.94 18.62 2.02 0.60 192 White Closed 35 1st 19.90 18.57 2.12 0.66 192 White Closed 35 2nd 19.80 18.48 2.16 0.58 85 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 192 White Closed 35 3rd 19.87 18.54 2.23 0.46 192 White Closed 45 1st 19.83 18.51 2.11 0.57 192 White Closed 45 2nd 19.78 18.46 2.09 0.68 192 White Closed 45 3rd 19.79 18.47 2.06 0.76 0 Mixed Initial Initial 1st 19.01 17.65 5.88 0.84 49.48 6.11 0.05 0 Mixed Initial Initial 2nd 19.24 17.90 3.27 1.03 50.19 6.17 0.05 0 Mixed Initial Initial 3rd 19.03 17.64 6.90 0.91 51.12 6.19 0.05 0 Mixed Initial Initial 4th 19.24 17.88 4.08 1.65 50.7 6.18 0.05 0 Mixed Initial Initial 5th 19.46 18.09 6.11 1.62 0 Mixed Initial Initial 6th 19.44 18.09 3.29 1.01 0 Mixed Initial Initial 7th 19.32 17.94 5.66 1.58 0 Mixed Initial Initial 8th 19.16 17.80 4.30 1.10 0 Mixed Initial Initial 9th 19.27 17.90 6.46 1.08 0 Mixed Initial Initial 10th 19.38 18.03 3.45 1.15 5 Mixed Open 25 1st 19.00 17.63 6.25 0.83 5 Mixed Open 25 2nd 18.96 17.59 7.10 0.67 5 Mixed Open 35 1st 19.23 17.89 4.26 2.68 5 Mixed Open 35 2nd 19.17 17.83 4.54 0.62 5 Mixed Open 45 1st 19.54 18.23 2.35 2.51 5 Mixed Open 45 2nd 19.72 18.40 2.37 0.94 9 Mixed Open 25 1st 19.18 17.83 5.65 0.81 50.5 6.17 0.05 9 Mixed Open 25 2nd 19.31 17.94 6.72 0.82 9 Mixed Open 35 1st 19.70 18.36 3.64 1.65 50.1 5.86 0.05 9 Mixed Open 35 2nd 19.58 18.24 4.84 1.20 9 Mixed Open 45 1st 20.01 18.71 1.22 1.77 50.19 6.17 0.05 86 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 9 Mixed Open 45 2nd 19.99 18.69 1.21 0.84 14 Mixed Open 25 1st 19.18 17.82 6.54 0.98 14 Mixed Open 25 2nd 19.11 17.74 6.47 1.22 14 Mixed Open 25 3rd 19.13 17.76 6.64 0.95 14 Mixed Open 35 1st 19.63 18.29 3.86 1.05 14 Mixed Open 35 2nd 19.70 18.36 3.76 1.36 14 Mixed Open 35 3rd 19.60 18.27 3.96 1.33 14 Mixed Open 45 1st 19.84 18.53 1.74 1.73 14 Mixed Open 45 2nd 20.11 18.80 1.78 1.02 14 Mixed Open 45 3rd 19.92 18.61 1.68 1.14 21 Mixed Open 25 1st 19.33 17.96 7.13 1.23 21 Mixed Open 25 2nd 19.02 17.66 6.43 1.35 21 Mixed Open 25 3rd 19.09 17.72 7.05 1.18 21 Mixed Open 45 1st 20.07 18.76 2.35 1.05 21 Mixed Open 45 2nd 19.91 18.59 2.33 0.95 21 Mixed Open 45 3rd 19.88 18.56 2.15 1.56 21 Mixed Open 35 1st 19.75 18.42 3.70 1.21 21 Mixed Open 35 2nd 19.74 18.41 4.03 1.03 21 Mixed Open 35 3rd 19.65 18.32 4.04 1.24 28 Mixed Open 25 1st 19.13 17.76 7.10 0.85 28 Mixed Open 25 2nd 19.01 17.64 6.82 1.07 28 Mixed Open 25 3rd 18.95 17.59 6.53 1.07 28 Mixed Open 35 1st 19.61 18.27 4.29 1.26 28 Mixed Open 35 2nd 19.73 18.39 4.37 1.00 28 Mixed Open 35 3rd 19.61 18.27 4.48 1.02 87 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 28 Mixed Open 45 1st 19.94 18.63 2.47 1.51 28 Mixed Open 45 2nd 19.94 18.63 2.46 0.99 28 Mixed Open 45 3rd 19.96 18.65 2.36 1.19 35 Mixed Open 25 1st 19.37 18.00 7.30 0.96 50.54 6.04 0.05 35 Mixed Open 25 2nd 19.13 17.76 6.81 1.01 35 Mixed Open 25 3rd 19.20 17.83 6.74 0.89 35 Mixed Open 35 1st 19.67 18.33 4.16 1.09 51.32 6.11 0.05 35 Mixed Open 35 2nd 19.71 18.38 4.16 1.18 35 Mixed Open 35 3rd 19.49 18.16 3.94 0.74 35 Mixed Open 45 1st 20.16 18.85 2.13 1.53 51.77 6.07 0.05 35 Mixed Open 45 2nd 20.08 18.77 2.05 1.24 35 Mixed Open 45 3rd 20.01 18.70 2.03 1.53 49 Mixed Open 25 1st 19.30 17.97 3.92 0.95 49.87 6.14 0.05 49 Mixed Open 25 2nd 19.03 17.66 6.89 1.18 49 Mixed Open 25 3rd 19.34 17.98 6.54 0.99 49 Mixed Open 35 1st 19.73 18.39 4.16 0.93 51.37 6.37 0.05 49 Mixed Open 35 2nd 19.74 18.41 3.95 0.70 49 Mixed Open 35 3rd 19.70 18.37 3.89 1.26 49 Mixed Open 45 1st 19.87 18.55 3.15 1.05 50.77 6.46 0.05 49 Mixed Open 45 2nd 19.74 18.42 3.11 1.03 49 Mixed Open 45 3rd 19.68 18.36 3.15 1.50 63 Mixed Open 25 1st 19.17 17.80 6.95 0.93 63 Mixed Open 25 2nd 19.53 18.16 6.87 0.73 63 Mixed Open 25 3rd 19.35 17.98 7.15 0.82 63 Mixed Open 35 1st 19.65 18.31 3.82 0.91 88 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 63 Mixed Open 35 2nd 19.77 18.44 3.71 0.95 63 Mixed Open 35 3rd 19.76 18.43 3.68 0.99 63 Mixed Open 45 1st 20.12 18.81 2.02 1.24 63 Mixed Open 45 2nd 20.04 18.73 1.82 1.28 63 Mixed Open 45 3rd 20.10 18.78 2.04 1.16 77 Mixed Open 25 1st 19.31 17.94 6.98 0.89 77 Mixed Open 25 2nd 18.95 17.59 6.71 0.90 77 Mixed Open 25 3rd 18.99 17.61 7.18 1.00 77 Mixed Open 35 1st 19.67 18.34 3.52 0.91 77 Mixed Open 35 2nd 19.67 18.33 3.56 0.86 77 Mixed Open 35 3rd 19.72 18.39 3.71 1.06 77 Mixed Open 45 1st 19.96 18.65 1.92 1.18 77 Mixed Open 45 2nd 20.05 18.73 1.89 1.84 77 Mixed Open 45 3rd 20.03 18.72 1.82 1.45 91 Mixed Open 25 1st 19.07 17.73 3.88 0.83 91 Mixed Open 25 2nd 19.13 17.80 3.71 0.58 91 Mixed Open 25 3rd 19.03 17.70 3.83 0.68 91 Mixed Open 35 1st 19.87 18.54 3.87 1.05 91 Mixed Open 35 2nd 19.57 18.24 3.79 0.95 91 Mixed Open 35 3rd 19.69 18.35 3.97 0.76 91 Mixed Open 45 1st 19.95 18.63 2.07 1.31 91 Mixed Open 45 2nd 20.09 18.77 2.10 1.27 91 Mixed Open 45 3rd 20.12 18.80 2.06 1.22 133 Mixed Open 25 1st 19.10 17.76 4.32 1.16 133 Mixed Open 25 2nd 18.83 17.50 3.56 0.85 89 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 133 Mixed Open 25 3rd 19.16 17.77 8.26 1.06 133 Mixed Open 35 1st 19.73 18.39 4.72 1.30 133 Mixed Open 35 2nd 19.78 18.44 4.73 0.81 133 Mixed Open 35 3rd 19.60 18.26 4.54 0.92 133 Mixed Open 45 1st 19.92 18.60 2.74 1.09 133 Mixed Open 45 2nd 19.86 18.54 2.72 0.99 133 Mixed Open 45 3rd 19.57 18.25 2.71 1.17 162 Mixed Open 25 1st 19.00 17.64 5.96 1.16 162 Mixed Open 25 2nd 19.11 17.75 6.53 1.10 162 Mixed Open 25 3rd 19.12 17.76 6.51 0.99 162 Mixed Open 35 1st 19.70 18.37 3.55 1.30 162 Mixed Open 35 2nd 19.74 18.41 3.59 0.90 162 Mixed Open 35 3rd 19.67 18.34 3.33 1.22 162 Mixed Open 45 1st 20.03 18.72 1.61 1.09 162 Mixed Open 45 2nd 19.91 18.60 1.64 1.10 162 Mixed Open 45 3rd 19.99 18.68 1.59 1.20 197 Mixed Open 25 1st 19.39 18.04 4.97 1.30 197 Mixed Open 25 2nd 19.00 17.65 5.36 1.40 197 Mixed Open 25 3rd 19.11 17.76 5.07 1.17 197 Mixed Open 35 1st 19.29 17.95 4.69 0.80 197 Mixed Open 35 2nd 19.70 18.36 3.96 0.83 197 Mixed Open 35 3rd 19.74 18.40 4.31 0.77 197 Mixed Open 45 1st 20.09 18.79 1.30 0.93 197 Mixed Open 45 2nd 20.03 18.72 1.49 0.96 197 Mixed Open 45 3rd 19.99 18.69 1.40 0.87 90 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 5 Mixed Closed 25 1st 19.15 17.79 5.79 1.21 5 Mixed Closed 25 2nd 19.26 17.91 5.55 0.55 5 Mixed Closed 35 1st 19.26 17.92 4.15 1.02 5 Mixed Closed 35 2nd 19.45 18.11 3.97 0.58 5 Mixed Closed 45 1st 19.26 17.93 3.38 0.48 5 Mixed Closed 45 2nd 19.24 17.91 3.55 1.20 9 Mixed Closed 25 1st 19.46 18.10 5.93 1.90 50.29 6.18 0.05 9 Mixed Closed 25 2nd 19.02 17.67 5.35 0.88 9 Mixed Closed 35 1st 19.72 18.39 3.55 1.09 50.28 6.04 0.05 9 Mixed Closed 35 2nd 19.65 18.32 3.62 0.58 9 Mixed Closed 45 1st 19.85 18.52 3.22 0.72 50.34 6.06 0.05 9 Mixed Closed 45 2nd 19.94 18.62 2.81 0.81 14 Mixed Closed 25 1st 19.35 17.99 5.71 1.02 14 Mixed Closed 25 2nd 19.15 17.79 6.07 1.00 14 Mixed Closed 25 3rd 19.36 18.00 6.02 1.04 14 Mixed Closed 35 1st 19.83 18.49 3.95 0.95 14 Mixed Closed 35 2nd 19.66 18.32 3.61 0.78 14 Mixed Closed 35 3rd 19.61 18.28 3.80 0.69 14 Mixed Closed 45 1st 19.54 18.20 4.42 1.20 14 Mixed Closed 45 2nd 19.45 18.11 4.44 1.14 14 Mixed Closed 45 3rd 19.42 18.08 4.36 1.45 21 Mixed Closed 25 1st 19.68 18.32 5.57 1.39 21 Mixed Closed 25 2nd 19.56 18.21 5.67 1.01 21 Mixed Closed 25 3rd 19.22 17.87 5.75 1.02 21 Mixed Closed 35 1st 19.80 18.46 4.02 0.94 91 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 21 Mixed Closed 35 2nd 20.00 18.66 3.98 0.65 21 Mixed Closed 35 3rd 19.67 18.33 4.05 0.79 21 Mixed Closed 45 1st 19.63 18.28 5.50 1.03 21 Mixed Closed 45 2nd 19.37 18.01 5.43 1.31 21 Mixed Closed 45 3rd 19.66 18.31 5.20 1.06 28 Mixed Closed 25 1st 19.37 18.01 5.87 1.04 28 Mixed Closed 25 2nd 19.30 17.94 6.07 0.93 28 Mixed Closed 25 3rd 19.42 18.06 6.19 1.09 28 Mixed Closed 35 1st 19.54 18.20 4.36 1.03 28 Mixed Closed 35 2nd 19.65 18.31 4.20 0.81 28 Mixed Closed 35 3rd 19.61 18.27 4.39 0.95 28 Mixed Closed 45 1st 19.48 18.14 4.93 1.30 28 Mixed Closed 45 2nd 19.41 18.06 5.09 1.46 28 Mixed Closed 45 3rd 19.41 18.07 5.00 0.97 42 Mixed Closed 25 1st 19.37 18.01 6.24 0.82 42 Mixed Closed 25 2nd 19.30 17.94 6.10 0.85 42 Mixed Closed 25 3rd 19.26 17.90 6.01 0.86 42 Mixed Closed 35 1st 19.43 18.09 3.92 0.82 42 Mixed Closed 35 2nd 19.69 18.36 3.69 0.66 42 Mixed Closed 35 3rd 19.88 18.55 3.87 0.71 42 Mixed Closed 45 1st 19.73 18.38 5.24 1.24 42 Mixed Closed 45 2nd 19.43 18.08 5.24 1.10 42 Mixed Closed 45 3rd 19.63 18.28 5.38 2.11 56 Mixed Closed 25 1st 19.28 17.92 5.56 0.79 56 Mixed Closed 25 2nd 19.38 18.02 6.00 0.93 92 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 56 Mixed Closed 25 3rd 19.33 17.97 6.10 1.29 56 Mixed Closed 35 1st 19.67 18.33 4.29 0.72 56 Mixed Closed 35 2nd 19.52 18.18 4.09 0.50 56 Mixed Closed 35 3rd 19.71 18.37 4.50 0.79 56 Mixed Closed 45 1st 19.48 18.13 4.82 1.21 56 Mixed Closed 45 2nd 19.62 18.28 4.82 1.00 56 Mixed Closed 45 3rd 19.63 18.28 4.86 1.34 70 Mixed Closed 25 1st 19.50 18.14 6.01 0.82 70 Mixed Closed 25 2nd 19.40 18.05 5.77 1.09 70 Mixed Closed 25 3rd 19.35 17.99 5.87 0.93 70 Mixed Closed 35 1st 19.44 18.09 5.04 0.96 70 Mixed Closed 35 2nd 19.64 18.29 5.09 1.08 70 Mixed Closed 35 3rd 19.56 18.21 5.23 0.90 70 Mixed Closed 45 1st 19.91 18.59 3.46 1.38 70 Mixed Closed 45 2nd 19.87 18.54 3.48 0.90 70 Mixed Closed 45 3rd 19.81 18.48 3.42 1.17 84 Mixed Closed 25 1st 19.31 17.95 5.56 0.91 84 Mixed Closed 25 2nd 19.25 17.90 5.77 0.87 84 Mixed Closed 25 3rd 19.35 18.00 6.01 0.90 84 Mixed Closed 35 1st 19.68 18.35 3.80 0.82 84 Mixed Closed 35 2nd 19.87 18.53 3.94 1.08 84 Mixed Closed 35 3rd 19.92 18.59 3.82 1.04 84 Mixed Closed 45 1st 19.62 18.29 3.57 1.08 51.35 6.08 0.05 0.01 84 Mixed Closed 45 2nd 19.84 18.51 3.81 1.49 84 Mixed Closed 45 3rd 19.85 18.52 3.78 1.35 93 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 118 Mixed Closed 25 1st 19.56 18.20 6.21 0.76 118 Mixed Closed 25 2nd 19.64 18.28 6.17 1.10 118 Mixed Closed 25 3rd 19.88 18.51 6.71 0.71 118 Mixed Closed 35 1st 19.50 18.14 5.98 0.99 118 Mixed Closed 35 2nd 19.68 18.33 5.69 0.86 118 Mixed Closed 35 3rd 19.70 18.34 5.72 0.79 118 Mixed Closed 45 1st 19.72 18.36 5.68 1.12 118 Mixed Closed 45 2nd 19.78 18.42 6.10 1.21 118 Mixed Closed 45 3rd 19.86 18.50 5.95 0.84 157 Mixed Closed 25 1st 19.68 18.31 6.31 0.91 157 Mixed Closed 25 2nd 19.62 18.25 6.69 1.41 157 Mixed Closed 25 3rd 19.50 18.14 5.99 0.92 157 Mixed Closed 35 1st 19.82 18.46 6.06 0.82 157 Mixed Closed 35 2nd 19.57 18.21 6.00 1.62 157 Mixed Closed 35 3rd 19.38 18.02 5.92 1.36 157 Mixed Closed 45 1st 19.81 18.47 4.44 1.04 157 Mixed Closed 45 2nd 19.68 18.34 4.11 1.05 157 Mixed Closed 45 3rd 19.82 18.48 4.16 1.07 192 Mixed Closed 25 1st 19.46 18.11 5.92 0.86 192 Mixed Closed 25 2nd 19.55 18.19 6.20 1.07 192 Mixed Closed 25 3rd 19.39 18.04 5.76 0.99 192 Mixed Closed 35 1st 19.55 18.19 5.62 1.04 192 Mixed Closed 35 2nd 19.57 18.21 5.68 0.84 192 Mixed Closed 35 3rd 19.43 18.08 5.83 0.89 192 Mixed Closed 45 1st 19.97 18.64 4.16 1.21 94 Day Pellet Type (White/Mixed) Storage configuration (Closed/Open) Temperature (\u00C2\u00B0C) Trial Number GCV (measured) (MJ/kg) NCV (calculated) (MJ/kg) M.C. (%w.b.) A.C. (%d.b.) %C (d.b.) %H (d.b.) %N (d.b.) %S (d.b.) 192 Mixed Closed 45 2nd 19.65 18.31 4.27 0.97 192 Mixed Closed 45 3rd 19.86 18.52 4.15 0.96 95 Table A-7: Example of the integration of calorific value over storage period using the coefficients in Table 4-3 \u00E2\u0080\u009COpen\u00E2\u0080\u009D storage configuration. The daily temperatures were obtained from an unnamed wood pellet voyage. This table is only for illustration. Duration of Storage (days) Temp (\u00C2\u00B0C) Number of days 1/days Cumulative Sum of fraction 1 7.3 2396 0.00042 0.00042 2 6.9 2416 0.00041 0.00083 3 5.6 2487 0.00040 0.00123 4 15.8 1963 0.00051 0.00174 5 17.9 1858 0.00054 0.00228 6 21.3 1685 0.00059 0.00287 7 20.9 1702 0.00059 0.00346 8 11.7 2175 0.00046 0.00392 9 5.4 2495 0.00040 0.00432 10 5.3 2498 0.00040 0.00472 11 3.3 2600 0.00038 0.00511 12 1.7 2685 0.00037 0.00548 13 1.9 2671 0.00037 0.00585 14 5.2 2507 0.00040 0.00625 15 10.3 2243 0.00045 0.00670 16 10.2 2252 0.00044 0.00714 17 2.8 2626 0.00038 0.00752 18 1.6 2691 0.00037 0.00790 19 9.6 2283 0.00044 0.00833 20 17.2 1895 0.00053 0.00886 21 11.1 2206 0.00045 0.00932 22 11.0 2209 0.00045 0.00977 23 16.8 1912 0.00052 0.01029 24 13.1 2101 0.00048 0.01077 25 8.3 2345 0.00043 0.01119 26 10.5 2235 0.00045 0.01164 27 12.2 2147 0.00047 0.01211 28 15.7 1968 0.00051 0.01262 29 7.2 2402 0.00042 0.01303 30 14.0 2056 0.00049 0.01352 31 15.6 1977 0.00051 0.01402 32 17.3 1889 0.00053 0.01455 33 19.2 1793 0.00056 0.01511 34 22.1 1642 0.00061 0.01572 The data from day 35 to day 313 are omitted. 96 Duration of Storage (days) Temp (\u00C2\u00B0C) Number of days 1/days Cumulative Sum of fraction 315 23.2 1589 0.00063 0.19477 316 21.6 1668 0.00060 0.19537 317 23.6 1569 0.00064 0.19601 318 18.3 1835 0.00054 0.19655 319 23.7 1563 0.00064 0.19719 320 25.3 1481 0.00068 0.19787 321 16.2 1943 0.00051 0.19838 322 11.1 2203 0.00045 0.19883 323 19.3 1784 0.00056 0.19940 324 22.2 1637 0.00061 0.20001 325 9.7 2277 0.00044 0.20045 326 8.7 2328 0.00043 0.20088 327 11.7 2172 0.00046 0.20134 328 9.8 2271 0.00044 0.20178 329 11.9 2164 0.00046 0.20224 330 19.4 1778 0.00056 0.20280 331 19.5 1776 0.00056 0.20336 332 21.4 1676 0.00060 0.20396 333 17.6 1875 0.00053 0.20449 334 17.0 1903 0.00053 0.20502 335 12.2 2147 0.00047 0.20548 336 20.0 1750 0.00057 0.20606 337 21.3 1682 0.00059 0.20665 338 6.7 2430 0.00041 0.20706 339 5.1 2512 0.00040 0.20746 340 8.6 2331 0.00043 0.20789 341 12.3 2141 0.00047 0.20836 342 13.6 2076 0.00048 0.20884 343 22.2 1637 0.00061 0.20945 344 23.1 1591 0.00063 0.21008 345 14.9 2011 0.00050 0.21057 346 8.4 2339 0.00043 0.21100 347 11.2 2198 0.00046 0.21146 348 12.4 2135 0.00047 0.21193 349 9.7 2277 0.00044 0.21236 350 11.3 2195 0.00046 0.21282 351 13.1 2101 0.00048 0.21330 97 Duration of Storage (days) Temp (\u00C2\u00B0C) Number of days 1/days Cumulative Sum of fraction 352 13.3 2093 0.00048 0.21377 353 13.5 2082 0.00048 0.21425 354 13.2 2099 0.00048 0.21473 355 17.9 1858 0.00054 0.21527 356 15.4 1985 0.00050 0.21577 357 6.8 2424 0.00041 0.21619 358 6.9 2419 0.00041 0.21660 359 7.5 2388 0.00042 0.21702 360 7.2 2402 0.00042 0.21743 361 6.9 2416 0.00041 0.21785 362 11.8 2167 0.00046 0.21831 363 14.9 2011 0.00050 0.21881 364 8.4 2339 0.00043 0.21923 365 6.1 2461 0.00041 0.21964 98 Table A-8: Example of the integration of calorific value over storage period using the coefficients in Table 4-3 \u00E2\u0080\u009CClosed\u00E2\u0080\u009D storage configuration. The daily temperatures were obtained from an unnamed wood pellet voyage. This table is only for illustration. Duration of Storage (days) Temp (\u00C2\u00B0C) Number of days 1/days Cumulative Sum of fraction 1 7.3 601 0.00166 0.00166 2 6.9 600 0.00167 0.00333 3 5.6 597 0.00167 0.00501 4 15.8 618 0.00162 0.00663 5 17.9 622 0.00161 0.00823 6 21.3 629 0.00159 0.00982 7 20.9 628 0.00159 0.01142 8 11.7 609 0.00164 0.01306 9 5.4 597 0.00168 0.01473 10 5.3 597 0.00168 0.01641 11 3.3 593 0.00169 0.01810 12 1.7 589 0.00170 0.01979 13 1.9 590 0.00170 0.02149 14 5.2 596 0.00168 0.02317 15 10.3 607 0.00165 0.02481 16 10.2 606 0.00165 0.02646 17 2.8 592 0.00169 0.02815 18 1.6 589 0.00170 0.02985 19 9.6 605 0.00165 0.03150 20 17.2 620 0.00161 0.03312 21 11.1 608 0.00164 0.03476 22 11.0 608 0.00164 0.03641 23 16.8 620 0.00161 0.03802 24 13.1 612 0.00163 0.03965 25 8.3 603 0.00166 0.04131 26 10.5 607 0.00165 0.04296 27 12.2 610 0.00164 0.04460 28 15.7 617 0.00162 0.04622 29 7.2 600 0.00167 0.04788 30 14.0 614 0.00163 0.04951 31 15.6 617 0.00162 0.05113 32 17.3 621 0.00161 0.05274 33 19.2 624 0.00160 0.05434 The data from day 35 to day 313 are omitted. 99 Duration of Storage (days) Temp (\u00C2\u00B0C) Number of days 1/days Cumulative Sum of fraction 34 22.1 630 0.00159 0.05593 315 23.2 632 0.00158 0.50182 316 21.6 629 0.00159 0.50341 317 23.6 633 0.00158 0.50499 318 18.3 623 0.00161 0.50660 319 23.7 633 0.00158 0.50818 320 25.3 637 0.00157 0.50975 321 16.2 618 0.00162 0.51137 322 11.1 608 0.00164 0.51301 323 19.3 625 0.00160 0.51461 324 22.2 630 0.00159 0.51620 325 9.7 605 0.00165 0.51785 326 8.7 603 0.00166 0.51951 327 11.7 609 0.00164 0.52115 328 9.8 606 0.00165 0.52280 329 11.9 610 0.00164 0.52444 330 19.4 625 0.00160 0.52604 331 19.5 625 0.00160 0.52764 332 21.4 629 0.00159 0.52923 333 17.6 621 0.00161 0.53084 334 17.0 620 0.00161 0.53245 335 12.2 610 0.00164 0.53409 336 20.0 626 0.00160 0.53569 337 21.3 629 0.00159 0.53728 338 6.7 599 0.00167 0.53895 339 5.1 596 0.00168 0.54062 340 8.6 603 0.00166 0.54228 341 12.3 611 0.00164 0.54392 342 13.6 613 0.00163 0.54555 343 22.2 630 0.00159 0.54714 344 23.1 632 0.00158 0.54872 345 14.9 616 0.00162 0.55034 346 8.4 603 0.00166 0.55200 347 11.2 608 0.00164 0.55364 348 12.4 611 0.00164 0.55528 349 9.7 605 0.00165 0.55693 350 11.3 609 0.00164 0.55858 100 Duration of Storage (days) Temp (\u00C2\u00B0C) Number of days 1/days Cumulative Sum of fraction 351 13.1 612 0.00163 0.56021 352 13.3 613 0.00163 0.56184 353 13.5 613 0.00163 0.56347 354 13.2 612 0.00163 0.56511 355 17.9 622 0.00161 0.56672 356 15.4 617 0.00162 0.56834 357 6.8 600 0.00167 0.57000 358 6.9 600 0.00167 0.57167 359 7.5 601 0.00166 0.57334 360 7.2 600 0.00167 0.57500 361 6.9 600 0.00167 0.57667 362 11.8 610 0.00164 0.57831 363 14.9 616 0.00162 0.57993 364 8.4 603 0.00166 0.58159 365 6.1 598 0.00167 0.58326 101 A.2 Derivation of constant pressure NCV from calorimetric GCV This derivation is adapted from The American standard ASTM D5865 - Standard Test Method for Gross Calorific Value of Coal and Coke. First step of this derivation involves calculating HHV at constant pressure from the calorimetrically determined GCV in the bomb at constant volume. From general thermodynamic understanding of a close system, we know that no work is performed in a constant-volume bomb during the determination of calorific value. However, when a fuel is burned at constant pressure, a change in the volume of the system occurs. When fuel is burned at constant pressure and the water formed condensed to the liquid state, the volume of the system contracts. This contraction is equal to the volume of oxygen required to burn the hydrogen. Work is done on the system by the atmosphere in order to maintain a constant pressure. This work \u00F0\u009D\u0091\u0084\u00F0\u009D\u0091\u00A3\u00E2\u0086\u0092\u00F0\u009D\u0091\u009D (which will be explained in the next paragraph) has to be added to HHV at constant volume \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 to convert the HHV to HHV at constant pressure \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D, as shown in the equation below. \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D = \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 +\u00F0\u009D\u0091\u0084\u00F0\u009D\u0091\u00A3\u00E2\u0086\u0092\u00F0\u009D\u0091\u009D (A-1) When carbon in the fuel reacts with oxygen, an equal volume of carbon dioxide results and no change in volume occurs. The oxygen and nitrogen in the fuel both result in an increase in volume as neither of these two elements reacts in oxygen gas in the atmosphere. The work associated with this change in the volume of gaseous phase for the combustion reaction may be expressed as follows: \u00F0\u009D\u0091\u0084\u00F0\u009D\u0091\u00A3\u00E2\u0086\u0092\u00F0\u009D\u0091\u009D = 0.01 \u00C3\u0097 \u00F0\u009D\u0091\u0085\u00F0\u009D\u0091\u0087 \u00C3\u0097 (\u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB2 \u00C3\u0097 2.016\u00E2\u0088\u0092\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u008231.9988\u00E2\u0088\u0092\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u008128.0134) (A-2) Substituting the value of the constants, we have \u00F0\u009D\u0091\u0084\u00F0\u009D\u0091\u00A3\u00E2\u0086\u0092\u00F0\u009D\u0091\u009D = 6.148\u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB \u00E2\u0088\u0092 0.775\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 \u00E2\u0088\u0092 0.885\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081 \u00E2\u0089\u0085 6.15\u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB \u00E2\u0088\u0092 0.8 \u00C3\u0097 (\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 + \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081) (A-3) 102 where \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB, \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082and \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081 are hydrogen, oxygen and nitrogen content (d.b.). R is the universal gas constant [8.3143 J/(mol-K)] and T is the standard thermochemical reference temperature (298.15 K). The hydrogen and oxygen contributed by the moisture associated with the sample have to be excluded using the equations below. \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB = \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB,\u00F0\u009D\u0091\u009A \u00E2\u0088\u0092 0.1119 \u00C3\u0097 \u00F0\u009D\u0091\u0080 (A-4) \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 = \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082,\u00F0\u009D\u0091\u009A \u00E2\u0088\u0092 0.8881 \u00C3\u0097 \u00F0\u009D\u0091\u0080 (A-5) where 0.1119 is the ratio of molar mass of hydrogen \u00F0\u009D\u0090\u00BB2 (2.016) to molar mass of water \u00F0\u009D\u0090\u00BB2\u00F0\u009D\u0091\u0082 (18), 0.8881 is the ratio of molar mass of oxygen \u00F0\u009D\u0091\u0082 (15.9994) to molar mass of water. \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB,\u00F0\u009D\u0091\u009A is the sample hydrogen content, including the hydrogen contributed from moisture. \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082,\u00F0\u009D\u0091\u009A is the sample oxygen content, including the oxygen content contributed by moisture. Second step is to convert the \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D to low heating value LHV at constant pressure \u00F0\u009D\u0090\u00BF\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D by subtracting the energy associated with the heat of vaporization of water that originates from hydrogen content of the sample \u00F0\u009D\u0091\u0084\u00F0\u009D\u0090\u00BB as shown in equation A-6. \u00F0\u009D\u0090\u00BF\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D = \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D \u00E2\u0088\u0092 \u00F0\u009D\u0091\u0084\u00F0\u009D\u0090\u00BB = \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 +\u00F0\u009D\u0091\u0084\u00F0\u009D\u0091\u00A3\u00E2\u0086\u0092\u00F0\u009D\u0091\u009D \u00E2\u0088\u0092 \u00F0\u009D\u0091\u0084\u00F0\u009D\u0090\u00BB (A-6) By applying the basis conversion to \u00F0\u009D\u0090\u00BF\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D and then subtracting the heat of vaporization of water that originated from as-received moisture value \u00F0\u009D\u0091\u0084\u00F0\u009D\u0091\u0080 , we get the as-received NCV at constant pressure \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D as shown in equation A-7. \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D = \u00F0\u009D\u0090\u00BF\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D \u00C3\u0097100 \u00E2\u0088\u0092\u00F0\u009D\u0091\u0080100\u00E2\u0088\u0092 \u00F0\u009D\u0091\u0084\u00F0\u009D\u0091\u0080 = [(\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 + \u00F0\u009D\u0091\u0084\u00F0\u009D\u0091\u00A3\u00E2\u0086\u0092\u00F0\u009D\u0091\u009D \u00E2\u0088\u0092 \u00F0\u009D\u0091\u0084\u00F0\u009D\u0090\u00BB) \u00C3\u0097100 \u00E2\u0088\u0092\u00F0\u009D\u0091\u0080100] \u00E2\u0088\u0092 \u00F0\u009D\u0091\u0084\u00F0\u009D\u0091\u0080 (A-7) 103 The heat of vaporization of water that originates from hydrogen content of the sample \u00F0\u009D\u0091\u0084\u00F0\u009D\u0090\u00BB, and the as-received moisture value \u00F0\u009D\u0091\u0084\u00F0\u009D\u0091\u0080 are calculated using the equations below. \u00F0\u009D\u0091\u0084\u00F0\u009D\u0090\u00BB = 0.01 \u00C3\u0097 \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u00A3\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u009D \u00C3\u0097 (\u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB2.016) = 218.180 \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB [\u00F0\u009D\u0090\u00BD\u00F0\u009D\u0091\u0094] (A-8) \u00F0\u009D\u0091\u0084\u00F0\u009D\u0091\u0080 = 0.01 \u00C3\u0097 \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u00A3\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u009D \u00C3\u0097 (\u00F0\u009D\u0091\u008018.0154) = 24.5 \u00F0\u009D\u0091\u0080 [\u00F0\u009D\u0090\u00BD\u00F0\u009D\u0091\u0094] (A-9) where \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u00A3\u00F0\u009D\u0091\u008E\u00F0\u009D\u0091\u009D is the constant pressure heat of vaporization of water at 25\u00E2\u0084\u0083, which equals to 43,985 J/mol. When all the energy terms are substituted into equation A-6, the \u00F0\u009D\u0090\u00BF\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D can be expressed only in terms of calorimetrically determined \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 , hydrogen \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB , oxygen \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 and nitrogen content \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081. \u00F0\u009D\u0090\u00BF\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D = \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 + [6.15\u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB \u00E2\u0088\u0092 0.8 \u00C3\u0097 (\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 + \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081)] \u00E2\u0088\u0092 218.18 \u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB = \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 \u00E2\u0088\u0092 212\u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB \u00E2\u0088\u0092 0.8 \u00C3\u0097 (\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 + \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081) (A-10) Equation A-10 can then be expressed in terms of the same set of parameters and moisture content \u00F0\u009D\u0091\u0080 (w.b.), as provided in equation A-7. NCVp is used in equation A-11 because the calorific value is now expressed in wet basis. \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D = [\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 \u00E2\u0088\u0092 212\u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB \u00E2\u0088\u0092 0.8 \u00C3\u0097 (\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 + \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081)] \u00C3\u0097100 \u00E2\u0088\u0092\u00F0\u009D\u0091\u0080100\u00E2\u0088\u0092 24.5\u00F0\u009D\u0091\u0080 (A-11) Or, we can express equation 10 in \u00F0\u009D\u0090\u00BA\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 (w.b.) as we know that, \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u009D = [\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 \u00E2\u0088\u0092 212\u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB \u00E2\u0088\u0092 0.8 \u00C3\u0097 (\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 + \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081)] \u00C3\u0097100 \u00E2\u0088\u0092\u00F0\u009D\u0091\u0080100\u00E2\u0088\u0092 24.5\u00F0\u009D\u0091\u0080 (A-12) \u00F0\u009D\u0090\u00BA\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 = \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089\u00F0\u009D\u0091\u00A3 \u00C3\u0097100 \u00E2\u0088\u0092\u00F0\u009D\u0091\u0080100 (A-13) Therefore, the resulting equation is shown in equation A-14 and omitting the subscripts \u00E2\u0080\u009Cp\u00E2\u0080\u009D and \u00E2\u0080\u009Cv\u00E2\u0080\u009D, we get 104 \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089 = \u00F0\u009D\u0090\u00BA\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089 \u00E2\u0088\u0092 [212\u00F0\u009D\u0091\u008B\u00F0\u009D\u0090\u00BB + 0.8 \u00C3\u0097 (\u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0082 + \u00F0\u009D\u0091\u008B\u00F0\u009D\u0091\u0081)] \u00C3\u0097100 \u00E2\u0088\u0092\u00F0\u009D\u0091\u0080100\u00E2\u0088\u0092 24.5\u00F0\u009D\u0091\u0080 (A-14) The steps to convert HHV at constant volume to NCV at constant pressure are summarized in Figure A-1. Figure A-1: The steps to convert high heating value \u00F0\u009D\u0090\u00BB\u00F0\u009D\u0090\u00BB\u00F0\u009D\u0091\u0089 (dry basis) to net calorific value \u00F0\u009D\u0091\u0081\u00F0\u009D\u0090\u00B6\u00F0\u009D\u0091\u0089 (wet basis) The subscripts \u00E2\u0080\u009C\u00F0\u009D\u0091\u00A3 \u00E2\u0080\u009D and \u00E2\u0080\u009C\u00F0\u009D\u0091\u009D \u00E2\u0080\u009D, which indicates the constant volume and constant pressure thermodynamic states, are omitted during the discussion of gross calorific value and net calorific value. The gross calorific value and net calorific value will be abbreviated as GCV without the subscript \u00E2\u0080\u009C \u00F0\u009D\u0091\u00A3 \u00E2\u0080\u009D and NCV without the subscript \u00E2\u0080\u009C \u00F0\u009D\u0091\u009D \u00E2\u0080\u009D 105 A.3 Principle of bomb calorimeter Calorimetry is the science of measuring the heat (gross calorific value) of chemical reactions or physical changes, with the use of a bomb calorimeter. Modern calorimeters consist essentially of a combustion vessel (inner vessel or bomb) in which the fuel is burned in pure oxygen (technical oxygen 99.95%), an outer vessel containing a known weight of water in which the combustion vessel is immersed and thermometers for measuring the rise in temperature of the water after combustion has taken place. There are two principles or modes of operating calorimeters: (1) isoperibol and (2) adiabatic. Isoperibol calorimeter is commonly used in North America. The temperature in its outer vessel is kept constant throughout the whole experiment by applying the concept of a thermostat. There are small heat exchanges between inner and outer vessel, resulting in a small error. The error is then corrected by a correction factor. The calorimeter measures the gross calorific value based on the temperature rise in the inner vessel. The net corrected temperature rise, t is calculated in the following equation. In Parr 6100 oxygen bomb calorimeter, gross calorific value is corrected automatically using sensors and controllers. )()( 21 bcrabrttt ac \u00EF\u0080\u00AD\u00EF\u0080\u00AD\u00EF\u0080\u00AD\u00EF\u0080\u00AD\u00EF\u0080\u00AD\u00EF\u0080\u00BD Where \u00F0\u009D\u0091\u008E is time of firing. \u00F0\u009D\u0091\u008F is time (to nearest 0.1 min) when the temperature reaches 60% of the total rise. \u00F0\u009D\u0091\u0090 is time at beginning of period (after the temperature rise in which the rate of temperature change has become constant. \u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u008E is temperature of time of firing, corrected for thermometer scale error. \u00F0\u009D\u0091\u00A1\u00F0\u009D\u0091\u0090 is temperature at time c, corrected for thermometer scale error. \u00F0\u009D\u0091\u009F1 is rate (temperature units per minute) at which temperature was rising during the 5-min period 106 before firing. r2 is rate (temperature units per minute) at which the temperature was rising during the 5-min period after time c. Due to the heat release by elements such as sulphur and nitrogen, as well as heat of combustion of fuse wire, the following correction to gross calorific value must be performed to achieve higher accuracy. Corrected gross calorific value,meeeqq measuredgrVcorrectedgrV )( 321,,,, \u00EF\u0080\u00AB\u00EF\u0080\u00AB\u00EF\u0080\u00AD\u00EF\u0080\u00BD Where \u00F0\u009D\u0091\u00921 is correction for heat of formation of nitric acid (HNO3), equivalent to millilitres of 0.0725N alkali used for titration. \u00F0\u009D\u0091\u00922 is correction for heat of formation of sulphuric acid (H2SO4), equivalent to weight of sample (g) x %sulphur in sample x 14. \u00F0\u009D\u0091\u00923 is correction for heat of combustion of fuse wire, equivalent to 2.3 x centimetres of fuse wire consumed in firing During the measurement, the inner vessel constantly moves into and out of the water body for cleaning and the repetition of the experiment. Thus, some unavoidable water loss occurs during the removal of the inner vessel. The loss of water causes the temperature rise to be higher, resulting in a higher than expected gross calorific value. It is recommended that after every 10 calorimeter tests, the water is re-measured to ensure a consistent volume of water. The fouling of the soot increases the heat capacity of the inner vessel. This would cause the gross calorific value to appear higher than it should be. Parr 6100 calorimeter uses a sensitive thermistor system where a temperature monitoring capability is built into the jacket. The thermistor system ensures an accuracy of 0.001 oC which is up to the standard set by ISO (EN) 14981:2009. Maintenance is need to maintain the accuracy. 107 A.4 Classification of wood pellets Table A-9: The quality parameters as required by EN-plus certification standard (A1, A2 and B), Industrial Wood Pellet Buyer (IWPB) standard (I1, I2 and I3) and corresponding testing standards. Parameter ISO Classification of Wood Pellets Testing Standard Measure EN-plus Certification Standard I1 I2 I3 ISO EN A1 A2 B Origin of material 1.1 Forest plantation and virgin wood X X X 17225-1 1.1.1 Whole tree X 17225-1 1.1.3 Stemwood X X 17225-1 1.1.4 Logging residue X 17225-1 1.2 Industrial by-products and residue X 17225-1 1.2.1 Chemically untreated wood X X X X 17225-1 1.2.1.5 Bark 17225-1 1.3 Used wood X 17225-1 1.3.1 Chemically untreated wood X 108 Parameter Measure EN-plus Certification Standard I1 I2 I3 ISO EN A1 A2 B Diameter / Length mm 6 \u00C2\u00B1 1 6 to 8 6 to 10 6 to 12 17829 16127 Length mm 3.15 \u00E2\u0089\u00A4 L \u00E2\u0089\u00A4 40 6 L \u00E2\u0089\u00A4 40 6 L \u00E2\u0089\u00A4 40 6 L \u00E2\u0089\u00A4 40 8 L \u00E2\u0089\u00A4 50 8 L \u00E2\u0089\u00A4 50 8 L \u00E2\u0089\u00A4 50 Moisture % of weight, ar \u00E2\u0089\u00A4 10 18134-1/2/3 14774-1/2/3 Total ash % of weight, dry \u00E2\u0089\u00A4 0.7 \u00E2\u0089\u00A4 1.5 \u00E2\u0089\u00A4 3.0 \u00E2\u0089\u00A4 1.0 \u00E2\u0089\u00A4 1.5 \u00E2\u0089\u00A4 3.0 18122 14775 Ash melting behaviour oC as stated WD XXX 15370 Durability % of weight, ar \u00E2\u0089\u00A5 97.5 \u00E2\u0089\u00A5 96.5 \u00E2\u0089\u00A5 97.5 \u00E2\u0089\u00A5 96.5 17831-1 15210-1 Fines content % of weight, ar \u00E2\u0089\u00A4 1.0 \u00E2\u0089\u00A4 4.0 \u00E2\u0089\u00A4 5.0 \u00E2\u0089\u00A4 6.0 18846 Particle size distribution % of weight >99% (3.15 mm) >98% (3.15 mm) >97% (3.15 mm) 17830 16126 >95% (2.0 mm) >90% (2.0 mm) >85% (2.0 mm) >60% (0.1 mm) >50% (0.1 mm) >40% (0.1 mm) Additives % of weight, ar \u00E2\u0089\u00A4 2.0 Max 2% of weight Net calorific value, constant pressure, ar MJ/kg GJ/tonne \u00E2\u0089\u00A5 16.5 \u00E2\u0089\u00A5 16.3 \u00E2\u0089\u00A5 16.0 \u00E2\u0089\u00A5 16.5 18125 14918 Bulk density kg/m3, ar \u00E2\u0089\u00A5 600 17828 15103 Bulk temperature oC \u00E2\u0089\u00A4 60.0 \u00C2\u00B1 1 Nitrogen % of weight, dry \u00E2\u0089\u00A4 0.3 \u00E2\u0089\u00A4 0.5 \u00E2\u0089\u00A4 1.0 \u00E2\u0089\u00A4 0.3 \u00E2\u0089\u00A4 0.6 16948 15104 Sulphur % of weight, dry \u00E2\u0089\u00A4 0.03 \u00E2\u0089\u00A4 0.04 \u00E2\u0089\u00A4 0.15 \u00E2\u0089\u00A4 0.2 \u00E2\u0089\u00A4 0.4 16994 15289 109 Parameter Measure EN-plus Certification Standard I1 I2 I3 ISO EN A1 A2 B Chlorine % of weight, dry \u00E2\u0089\u00A4 0.02 \u00E2\u0089\u00A4 0.03 \u00E2\u0089\u00A4 0.03 \u00E2\u0089\u00A4 0.05 \u00E2\u0089\u00A4 0.1 Elemental components in mg/kg substance of raw biomass material Arsenic mg/kg dry \u00E2\u0089\u00A4 1 \u00E2\u0089\u00A4 2 16968 15297 Cadmium mg/kg dry \u00E2\u0089\u00A4 0.5 \u00E2\u0089\u00A4 1 Chromium mg/kg dry \u00E2\u0089\u00A4 10 \u00E2\u0089\u00A4 15 Copper mg/kg dry \u00E2\u0089\u00A4 10 \u00E2\u0089\u00A4 20 Lead mg/kg dry \u00E2\u0089\u00A4 10 \u00E2\u0089\u00A4 20 Mercury mg/kg dry \u00E2\u0089\u00A4 0.1 \u00E2\u0089\u00A4 0.1 Nickel mg/kg dry \u00E2\u0089\u00A4 10 Zinc mg/kg dry \u00E2\u0089\u00A4 100 \u00E2\u0089\u00A4 200 "@en . "Thesis/Dissertation"@en . "2015-02"@en . "10.14288/1.0135651"@en . "eng"@en . "Chemical and Biological Engineering"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "Attribution-NonCommercial-NoDerivs 2.5 Canada"@en . "http://creativecommons.org/licenses/by-nc-nd/2.5/ca/"@en . "Graduate"@en . "Calorific value of wood pellets"@en . "Text"@en . "http://hdl.handle.net/2429/51791"@en .