UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Bio-oil upgrading through biodiesel emulsification and catalytic vapour cracking Yu, Joyleene Ruth 2014

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

Item Metadata

Download

Media
24-ubc_2014_september_yu_joyleene.pdf [ 6.74MB ]
Metadata
JSON: 24-1.0167466.json
JSON-LD: 24-1.0167466-ld.json
RDF/XML (Pretty): 24-1.0167466-rdf.xml
RDF/JSON: 24-1.0167466-rdf.json
Turtle: 24-1.0167466-turtle.txt
N-Triples: 24-1.0167466-rdf-ntriples.txt
Original Record: 24-1.0167466-source.json
Full Text
24-1.0167466-fulltext.txt
Citation
24-1.0167466.ris

Full Text

 BIO-OIL UPGRADING THROUGH BIODIESEL EMULSIFICATION AND CATALYTIC VAPOUR CRACKING  by  JOYLEENE RUTH YU  B.A.Sc., The University of British Columbia, 2011  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)  May 2014  © Joyleene Ruth Yu, 2014      ii Abstract With our limited fuel supplies struggling to keep up with our ever-increasing demand for energy, and the rising trend towards sustainable and cleaner technologies, the need to harness the potential of bio-oil as an alternative source of energy has never been more compelling. Although crude bio-oil can already be utilized to supplement heating oils and boiler fuels, its greater value lies in its potential as a source of transportation fuels and chemicals after upgrading.  In collaboration with Diacarbon Energy Inc., the main objectives of this project were twofold: (1) investigating the effect of extraction location from their proprietary pyrolysis unit on crude bio-oil quality prior to its emulsification with biodiesel, and characterizing the resulting biodiesel- and lignin-rich layers; and (2) designing and building a catalytic test unit to perform in situ cracking of slow pyrolysis vapours. Experimental results confirmed that extraction location does affect the crude bio-oil quality. The effect of the surfactant on the emulsification was minimal as the resulting biodiesel-rich layer from the emulsification without the surfactant showed similar improvements in terms of water content, viscosity, TAN and HHV. A water mass balance confirmed that the majority of the water (~97%) is retained in the lignin-rich phase after emulsification. This is significant because the solvency of biodiesel can be utilized to upgrade bio-oils by selectively extracting its desirable fuel components into a biodiesel-rich phase, which can then be easily separated from the lignin-rich phase where the higher molecular weight compounds, such as pyrolytic lignin, as well as the majority of the water, are retained.  The bio-oil samples obtained from the non-catalytic and catalytic vapour cracking experiments separated into two distinct layers – an aqueous and organic layer. While the aqueous layers were fairly similar in nature, the organic layer from the catalytic experiment showed a significant decrease in viscosity (94.3% less) and water content (64.3% less). The organic layer from the catalytic pyrolysis remained homogeneous while that from the non-catalytic pyrolysis split into a hazy aqueous layer (with suspended oil droplets) sandwiched between a thin organic layer on top and a thicker organic layer at the bottom.       iii Preface All of the work presented in this thesis was carried out and completed by the author under the supervision of Dr. Naoko Ellis in the Department of Chemical and Biological Engineering at the University of British Columbia.  Results from Chapter 3 were presented on October 15th, 2012 at the 62nd Canadian Chemical Engineering Conference as part of the International Symposium on Biomass and Bioenergy held in Vancouver, BC, as well as during the poster session for Research Day 2013 hosted by the Department of Chemical and Biological Engineering at the University of British Columbia on October 9, 2013. Parts of Chapters 2 and 3 of this thesis are being included in a manuscript entitled, “Bio-oil Upgrading through Biodiesel Emulsification: Surfactant-free Mixtures” (Yu, J. and Ellis, N.), that is currently being prepared for journal submission.      iv Table of Contents Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iii Table of Contents ......................................................................................................................... iv List of Tables ............................................................................................................................... vii List of Figures ............................................................................................................................... ix List of Symbols and Abbreviations ............................................................................................. x Glossary ........................................................................................................................................ xi Acknowledgements ..................................................................................................................... xii Dedication ................................................................................................................................... xiv Chapter 1 Introduction ............................................................................................................... 1 1.1 The State of Our Energy Economy ..................................................................................... 1 1.1.1 The Cost of Petroleum and Why We Need Cleaner Energy Sources ....................................... 1 1.1.2 Trends and Opportunities in Renewable Energy ...................................................................... 2 1.2 Biomass Pyrolysis ............................................................................................................... 3 1.2.1 Biomass and Its Major Components ......................................................................................... 3 1.2.2 The Pyrolysis Process ............................................................................................................... 5 1.2.3 Types of Pyrolysis .................................................................................................................... 6 1.3 Bio-oil ................................................................................................................................. 8 1.3.1 Composition and Properties ...................................................................................................... 8 1.3.2 Practical Applications ............................................................................................................. 11 1.3.3 Upgrading Methods/Technologies .......................................................................................... 12 1.4 Project Justification and Scope of Work .......................................................................... 13 Chapter 2 Materials and Methodology ................................................................................... 18 2.1 Bio-oil Sample Collection, Storage and Handling ........................................................... 18     v 2.2 Chemicals and Other Materials ........................................................................................ 19 2.3 Sample Homogenization ................................................................................................... 19 2.4 Characterization Methods ................................................................................................. 19 2.4.1 Water Content ......................................................................................................................... 19 2.4.2 Total Acid Number ................................................................................................................. 20 2.4.3 Density and Viscosity ............................................................................................................. 20 2.4.4 Elemental Analysis and Energy Content (High Heating Value) ............................................ 21 Chapter 3 Emulsification with Biodiesel ................................................................................. 22 3.1 Crude Bio-oil Characterization (Pre-emulsification) ....................................................... 22 3.2 The Emulsification Process .............................................................................................. 23 3.3 Experimental Results and Discussion ............................................................................... 27 3.3.1 Biodiesel-rich Phase (Upper Layer) ....................................................................................... 28 3.3.2 Lignin-rich Phase (Bottom Layer) .......................................................................................... 30 3.3.3 Water Mass Balance ............................................................................................................... 33 Chapter 4 Design and Construction of Catalytic Reactor Test Unit .................................... 35 4.1 Objectives and Overall Design Considerations ................................................................ 35 4.2 Process Design Details ..................................................................................................... 36 4.3 Safety Considerations ....................................................................................................... 39 4.4 Catalyst Selection and Incorporation ................................................................................ 40 4.5 Experimental Set-up ......................................................................................................... 42 4.6 Experimental Results and Discussion ............................................................................... 46 Chapter 5 Conclusions and Recommendations ...................................................................... 52 References .................................................................................................................................... 54 Appendix A:  Experimental Data and Results ......................................................................... 62 A.1 Sample Data and Results for Biodiesel Emulsification Experiments ................................. 62     vi A.2 Sample Data and Results for Catalytic Reactor Experiments ............................................ 74 Appendix B: Catalytic Reactor Design Supplementary Documentation ............................... 82 B.1 Experimental Procedure ...................................................................................................... 82 B.2 Legend for Process Flow Diagram ..................................................................................... 85 B.3 HAZOPS ............................................................................................................................. 87 B.4 Emergency Shut Down ....................................................................................................... 89 B.5 Test Unit Diagrams and Photos .......................................................................................... 89 Appendix C: Sample Calculations ............................................................................................. 94         vii List of Tables Table 1-1. Typical range of operating parameters and product yields for the different pyrolysis processes – may still vary considerably depending on feedstock, process and actual operating conditions.17,22–24 ..................................................................................................... 7 Table 1-2. Summary of commonly used commercially available catalysts. ................................. 16 Table 3-1. Comparison of the TAN of the bio-oil samples obtained from the different extraction points on Diacarbon’s pyrolysis unit; the error range indicates the standard deviation for triplicate measurements of the same sample. ........................................................................ 22 Table 3-2.Summary of density and dynamic and kinematic viscosities for the bio-oil samples obtained from different extraction points; the error range indicates the standard deviation for triplicate viscosity and at least triplicate density measurements of the same sample. .......... 22 Table 3-3. Comparison of water content in the bio-oil samples obtained from the different extraction points on Diacarbon’s pyrolysis unit; the error range indicates the standard deviation for triplicate measurements of the same sample. .................................................. 23 Table 3-4. Preparation methods for bio-oil/biodiesel mixture production ................................... 24 Table 3-5. Comparison of stability and volume ratios of the biodiesel-rich and lignin-rich layers after emulsification; the error range indicates the standard deviation for triplicate measurements of the same sample. ....................................................................................... 27 Table 3-6. Comparison of the total acid number and water content for the samples – pre- and post-emulsification; the error range indicates the standard deviation for triplicate measurements of the same sample. ....................................................................................... 28 Table 3-7. Comparison of the density and dynamic and kinematic viscosities for the crude bio-oil and biodiesel-rich layers after emulsification; the error range indicates the standard deviation for at least triplicate measurements of the same sample. For detailed viscosity results, refer to Tables A22-24. ............................................................................................. 29 Table 3-8. Elemental analysis and calculated HHVs for crude and emulsified samples. ............. 30     viii Table 4-1. Comparison of the water content for the top and bottom layers from the non-catalytic and catalytic experiments; the error range indicates the standard deviation for quadruplicate measurements of the same sample. ....................................................................................... 51 Table 4-2. Summary of density and dynamic and kinematic viscosities for the top and bottom layers obtained from the non-catalytic and catalytic experiments; the error range indicates the standard deviation for triplicate viscosity measurements and at least triplicate density measurements of the same sample. ....................................................................................... 51        ix List of Figures Figure 2-1. Simplified schematic diagram of Diacarbon's pyrolysis unit. .................................... 18 Figure 3-1. Results from the different emulsification methods after one week ............................ 25 Figure 3-2. Distinct layers shown during emulsification using Dynamotive bio-oil. ................... 31 Figure 3-3. Lignin-rich phase after emulsification of Diacarbon’s bio-oil. .................................. 31 Figure 3-4. Lignin extraction using bio-oil from Dynamotive. .................................................... 32 Figure 3-5. Diacarbon bio-oil on filter surface after water precipitation method. ........................ 33 Figure 3-6. Distribution of water among the components: (a) pre-emulsification; and (b) post-emulsification. ....................................................................................................................... 34 Figure 4-1. Pre-assembled tubes with nuts and ferrules attached (these tubes have already been used, hence the discoloration). .............................................................................................. 37 Figure 4-2. Pre-formed zeolite pellets (left) and SiC granules (right). ......................................... 42 Figure 4-3. Process flow diagram for catalytic reactor unit .......................................................... 45 Figure 4-4. Vapour build-up inside the collection flask during the experiment. .......................... 46 Figure 4-5. Samples collected from the non-catalytic (left) and catalytic (right) experiments shown in 600-mL jars. .......................................................................................................... 47 Figure 4-6. Residue left behind in collection flask after the contents were transferred. .............. 48 Figure 4-7. Samples extracted from the aqueous (top) and organic (bottom) layers of the non-catalytic (SiC only) experiment collected into 118-mL jars. ................................................ 49 Figure 4-8. Samples extracted from the aqueous (top) and organic (bottom) layers of the catalytic experiment (with HZSM-5) collected into 118-mL jars. ...................................................... 49 Figure 4-9. The organic (bottom) layers from the non-catalytic (left) and catalytic (right) experiments, ~10 minutes after being extracted from their respective sample mixtures. ..... 50     x List of Symbols and AbbreviationsACS  American Chemical Society ASTM American Society for Testing and Materials CTA International Centre for Technology Assessment DET dynamic equivalence-point titration DN diameter nominal, mm EBRI European Bioenergy Research Institute GC-MS  Gas Chromatography-Mass  Spectroscopy GHSV  gas hourly space velocity, hr-1 HAZOP Hazard and Operability HDPE   high-density polyethylene HHV  high heating value, MJ/kg HPLC High Performance Liquid Chromatography IEA  International Energy Agency ID  inner diameter, mm KOH  potassium hydroxide NRC  National Research Council MDF  medium density fibreboard MET monotonic equivalence-point titration OD  outer diameter, mm OPEC Organization of the Petroleum Exporting Countries PE   polyethylene PP   polypropylene PTFE   polytetrafluoroethylene SiC  silicon carbide SPF  spruce-pine-fir TAN total acid number, mg KOH/g sample TBN total base number, mg KOH/g sample WEC  World Energy Council ZSM-5  Zeolite Socony Mobil-5                xi Glossary Atomization – in automotive systems, the process of forcing fuel through a nozzle under high pressure to create a fine misted spray Cetane Number – refers to the combustion quality of diesel fuel, it represents the time delay between the start of the injection process and the point when the fuel ignites; the value is based on the vol.% of cetane that provides the identical ignition time delay of the fuel sample Energy equity – the accessibility and affordability of energy supply across the population Energy security – the effective management of primary energy supply from domestic and external sources, the reliability of energy infrastructure, and the ability of energy providers to meet current and future demand Environmental sustainability – the achievement of supply and demand-side energy efficiencies and the development of energy supply from renewable and other low-carbon sources Structural water – water that cannot be removed through evaporation          xii Acknowledgements I would like to express my gratitude to my supervisor, Dr. Naoko Ellis, for supporting me personally and professionally in my research, for challenging me to think, for pushing me outside my comfort zone and for allowing me the freedom to stumble as I sought out answers under her guidance. To my supervisory committee – Dr. Jared Taylor and Dr. Glenn Sammis, thank you for giving of your time and expertise to supervise my work and review my thesis.  To the past and present members of my Biofuels Research Group –Steve Reaume, Soojin Lee, Jidon Janaun and especially Amir Dehkhoda – thank you for taking me under your wing, for sharing your wisdom and experience and the gift of your friendship. To my research assistant, Manu Suvarna, although your time in the lab was short, thank you for helping me run my experiments and for keeping me company during those long days in the lab. To my colleagues and friends around the department, thank you for being generous with your time, expertise, resources and advice, your doors always seemed to be open when I needed it. This project would not have been possible without the help of the many people who work tirelessly behind the scenes to make things happen. To the workshop crew – Doug Yuen, Gordon Cheng, David Roberts, Serge Milaire and Alex Thng– thank you for being patient with me as I fumbled around trying to figure things out, for being sounding boards, problem solvers and miracle workers, in more ways than one. To the safety committee – Marlene Chow and Ivan Leversage– thank you for always keeping my safety and well-being in mind, and especially for helping me clean up my messes. To Richard Ryoo, the one-man army in the shipping department, thank you for always making sure I was able to get what I needed when I needed it. To the administration department – Lori Tanaka, Helsa Leong, Amber Lee and Joanne Dean, thank you for taking care of the not-so-little things like finances, paperwork, etc. so that I could focus on my research.   To my parents, Thomas and Ruthia, thank you for your unconditional love, support and encouragement, then, now and for always. To my brother, Donald, your believing in me always enabled me to go further than I thought I could. To God, my heavenly Father, thank you for being with me through all the ups and downs, for picking me up when I was down, for giving me     xiii that extra push to get things done and for the wisdom to see and understand things that I would otherwise not.  I would also like to acknowledge NSERC and MITACS for the funding of this project and for the Department of Chemical and Biological Engineering as well as the University of British Columbia for their financial support.       xiv Dedication      For my parents,  for instilling in me  a near-bottomless thirst for knowledge,  and for my brother,  who lights up my life unlike any other.     1 Chapter 1 Introduction 1.1 The State of Our Energy Economy 1.1.1 The Cost of Petroleum and Why We Need Cleaner Energy Sources The economics of petroleum is a rather complex affair; being the dominant source of energy for the better part of the last 65 years, the influence of crude oil on international politics and strategies has been keenly felt, perhaps more so than any other commodity – often proving to be the source of instability, dispute and even war.1 Indeed, as Herrmann, et al. so aptly put it, “…the modern day oil industry is a remarkable amalgam[ation] of politics, economics, science and technology, with the companies involved often achieving enormous profits and [an inordinate amount of influence] in the nations in which they are based.”1  The figures we see when we gas up at the pump or when we receive our utility bills are due to more than global supply and demand, they are a result of various factors such as federal and state (or provincial) taxes, costs from refining, distribution and marketing, the price of crude oil, the commodities market and even seasonal conditions. However, these prices fail to reflect the true cost of fossil fuels. The National Research Council (NRC) estimates that the hidden costs of energy production and use in the United States amounted to $120 billion in 2005, and that only reflects the impact on public health from air pollution associated with electricity generation and motor vehicles. It does not include damages from climate change, ecosystem destruction or even air pollutants such as mercury,2 much less the amount of taxpayer dollars that go into cleaning up the industry’s accidents and disasters. According to a report published by the International Centre for Technology Assessment (CTA) in 1998, if we accounted for externalities such as: (1) tax subsidization of the oil industry; (2) government program subsidies; (3) protection costs involved in oil shipment and motor vehicle services; (4) environmental, health and social costs; and (5) other external costs due to travel delays, car accidents and insurance losses, we would be paying an estimated $10-$12 more per gallon at the pump.3  With the recent application of combined horizontal drilling and hydraulic fracturing technologies to North American shale gas reserves, it would seem that ‘peak oil’, and the accompanying global energy crisis, has been postponed once again, perhaps by a few years or until some more     2 advanced technology comes along. Like its predecessor – the development of the tar sands, this unexpected boost to oil supplies is largely controversial,4–6 largely due to similar issues in water management, health and environmental concerns. While proponents and detractors each have valid arguments for and against the continued exploitation of our fossil fuel supplies, it does not change the fact that our supply of fossil fuels is limited, and it is not so much a matter of if we will exhaust these resources, but when, and at what cost to our environment and future generations. At present, 36% of all the energy in the United States is derived from petroleum, but the alarming statistic is that it uses 141% of its total petroleum production for transportation alone.7 With the energy consumption of countries such as China and India growing at an accelerated rate, the need for developing sustainable energy sources only becomes even more compelling. Unfortunately, developing the technology to harness and/or create sustainable energies is only one piece of the puzzle. Recent reports released by the World Energy Council (WEC) have identified the need to address what they refer to as the “energy trilemma” – the triple challenge of finding solutions to support the three key aspects of energy security, energy (or social) equity and environmental sustainability.8  1.1.2 Trends and Opportunities in Renewable Energy The International Energy Administration (IEA) forecasts a 56% increase in world energy consumption between 2010 and 2040, with the industrial sector consuming about half of our global delivered energy. Even with current policies and regulations limiting the use of fossil fuels, worldwide energy-related CO2 emissions are still predicted to rise from 31 billion metric tons in 2010 to 36 billion metric tons in 2020 and further up to 45 billion metric tons in 2040, a 46% increase altogether.9 Fortunately, renewable energy is increasingly becoming a larger part of the global energy mix and is projected to grow to 18% in 2035, provided that the necessary support measures for deployment remain in place – the sector’s continued growth hinges on subsidies to facilitate deployment and drive further cost reductions.10  In recent years, the requirements of the European Union’s Renewable Energy Directive and mandatory national targets (20% renewables by 2020) have caused the rapid expansion of renewable power generation in Europe, particularly wind and solar; the market for renewables in the United States has also been growing strongly, in large part due to the continuation of     3 stimulus policies directed at renewable energy, such as the provision of cash grants of up to 30% of investment costs for eligible renewable energy projects (US Treasury 1603 Program).10 However, the majority of our global energy infrastructure is still dependent on coal, oil and natural gas. With fuel supplies struggling to keep up with increasing energy demands, the upward trend in fuel prices shows no sign of abating anytime soon, even with the recent market fluctuations. Add to that increasing shifts towards sustainability and cleaner technologies, the need for alternative sources of energy only becomes more compelling. The ecological advantages of biomass utilization in conjunction with oil prices and reserves and the political instability that characterizes many of the oil-producing countries have made biomass an integral part of our primary energy sources.11 Whilst the research community continues to explore different options, bio-oil, the liquid product obtained from the pyrolysis of biomass, has been emerging as one of the more viable avenues for sustainable energy. Although crude bio-oil can already be utilized as an alternative to augment heating oils and boiler fuels, it has the added potential of being used for producing transportation fuels and chemicals after upgrading. Having been sourced from biomass feedstock, depending on the efficiency with which bio-oil is produced and used, it can be considered a carbon-neutral fuel. This growing interest in bio-oil production is also due to the possibilities of decoupling liquid fuel production (on a spatial and temporal basis) from utilization as well as the on-site separation of minerals from the liquid fuel for use towards soil amendment.12 1.2 Biomass Pyrolysis 1.2.1 Biomass and Its Major Components Biomass generally refers to plant-derived organic matter; in the context of energy sources it can refer to wood, agricultural or energy crops, wood and crop residues as well as animal and municipal solid wastes. The value of any type of biomass is dependent on its physical and chemical properties and mankind has long been utilizing it for food, fibre, fuel as well as building material. Even coal and oil can be considered as biomass, albeit the fossilized kind, but since it takes millions of years for biomass to be converted into fossil fuels, they cannot be     4 considered renewable within a timeframe that can accommodate our rate of consumption.13 While biomass used for the first-generation of biofuels came primarily from food crops such as wheat, sugarcane, and corn, raising concerns over food shortages as crops were diverted towards fuel supplies, second-generation biofuels are sourced from lignocellulosic feedstocks (e.g., wood, energy crops, residues and organic waste), which have a lesser impact on food security and land use. Currently, biomass makes up about 10% of all energy consumed worldwide, but its consumption varies greatly among nations, with industrialized countries like the United States relying on it for only about 3% of their energy needs.14  Biomass is composed primarily of cellulose, hemicellulose and lignin, with various other compounds such as proteins, acids, salts and minerals – essentially non-structural materials that are collectively referred to as extractives (e.g., waxes, essential oils, resins), present in much lower concentrations. Actual biomass compositions can be so highly variable that handbooks on biomass properties have been published and searchable databases can be found online (e.g. http://www.ecn.nl/phyllis). Vassilev et al.15 have identified various factors that affect biomass composition, with some being more important (e.g. plant species) than others (e.g. growing, harvesting time, transport and storage conditions).  Cellulose, a glucose polysaccharide connected linearly by !-1,4-glycoside linkages, typically comprises 40-50 wt.% of dry biomass. Its basic building block is cellobiose, a glucose-glucose dimer; the aggregation of the linear cellulose chains provides a crystalline structure that stabilizes the plant cell walls and makes it resistant to chemical attack. Unlike cellulose, hemicellulose is amorphous and is a shorter, highly branched polymer composed of five-carbon (xylose and arabinose) and six-carbon sugars (glucose, mannose and galactose), accounting for 20-40 wt.% of the biomass. Lignin is a polyphenolic polymer bound together by ether and carbon-carbon bonds that constitutes 15-25wt.% of the lignocellulose source. It is considered an amorphous cross-linked resin with no exact structure and is the main binder for agglomerating the fibrous cellulosic components as well as shielding them against rapid microbial or fungal destruction.      5 1.2.2 The Pyrolysis Process Pyrolysis is the thermal decomposition of organic material that occurs in the absence of oxygen or when oxygen levels are simply insufficient to achieve combustion.16,17 This is unlike the gasification process where the oxygen supply is carefully controlled in order to produce syngas. Although the application of pyrolysis to wood can be traced back to ancient Egyptian times when it was used for making tar pitch to use as waterproofing for boats and embalming dead bodies, most of the research done in the area only started a little over 40 years ago. Interest in biomass as an energy source was renewed in the 1970s when oil production in the US supposedly ‘peaked’ and fuel prices increased significantly as a result of the OPEC embargo.18 Over the last two decades, fundamental research has shown that high yields of liquids and gases, which could be used as sources of valuable chemicals, chemical intermediates, petrochemicals and fuels, could be obtained from biomass; thus, the focus has shifted from developing the traditional char-producing pyrolysis towards the development of technology to produce the more valuable syngas, bio-oil or chemicals from fast pyrolysis.17 The pyrolysis process has three main products – solid biochar, non-condensable gases and condensable vapours (which result in bio-oil), the proportions of which are dependent on the process conditions. Physically, the changes that occur during pyrolysis can be broken down as follows:19  (1) Heat is transferred from the heat source, increasing the temperature inside the feedstock;  (2) Primary pyrolysis reactions at this heightened temperature are initiated, leading to the release of volatiles and char formation; (3) Hot volatiles flow towards cooler solid particles causing heat transfer between them; (4) Some of the volatiles start to condense amidst the cooler particles, producing tar; (5) Autocatalytic secondary pyrolysis reactions take place simultaneously with the primary pyrolysis reactions; and (6) Further thermal decomposition, reforming, water gas shift reactions, recombination of radicals, and dehydration can also occur – depending on residence time, temperature or pressure profile of the system.     6 The chemical changes that occur are not as well understood, and may be considered to be a combination of the reactions and cross-reactions occurring in all three of the main biomass constituents – cellulose, hemicellulose and lignin. Cellulose degradation occurs from 240-350°C, producing anhydrocellulose and levoglucosan – at temperatures less than 300°C, depolymerisation is the dominant process, at temperatures above 300°C, char, tar and gaseous products are formed.19 Hemicellulose decomposes at 200-260°C and yields more volatiles but lesser char and tar compared to cellulose; while, lignin decomposition occurs from 280-500°C and produces more residual char than cellulose.17 According to a pathway proposed by Lange,20 at low temperatures (<200°C) the carbohydrates depolymerize resulting in formation of smaller units; at ~300°C this depolymerization is accompanied by slow dehydration to unsaturated species, which can undergo subsequent reactions to form unsaturated polymers and char and at higher temperatures ("700°C) C-C and C-H bonds cleavage is severe, producing smaller oxygenates and non-condensable gases (CO, CO2, H2, and CH4). Characteristics of wood pyrolysis products are also dependent on the hardwood or softwood species; however, the terms hardwood and softwood, which actually refer to angiosperms (e.g. oak, maple, birch) and gymnosperms (e.g. pine, spruce, fir), respectively, can be misleading, as they have little to do with the hardness of the wood, but instead indicate the seed structure of the trees. 1.2.3 Types of Pyrolysis Pyrolysis processes are typically differentiated based on operating conditions, particularly temperature, heating rate and residence time. While the processes are usually divided into three main categories: slow, fast and intermediate pyrolysis, the terms are somewhat arbitrary and are poor indicators of actual residence times or heating rates; many pyrolysis processes have been conducted in a range between the two extremes.17 Generally speaking, lower process temperatures and longer residence times favour char production; high temperatures and longer residence times increase the production of non-condensable gases; while moderate temperatures (400-500°C) and short vapour residence times (0.5 – 2 s) maximize the production of the liquid bio-oil.21 A summary of the typical operating conditions and product yields for the different pyrolysis processes is shown in Table 1-1.     7 Table 1-1. Typical range of operating parameters and product yields for the different pyrolysis processes – may still vary considerably depending on feedstock, process and actual operating conditions.17,22–24  Pyrolysis Type Operating Temperature (°C) Heating Rate (°C/s) Vapour Residence Time Product Yield (wt.%) Char Liquid Gas Slow 400 - 600 0.1 - 2 5 – 30 min ~35 ~30 ~35 Fast 300 - 750 #1000 1 – 2 s 15-25 60-75 10-20 Intermediate 300 - 700 100 - 500 10 – 30 s 20-25 50-55 30-35  Slow Pyrolysis Slow (or conventional) pyrolysis, sometimes referred to as carbonization, has been used for thousands of years to produce charcoal. It is characterized by slow heating rates (0.1 to 2°C/s), temperatures of ~500°C and longer vapour residence times (usually 5-30 min, and sometimes up to several hours) with the main objective of maximizing the production of biochar, which has been found to be particularly valuable for soil amendment and carbon sequestration.25 Moisture content and particle size are generally less important although some continuous systems do specify a certain size reduction and drying time to optimize results.26 The typical product distribution is approximately 35 wt.% char, 35 wt.% gas and 30 wt.% liquid bio-oil.  Fast Pyrolysis In fast pyrolysis, the liquid yield is maximized by preventing the primary decomposition products from being: (1) cracked thermally or catalytically (by char that has already formed) to small non-condensable gas molecules; or (2) recombined or polymerized into char.12 To address the critical issue of raising the temperature of the biomass particle to the optimum process temperature as quickly as possible and minimize its exposure to lower temperatures that favour the formation of char particles, the following conditions are necessary:27 (1) the use of very high heating and heat transfer rates, which usually calls for finely ground biomass feed; (2) careful maintenance of pyrolysis temperatures within the 425-500°C range; (3) short vapour residence times (typically  <2 s); and (4) rapid cooling of vapours and aerosols to produce bio-oil.     8 Depending on the feedstock, the process can produce about 60-75 wt.% liquid bio-oil, 15-25 wt.% biochar and 10-20 wt.% non-condensable gases. Variations of this process, referred to as flash or ultra pyrolysis, with heating rates up to 10000°C/s, high reaction temperatures (900-1300°C) and residence times of <0.5 s have also been reportedly developed, and under some conditions, very minimal to no char is formed.28 Much has been written on the development of fast pyrolysis and its variants over the last three decades, particularly by Bridgwater,29 Butler et al.,30 and Vendersboch & Prins.31 Intermediate Pyrolysis Compared to fast pyrolysis, the heating rates for intermediate pyrolysis are much lower, usually in the range of 100-500°C/min, which leads to a decrease in tar formation since the chemical reactions are more controlled instead of simply just thermal cracking.24 Intermediate pyrolysis is considered a modified method between slow and fast pyrolysis and literature on the process is relatively sparse compared to the aforementioned methods discussed above. Research and development on the process is mostly linked to the Halocean rotary kiln reactor24 and the Pyroformer developed by the European Bioenergy Research Institute (EBRI) at Aston University.32 1.3 Bio-oil 1.3.1 Composition and Properties In its raw form, bio-oil is a dark and viscous liquid, with an elemental composition fairly close to the biomass from which it originated; its colour may range from dark reddish brown to jet-black depending on chemical composition as well as amount of char present. It also has a distinct acridly smoky smell due to the presence of low molecular weight aldehydes and other volatile organics21 that tends to linger, which can irritate the eyes and nose upon prolonged exposure. While most bio-oils appear fairly homogeneous, it has been reported that those produced from biomass containing substantial amounts of extractives exhibit some degree of phase separation – featuring an extractive-rich top phase and a bottom phase resembling normal bio-oil, although this is quite uncommon given that the majority of biomass contains very low amounts of extractives, which are dispersed throughout the bio-oil matrix.33 A two-phase product with a     9 large aqueous phase and a viscous oily phase may also be obtained if feedstock moisture levels are  >10 wt.% or if alkaline metals such as potassium, which have been shown to catalyze pyrolysis reactions, are present.34  Bio-oil can be considered to be a complex mixture of water-soluble derivatives of cellulose and hemicellulose degradation and water-insoluble pyrolytic lignin.21 According to Garcìa-Pérez et al.,35 the complexity of its multiphase structure can be attributed to the presence of char particles, waxy materials, aqueous droplets, droplets of different natures and micelles formed of heavy compounds in a matrix of hollocellulose-derived compounds and water. Of the numerous organic compounds present, more than 300 have been identified and classified into the following general categories:36 (1) hydroxyaldehydes, (2) hydroxyketones, (3) sugars and dehydrosugars, (4) carboxylic acids, and (5) phenolic compounds. So far only about 40 wt.% of the compounds in bio-oil is detectable by GC-MS and 15-20 wt.% is detectable by HPLC (High Performance Liquid Chromatography).11  Water and Oxygen Content Given its high structural water content and highly viscous and acidic nature, bio-oil is a relatively difficult fuel to burn. Although the significant water content (ranging from 15 to 35 wt.%) lowers its viscosity – facilitating transport, pumping and atomization, it also reduces its higher heating value (HHV) to < 19 MJ/kg compared to 42-44 MJ/kg for conventional fuel oils,31 thereby limiting its application as a fuel substitute without prior upgrading. The presence of water has been shown to have a positive effect on phase stability, but once the water content exceeds ~30 wt.%, phase separation occurs.37  Bio-oil’s high oxygen content (35-40 wt.%) can be attributed to the large number of oxygenated compounds that make up the majority of the more than 300 compounds that have been identified in it. The distribution of these compounds is largely dependent on biomass type and process severity – temperature, residence time and heating rate profiles, increasing the severity of the pyrolysis reduces the yield of the organic liquid due to increased vapour cracking and gas formation but produces less oxygen in the final product.38 This oxygen content is one of the major obstacles to utilizing bio-oil as a substitute for conventional fuels, as it lowers its energy density and makes it immiscible with hydrocarbon fuels.      10 Once pyrolysis liquids have condensed, they can no longer be completely re-vapourized to evaporate the water or distil off the lighter fractions. The liquids’ complex composition results in a wide range of boiling point temperatures; during distillation, the slow heating induces the polymerization of some reactive components, and boiling starts below 100°C, stopping between 250–280°C, and leaving behind 35–50 wt.% of solid residue together with volatile organic compounds and water.21,39 Viscosity and Stability  The dynamic viscosity of bio-oil generally ranges between 0.025-1Pa$s (at 40°C), which is fairly high when compared with crude oil and diesel fuel, an important consideration especially when it comes to designing pumping systems and pipelines. High viscosities are generally undesirable in fuel systems since they can cause poor atomization leading to incomplete combustion; the higher injection pressure required could also cause accelerated wear to the injector pump system.40  Since bio-oil is formed by the rapid quenching of vapours and aerosols to prevent further reactions, it is not as stable as conventional liquid fuels and its physical and chemical characteristics slowly change over time – the process, referred to as ‘aging’, is marked by a gradual but noticeable increase in viscosity due to polymerization reactions. At room temperature, the aging process can occur over months or years, depending on the initial quality of the bio-oil as well as the type of feedstock and may be accelerated at elevated temperatures (>50°C);29 the presence of char also appears to catalyze the polymerization reactions during storage. The physical and chemical aging mechanisms affecting the storage stability of bio-oils has been extensively reviewed by Diebold.41  Acidity Bio-oil’s high concentration of carboxylic acids gives it a pH value between 2-3 and a total acid number (TAN) that ranges between 50-100 mg KOH/g bio-oil, making it highly corrosive particularly to carbon steel, cast iron, mild steel, aluminium, and copper.11 Stainless steels (e.g. 304L, 316L, and 430), glass and many plastics like PTFE (polytetrafluoroethylene), PE (polyethylene), PP (polypropylene), HDPE (high-density polyethylene) and polyester resins have been found to be very resistant to bio-oils, making them suitable as materials of construction for storage, transportation and sampling purposes.42      11 1.3.2 Practical Applications Bio-oil’s aforementioned properties have made it challenging to use as a fuel in standardized equipment such as boilers, engines, and turbines that have been built for combusting petroleum-derived fuels. The variability of its composition (due to different feedstocks, reactor types, operating parameters, recovery systems, etc.) makes the standardization of its physical and chemical properties as well as the prediction of combustion behaviour difficult, making large-scale applications even more challenging. The added difficulty of isolating and identifying the highly variable compounds that comprise it have further limited its range of applications.  However, compared to traditional biomass fuels such as black liquor or hog fuel, bio-oil presents a much better opportunity for high-efficiency energy production, and significant effort has been spent on the research and development of bio-oil applications,38 the most notable of which are discussed below.  Generation of Heat and Electricity While bio-oil’s high oxygen and water content gives it a much lower heating value compared to fossil fuels, combustion tests have shown that bio-oil from fast pyrolysis can be used in the place of heavy and light fuel oils for industrial boiler applications, with little to no modification of existing equipment required. Combustion of pure bio-oil on its own, as well as its co-combustion with fossil fuels and with natural gas have been demonstrated on larger scales, usually for district heating purposes.37,43–45 Fortum, a Finnish energy company, has recently commissioned an integrated bio-oil plant with a combined heat and power (CHP) production plant in Joensuu in order to produce 50,000 tonnes of bio-oil per year to supply the heating needs of more than 10,000 households.46 Production of Chemicals Companies such as Wright’s, Figaro (now a division of Baumer Foods) and Red Arrow Products have patented and commercialized a line of natural smoke condensates from bio-oil’s characteristic smoky odour to artificially create the taste, colour and smell of meat through direct injection, coatings, marinades, etc. However, bio-oil’s rich organic composition also makes it an attractive source of chemicals such as phenols, carboxylic acids, aldehydes and ketones as well     12 as compounds with the potential to be synthesized into biopolymers. Research has been done to investigate the potential of utilizing the whole bio-oil as a substitute for the phenol-formaldehyde resins used in manufacturing wood panels such as plywood, medium-density fibreboard (MDF) and particleboards; however, it appears that the challenge of cost and quality requirements have yet to be overcome.47  Transportation Fuels The simplest method of introducing bio-oil as a transport fuel appears to be in some combination with diesel fuel, similar to how biodiesel blends are currently used. Although biomass pyrolysis oils are immiscible with hydrocarbons, they can be emulsified with diesel fuel with the aid of surfactants. Stable emulsions containing 25-75 wt.% bio-oil in diesel have been produced and tested in diesel engines, showing promising ignition characteristics; the main drawbacks are the cost of surfactants and the energy required for emulsification, the emulsions also produced significantly higher levels of corrosion/erosion in engine applications compared with diesel alone.48,49  1.3.3 Upgrading Methods/Technologies Bio-oil can be upgraded through physical, chemical or catalytic methods depending on the desired end use for the final product(s). The degree of upgrading can usually be tied to one or more of these objectives: chemical recovery, fuel for heat or electricity generation and transportation purposes. Physical upgrading techniques generally aim to separate out unwanted particles and/or reduce viscosity and include: hot gas filtration, solvent addition,50 emulsification,51,52 centrifugation, and steam stripping, although not all of them may be practical on a larger scale (e.g. centrifugation, hot gas filtration).  The major differences between bio-oil and petroleum-derived fuels is the presence of oxygenated (aliphatic and aromatic) compounds and lower sulphur levels in bio-oil; hence, chemical and catalytic upgrading processes are mainly concerned with deoxygenation which can be done via hydrotreatment,53–55 catalytic cracking of pyrolysis vapours,56,57 esterification58 and steam reforming. Aspects of the emulsification and catalytic cracking methods in relation to this project     13 will be discussed in Section 1.4, but comprehensive reviews of bio-oil upgrading methods can be found in the publications by Maggi & Elliott,59 Bridgwater21,29 and Zhang et al.39  1.4 Project Justification and Scope of Work As alluded to in the earlier sections, bio-oil’s utility as a liquid fuel can be extended towards the production of transportation fuels and chemicals after upgrading. However, the former first requires bio-oil to undergo some form of chemical transformation to increase its thermal stability and volatility and reducing its viscosity through oxygen removal and molecular weight reduction.60 To avoid extensive engine modifications, fuel properties such as HHV, corrosivity, and cetane ratings would also need to be satisfied. Upgrading bio-oil via emulsification with diesel51 or biodiesel61,62 has been shown to produce emulsions with lower viscosities and acid numbers as well as increased energy content compared to the original bio-oil raw material. However, the surfactants required to facilitate these emulsifications add to the cost and energy requirements. Ikura et al. estimated that the cost of using Hypermers (commercial surfactants) would range from 5.2 cents/L for a 10% emulsion to 8.9 cents/L for a 30% emulsion.51 It is interesting to note that bio-oil itself can be considered a micro-emulsion of sorts, wherein the discontinuous phase (the pyrolytic lignin macromolecules) is stabilized by the continuous phase (the aqueous solution of hollocellulose decomposition products).21  The rationale for using biodiesel for the emulsification can be attributed to several factors: (1) it has been proven to be an acceptable substitute for diesel fuel*, and is already an EPA-registered fuel and fuel additive; (2) it is a renewable fuel itself, having been derived from animal fats and vegetable oils; (3) it results in lower greenhouse gas (GHG) and particulate emissions and is biodegradable and non-toxic; and (4) it has really good solvent properties. The emulsification process itself utilizes the solvency of biodiesel, extracting the desirable fuel components into a biodiesel-rich phase, leaving behind a lignin-rich layer containing the pyrolytic lignin as well as other small molecular compounds for potential value-added co-product streams.                                                  * It is usually used in blended form (e.g., B20 – 20% biodiesel, 80% diesel, in the US) due to lack of regulatory incentives and pricing; B20 or lower-level blends generally do not require engine modifications.63     14 While a recent characterization study by Jiang and Ellis61 has shown a rather significant improvement in the fuel properties of a bio-oil/biodiesel emulsion in comparison to the crude bio-oil, the quality still falls short of the standards set for diesel fuel as well as biodiesel, indicating that there is still more work to be done, as shown in Table 1-2.  Table 1-2. Fuel property comparison for diesel, biodiesel, bio-oil and a bio-oil/biodiesel emulsion. Fuel Property No. 2 Diesel64 Biodiesel (B100)65 Crude Bio-oil61,66 40 vol.% Bio-oil/Biodiesel emulsion61 Fuel Standard ASTM D975 ASTM D6751 N/A N/A Viscosity (mm2/s) 1.2 – 4.1 (@40°C) 1.9 – 6.0 (@40°C) 19.0 (@40°C) 5.2 (@25°C) Water content 0.05 vol.% (max) 0.05 vol.% (max) 15 to 35 wt.% 0.46 wt.% HHV (MJ/kg) 45.3 41.43 15.3 35.8 Density (kg/m3) @20°C 853 880 1200 895 Acid number (mg KOH/g) N/A 0.50 79.23 14.01  Bio-oil deoxygenation is usually performed via two conventional petroleum-refining processes – hydrotreating (HDT) and catalytic cracking. The HDT of bio-oil incorporates the same principles and process technology as that used for petroleum feeds, however, instead of removing sulphur and nitrogen (neither of which are present in significant quantities in bio-oils), the process is used to remove oxygen, increasing the HHV of bio-oil. The primary drawbacks of the process are the high pressure (up to 200 bars) and significant hydrogen requirements; a dual-stage process – mild HDT followed by more severe HDT has been found to be necessary to overcome the reactivity of the bio-oil.67  In conventional petroleum refining, catalytic cracking decomposes heavy crude fractions into medium and light distillates at high temperatures in the presence of a cracking catalyst (usually synthetic crystalline zeolites) to create more desirable products and reduce the amount of residuals. In bio-oil upgrading, the process produces similar deoxygenation results as HDT but without the need for high pressures or hydrogen. While the process tends to suffer from the     15 problem of catalyst deactivation, the use of high heating rates, high catalyst to feed ratios and proper catalyst selection have been shown to minimize coke formation and delay the deactivation.68 Catalytic cracking can be integrated into the pyrolysis process in a number of ways: catalytic pyrolysis – where the biomass is pyrolyzed in the presence of the catalyst, catalytic vapour cracking (or close coupled vapour upgrading), decoupled liquid bio-oil upgrading and decoupled vapour upgrading from volatilization of bio-oil.21 The main objective of catalytic vapour cracking is to conduct in situ upgrading of the pyrolysis vapours prior to their condensation into the actual bio-oil. Similar to the catalytic cracking process used in the petroleum industry, the pyrolysis vapours undergo a variety of chemical reactions (depending on the choice of catalyst), which affect the final composition of the condensed liquid (bio-oil). By simply altering the combination of catalyst and feedstock, the quality of the crude bio-oil and consequently, its fuel properties and chemical composition, could be improved, potentially making subsequent upgrading steps easier and more economical. Currently, different catalysts are being studied in order to determine their suitability based on the type of pyrolysis and the feedstock being used. A summary of some commonly used commercially available catalysts and their experimental applications are shown in Table 1-2. The most commonly used catalysts are zeolites, however, with ZSM-5 types being the zeolites of choice for pyrolysis due to their activity, selectivity (towards aromatic compounds), limited deactivation by coke and high thermal stability.69 ZSM-5 has been found to dramatically change the composition of bio-oils by reducing the amounts of oxygenated compounds via deoxygenation reactions and simultaneously increasing the yield of aromatic compounds, producing a lighter (gasoline-like) fraction and decreasing the bio-oil’s molecular weight.70 However, the improvement in the oil quality does come at the expense of the yield and undesirable by-products such as the extra water and CO2 present additional challenges.          16 Table 1-2. Summary of commonly used commercially available catalysts. Catalyst Name Type Producer Composition Operating Temperature System ZSM-5 Zeolite BDH (UK), Zeolyst International Varying Si/Al ratio 400-600°C Fast catalytic pyrolysis of biomass56 H-Beta zeolites Zeolyst International Varying SiO2/Al2O3 ratios  450°C Catalytic pyrolysis of biomass in a fluidized bed reactor71 HY zeolite Zeolyst International Varying Si/Al ratios 425 and 500°C  Pyrolysis of tires in a conical spouted bed reactor72  Zinc Oxide Metal Acros Organics, Fisher Scientific 99.5% ZnO 500°C  Catalytic fixed -bed pyrolysis of cassava73,74 Dolomite Natural Närke-Ernströms AB CaCO3MgCO3  700-900°C  Catalytic tar decomposition75  Olivine Magnolithe GMBH Company, Reade Advanced Materials (Mg,Fe)2SiO4  800°C Methane and tar-cracking76  Utilizing the previous studies on various zeolite catalysts77–79 in conjunction with literature showing the effect of surfactant dosage, bio-oil/biodiesel ratios, and emulsification conditions on emulsion stability61 and favourable storability of the emulsion,80 this thesis project is divided into two main phases – (1) the emulsification of bio-oil with biodiesel followed by separation and characterization of the biodiesel-rich and lignin-rich phases; and (2) the design and building of a catalytic test unit for catalytic vapour cracking followed by the characterization of the catalyzed and uncatalyzed crude bio-oil samples. This thesis was done in collaboration with Diacarbon Energy Inc. (Burnaby, BC, Canada) and investigates whether their extraction location of the sample has any effect on crude bio-oil quality prior to being upgraded via emulsification. The     17 catalytic test unit was built based on their pilot plant’s operating conditions, system restrictions/limitations and site logistics. Since most research studies done on catalytic pyrolysis have focused on incorporating it into the fast pyrolysis process – where bio-oil production is maximized, this project attempts to extend the concept to the slow pyrolysis process where bio-oil is also produced as a by-product, albeit in smaller amounts. Given bio-oil’s complex composition, extracting out its desirable fuel components to improve the fuel quality of biodiesel and simultaneously leaving behind its lignin-rich components potentially provides more than two valuable product co-streams. Applied successfully, the results of this project would not only add value to Diacarbon’s pyrolysis units/products, but the fundamentals could then be extended towards expanding the knowledge base in the areas of alternative energy and renewable resources – indirectly contributing to the growth and development of our bio-based economy. Given the proliferation of biomass from Canada’s forestry and agricultural sectors, the development of technologies to utilize biomass-based alternative fuels will enable us to convert what was originally a waste stream into an energy stream. This not only creates the opportunity to reduce waste (and net CO2 emissions) but also produces an alternative and renewable source of fuel (as well as potential value-added chemicals).       18 Chapter 2 Materials and Methodology 2.1 Bio-oil Sample Collection, Storage and Handling Crude bio-oil samples from three different extraction points were obtained from Diacarbon’s pyrolysis unit during a site visit on May 8, 2012. The biomass feedstock for their slow pyrolysis process was a spruce-pine-fir (SPF) mixture and the samples were extracted from the sample ports shown in Figure 2-1 below. The entire tank system acts like a series of condensers where the condensed bio-oil is progressively cooled from Tank 1 (~170°C) to Tank 3 (~80°C); tar and other heavier compounds end up in Tanks 1 and 2 while the lighter fractions get drawn through the pump and into Tank 3. Based on the location of the sample extraction point, the bio-oil samples will be referred to hereafter as Diacarbon Tank 2, Diacarbon Vacuum Pump, and Diacarbon Tank 3 for the remainder of this report. The collected samples were stored in 500-mL glass jars and placed in the cold room at 4°C. An additional 5 gallons of crude bio-oil from Tank 3 was received on June 15, 2012 (collected on June 13, 2012) and stored unopened in the polyethylene pail it came in inside the cold room at 4°C. A few weeks later, the container was opened, warmed up using a pail heater and the contents thoroughly mixed using a top-mounted propeller stirrer then pumped and portioned out into 600-mL glass jars (to facilitate easier sampling and handling in subsequent experiments). Each of the jars was blanketed with nitrogen and stored at 4°C until ready for use.   Figure 2-1. Simplified schematic diagram of Diacarbon's pyrolysis unit.     19 2.2 Chemicals and Other Materials 2-propanol, methyl alcohol (99.9%, extra dry) and potassium hydroxide (0.1 N) were obtained from Fisher Scientific Canada while the other chemicals – Toluene (anhydrous, 99.8%), Hydranal® Composite 5, and 1-octanol (ACS Reagent) were from Sigma Aldrich Canada.  While the bio-oil was supplied by Diacarbon, the biodiesel used for the emulsification process was generously donated by Milligan Biofuels Inc. (formerly Milligan Bio-Tech, Foam Lake, SK, Canada). Pre-formed zeolite pellets were purchased from ACS Materials (Medford, MA, USA) and silicon carbide grit (#8 Mesh) was obtained from Kramer Industries, Inc. (Piscataway, NJ, USA).  2.3 Sample Homogenization  The crude bio-oil samples were homogenized using a Polytron PT 10-35 homogenizer, which utilizes the rotor/stator principle to evenly distribute different components (solids, liquids, or gases) throughout a liquid system. In a rotor/stator system, the tip of the dispersion aggregate is immersed into the sample and the spinning rotor generates a vacuum that draws sample in then discharges it back to the bulk through the stator’s slots, the high shear mixing action quickly homogenizes samples. For this project, this same process was applied for creating the emulsions. 2.4 Characterization Methods 2.4.1 Water Content The water content of the samples was analyzed on a Metrohm 794 Basic Titrino using the user-installed Karl Fisher titration method.  From the cold room, the crude bio-oil samples were warmed up to 40°C using a water bath then mixed by hand prior to sampling. The emulsified samples were stored at room temperature and were only mixed prior to sampling.  Due to the nature of the samples, the solvent used in the titration vessel was a mixture of 40% toluene and 60% extra-dry methanol by volume. Using methanol alone was not sufficient to fully dissolve the samples and the electrode was unable to sense any water present, resulting in an erroneous reading of 0% or a straight up error on the instrument. Alternatively, xylene or chloroform can also be used as co-solvents in place of the toluene portion. The method was also     20 modified to include an extra 30 seconds of extraction time to give the bio-oil sample sufficient time to dissolve in the solvent and allow the electrode to properly sense the water present. A few preliminary runs were performed to get a rough estimate of the water content in the samples and the sample size was adjusted accordingly to the recommended ASTM sample amount for testing.81  2.4.2 Total Acid Number The total acid number (TAN) of the bio-oil samples was measured using the Metrohm 794 Basic Titrino with 2-propanol as the solvent using the pre-loaded monotonic equivalence-point titration (MET) method as opposed to the dynamic equivalence-point titration (DET) method more commonly used for acid/base titrations. The MET compensates for background noise that might result from the nature of the sample matrix and is particularly suitable for determining acid or base numbers (TAB & TBN) in oils. The TAN of the 2-propanol solvent is first determined then subtracted from the TAN measured for the samples. From the cold room, the crude bio-oil samples were warmed up to 40°C in a water bath and mixed prior to sampling. The emulsified samples were stored at room temperature and were only mixed prior to sampling. A few preliminary runs were performed to get a rough estimate of the acid number in the samples and the sample size was adjusted accordingly to the recommended ASTM sample amount for testing.82  2.4.3 Density and Viscosity The dynamic viscosities of the samples were measured using a Brookfield DV-E Viscometer fitted with an ultra-low adapter (ULA) and the accompanying spindle. The viscosities were measured at 40°C and the reported values are the average of triplicate measurements. For the ULA spindle, the shear rate is the product of the spindle speed (in rpm) x 1.223. Spindle speeds are chosen to give the highest torque reading possible to minimize the error. From the cold room, the crude bio-oil samples were warmed up to 40°C in a water bath and mixed prior to sampling. The emulsified samples were stored at room temperature and were only mixed prior to sampling. Samples were allowed a 10-minute equilibration time after being loaded into the ULA sample holder with the spindle inserted and the water jacket slipped on.      21 To determine the kinematic viscosity, the density of the samples (averaged over at least three replicates) was measured using a micropipette and an analytical balance. Using a densitometer (Anton Paar DMA 35n) produced inconsistent results given that it was difficult to prevent air bubbles from forming in the U-tube after drawing the sample into the syringe. Even with the micropipette, the values reported are likely somewhat lower than what they actually are since it is impossible to fully dispense the samples as some of it invariably remains on the walls of the pipette tip. 2.4.4 Elemental Analysis and Energy Content (High Heating Value) The crude and emulsified samples were sent to Canadian Microanalytical Service, Ltd. (Delta, BC, Canada) for elemental analysis and the high heating value (HHV) was calculated for each of the samples based on the C, H and O values from the analysis. In the absence of an oxygen-bomb calorimeter, the HHV was calculated using this formula: HHV (MJ/kg) = [338.2 %C + 1442.8 (%H - (%O/8))] x 0.001.61,83         22 Chapter 3 Emulsification with Biodiesel 3.1 Crude Bio-oil Characterization (Pre-emulsification) A preliminary characterization of the bio-oil was conducted in order to determine which extraction point would provide the sample that would be most suitable for upgrading via emulsification with biodiesel. The samples were characterized on the basis of total acid number, water content, and viscosity.  Table 3-1. Comparison of the TAN of the bio-oil samples obtained from the different extraction points on Diacarbon’s pyrolysis unit; the error range indicates the standard deviation for triplicate measurements of the same sample. Sample Total Acid Number (mg KOH/g sample) Diacarbon Tank 2 71.07 ± 0.96 Diacarbon Vacuum Pump 42.83 ± 1.77 Diacarbon Tank 3† 53.47 ± 3.97  Table 3-2.Summary of density and dynamic and kinematic viscosities for the bio-oil samples obtained from different extraction points; the error range indicates the standard deviation for triplicate viscosity and at least triplicate density measurements of the same sample. Sample Spindle Speed (RPM) Dynamic Viscosity at 40°C (cP) Density (g/L) Kinematic Viscosity at 40°C (mm2/s) Diacarbon Tank 2 100 2.021 ± 0.002 1087.3 ± 0.1 1.859 Diacarbon Vacuum Pump 1.5 383.2 ± 0.8 1035.1 ± 3.1 370.2 Diacarbon Tank 3‡ 2.5 226.9 ± 0.1 988.1 ± 2.6 229.6                                                  † After collecting the bio-oil samples during the May 8, 2012 site visit, Diacarbon informed us that there had been a water leak in Tank 3 (which resulted in a TAN of 47.70 ± 0.77 mg KOH/g) hence, the value reported here is for the Tank 3 sample collected on June 13, 2012. ‡ After collecting the bio-oil samples during the site visit on May 8, 2012, Diacarbon informed us that there had been a water leak in Tank 3. Hence the value reported here is for the Tank 3 sample collected on June 13, 2012; in addition, the remainder of the sample collected in May was insufficient to conduct the viscosity tests in triplicate.     23 Table 3-3. Comparison of water content in the bio-oil samples obtained from the different extraction points on Diacarbon’s pyrolysis unit; the error range indicates the standard deviation for triplicate measurements of the same sample. Sample Water Content (wt.%) Diacarbon Tank 2 71.5 ± 1.5 Diacarbon Vacuum Pump 22.9 ± 0.7 Diacarbon Tank 3§ 7.40 ± 0.09   3.2 The Emulsification Process From the characterization tests done on the crude samples, the bio-oil sampled from tank 3 was found to have the lowest water content (refer to Table 2-3) even though the bio-oil extracted from the vacuum pump had the lowest total acid number (see Table 2-1). While both water content and acidity are important parameters, the water content of the crude bio-oil from tank 3 is at least three times less than that of the sample drawn from the vacuum pump; while, the difference between their total acid numbers is relatively smaller (~1.5 times). With advances in engine technology resulting in a tightening of tolerances around fuel quality, engines have become more sensitive to fuel contaminants – mainly in the form of particulates and water. The presence of water reduces the combustibility of the fuel and can have adverse effects on its lubricating properties. It promotes corrosion of tanks and equipment and microbial growth can occur in fuel systems at the fuel-water interface,84 increasing the likelihood of plugged fuel filtration systems. Plugging is also exacerbated at colder temperatures from ice formation, and even more so when wax forms around the ice crystals. Hence, the bio-oil from tank 3 was used for the succeeding emulsification experiments. In the previous study by Jiang and Ellis,61 the optimal parameters for producing a stable bio-oil/biodiesel emulsion were found to be: a surfactant (octanol) concentration of 4% by volume, an initial bio-oil/biodiesel volume ratio of 4:6, a homogenization temperature of 30°C and a                                                 § After collecting the bio-oil samples during the site visit on May 8, 2012, Diacarbon informed us that there had been a water leak in Tank 3 (the water content measured was 83.21± 0.70 wt.%); hence the value reported here is for the Tank 3 sample collected on June 13, 2012.     24 mixing time of 15 minutes at a stirring intensity of 1200 rpm. While the original study utilized bio-oil obtained from VTT and a soybean-based biodiesel (World Energy), this project uses bio-oil from Diacarbon and a canola-based biodiesel, thus, based on the optimal conditions determined from the previous study, the following methods of preparing the mixtures of bio-oil and biodiesel were tested, as shown in Table 3-4: Table 3-4. Preparation methods for bio-oil/biodiesel mixture production Method Procedure Results and Observations (refer to Figure 3-1 for images below) 1 Centrifuge bio-oil for 5 minutes at 10000 rpm prior to mixing with biodiesel (at an initial 4:6 volume ratio) using a homogenizer without the addition of octanol. Stratification started almost immediately (within the first 5 minutes), two distinct layers were formed and the development of the layers was easily observable.  2 Homogenize bio-oil for 5 minutes at 17200 rpm prior to mixing with biodiesel (at an initial 4:6 volume ratio) using a homogenizer without the addition of octanol. Stratification was more gradual compared to Method 1, the container had to be held to the light at a certain angle in order to distinguish between the two layers.  3 Centrifuge bio-oil for 5 minutes at 10000 rpm prior to mixing with biodiesel (at an initial 4:6 volume ratio) and 4 vol% octanol using a homogenizer.  Similar to the first method, except the stratification appeared to proceed more rapidly.  4 Homogenize bio-oil for 5 minutes at 17200 rpm prior to mixing with biodiesel (at an initial 4:6 volume ratio) and 4 vol.% octanol using a homogenizer. After homogenization, a small amount of the bio-oil clumped together, settling to the bottom of the flask after emulsification, apart from the apparent division between the dark fluid and the lighter golden brown liquid, no other phases could be distinguished even under strong light.     25  Figure 3-1. Results from the different emulsification methods after one weekMethod 1. Centrifuged bio-oil emulsified with biodiesel (no octanol) Volume of Upper Layer:  ~18.2 mL (after 24 hours) ~26.7 mL (after 1 week) Volume of Lower Layer:  30 mL (after 24 hours) 21.5 mL (after 1 week) Method 2. Homogenized bio-oil emulsified with biodiesel (no octanol) Volume of Upper Layer:  ~20 mL (after 24 hours) ~25 mL (after 1 week) Volume of Lower Layer:  29 mL (after 24 hours) 24 mL (after 1 week)  Method 4. Homogenized bio-oil emulsified with biodiesel (with 4% octanol) Volume of Upper Layer:  ~39 mL (after 24 hours) ~39 mL (after 1 week) Volume of Lower Layer:  9 mL (after 24 hours) 9 mL (after 1 week)  Method 3. Centrifuged bio-oil emulsified with biodiesel (with 4% octanol) Volume of Upper Layer:  ~25.7 mL (after 24 hours) ~29.7 mL (after 1 week) Volume of Lower Layer:  23.5 mL (after 24 hours) 19.5 mL (after 1 week)     26 For the homogenization step in all the methods, a Polytron PT 10-35 homogenizer was used. After performing the homogenization in conical flasks, the mixtures were transferred to graduated cylinders, covered with parafilm and allowed to stand at room temperature over 24 hours to allow the layers to develop. Furthermore, the mixtures were left for up to one week to determine if additional stratification would take place.  Bio-oil typically contains 40-45 wt.% oxygen from the diverse oxygenated organic compounds present, making it immiscible with hydrocarbons, but miscible with polar solvents like methanol, ethanol, and acetone.21 In this case, as the emulsification mixtures were formed with and without the use of octanol, it was hypothesized that the surfactant might be unnecessary due to the polarity of biodiesel. When left at a standstill (usually within weeks/months), bio-oil slowly begins to develop a thick bottom layer (usually composed of the heavier lignin components) that is several inches thick and viscous like honey, a thick floating layer (~10–50 mm) may also form on the surface; however, the bottom layer shows no distinct separation from the bulk but rather a gradual transition.85 This was observed in the sample jars containing the samples from Diacarbon (collected on May 8, 2012), heating the contents to temperatures between 35 and 40°C and/or vigorously stirring with the homogenizer at 17200 rpm re-establishes homogeneity but the samples cool down relatively quickly and the bottom layer reforms rapidly.  While Ikura et al.51 centrifuged their crude bio-oil to remove the heavy fraction and solids and utilized only the light top fraction for the emulsification mixtures; for practical reasons, it might be more reasonable to heat, homogenize and use the whole bio-oil to prepare the emulsification, as there might be more than char and small fines centrifuged out. Also, once centrifuged it is quite difficult to recover the solids deposited on the bottom of the tubes for analysis. Since these do end up settling to the bottom in the lignin-rich phase, it might be more practical to filter them out after the emulsification instead. After the preliminary emulsifications, emulsification replicates were performed based on methods 2 and 4 (homogenization of the crude bio-oil prior to emulsification, with and without octanol). However since the latter batch of bio-oil from tank 3 (collected on June 13, 2012 – no water leak) did not exhibit separation, unlike the tank 3 sample collected on May 8, 2012, the first homogenization step was deemed unnecessary. The sample only needed to be warmed up to     27 40°C in a water bath and mixed by hand (with a stirring rod or shaken in the jar) prior to sampling.  The emulsification conditions used were as follows: a 10-minute mixture equilibration at 30°C, 15-min homogenization at 30°C and 17200 rpm (setting 4 on Polytron PT 10-35), the mixture was then covered and left to separate out into layers over 24 hours at room temperature. In the original study the homogenizer was set at 1200 rpm, although it did conclude that increasing the stirring intensity resulted in more stable mixtures. The homogenizer used for this project is a different brand and model altogether – with different rpm speed control settings, thus the setting chosen was one which maximized mixing without causing sample loss (from getting flung out of the container). The emulsification triplicates were done with and without octanol. Table 3-5. Comparison of stability and volume ratios of the biodiesel-rich and lignin-rich layers after emulsification; the error range indicates the standard deviation for triplicate measurements of the same sample. Sample Volume Ratio of Upper to Lower Layer S (Stability)* VTT Finland Bio-oil/World Energy Biodiesel (4:6 ratio), with 4 vol% octanol61 73:27 0.21 ± 0.05 Diacarbon Tank 3/Milligan Biotech Biodiesel (4:6 ratio) with 4 vol% octanol 89:11 0.474 ± 0.006 Diacarbon Tank 3/Milligan Biotech Biodiesel (4:6 ratio) no octanol 84:16 0.454 ± 0.007 *Stability is defined as the volume of bio-oil dissolved per volume of biodiesel over a 24-hour period. As shown in Table 3-5, with and without the use of octanol, the volume of bio-oil dissolved in the biodiesel was more than double in comparison to the original study. The resulting upper to lower volume ratios were also higher. In addition, after a week, the volume ratios appear to remain unchanged.  3.3 Experimental Results and Discussion After the emulsification process, the biodiesel- and lignin-rich layers were isolated and analyzed separately. However due to the insolubility of the bottom layer, only the top biodiesel-rich layer     28 could be characterized on the basis of total acid number, water content, and viscosity. The fate of the lignin-rich layer is discussed in Section 3.3.2. 3.3.1 Biodiesel-rich Phase (Upper Layer) Table 3-6. Comparison of the total acid number and water content for the samples – pre- and post-emulsification; the error range indicates the standard deviation for triplicate measurements of the same sample. Sample Total Acid Number (mg KOH/g sample)  Water Content (wt.%) Diacarbon Tank 3 crude bio-oil 53.47 ± 3.97  7.40 ± 0.09  Biodiesel 5.92 ± 0.40  0.21 ± 0.01  Biodiesel-rich layer, with octanol 16.56 ± 0.18  1.30 ± 0.03  Biodiesel-rich layer, no octanol 14.50 ± 0.64  0.94 ± 0.01   After emulsification, the biodiesel-rich phase showed significant reductions in total acid number, with and without the use of the octanol surfactant (75.3% and 78.4%, respectively), compared to the original crude bio-oil sample. Likewise, the water content als o showed a significant decrease, by 82.4% and 87.3%, respectively. Having the emulsions form without the use of octanol and obtaining similar (if not better) reductions in the TAN and water content is both interesting and important, especially as one of the main obstacles of developing any new technology is cost, and surfactants tend to make up a large portion of the processing cost. Even assuming the use of a proprietary CANMET s urfactant, Ikura et al. calculated the cost of producing stable emulsions to range from 2.6 cents/L for a 10% emulsion to 4.1 cents/L for a 30% emulsion. 51  While the reduction in density is not as remarkable, the improvement in densities after emulsification, again with and without the use of octanol, is worth noting. As with the TAN and water content, the results are quite similar, with the no-octanol emulsion having a slightly lower viscosity.        29 Table 3-7. Comparison of the density and dynamic and kinematic viscosities for the crude bio-oil and biodiesel-rich layers after emulsification; the error range indicates the standard deviation for at least triplicate measurements of the same sample. For detailed viscosity results, refer to Tables A22-24. Sample Spindle Speed (RPM) Dynamic Viscosity at 40°C (cP) Density (g/L) Kinematic Viscosity at 40°C (mm2/s) Diacarbon Tank 3  2.5 226.9 ± 0.1 988.1 ± 2.6 229.6 Biodiesel-rich layer (4vol% octanol) 60 9.19 ± 0.00 936.8 ± 6.9 9.81 Biodiesel-rich layer (no octanol) 60 8.94 ± 0.00 923.5 ± 0.6 9.68  The results from the elemental analysis of the crude and emulsified top-layer samples and the calculated HHVs are presented in Table 3-8. Comparing amongst the crude bio-oil samples from Diacarbon, the oxygen content is the highest for Tank 2 and lowest in Tank 3 (June 13, 2012 sample), which is the sample used for the emulsions. However, when compared to the bio-oil from VTT Finland, the oxygen content in two out of the three Diacarbon samples is considerably lower. The calculated HHVs reflect the impact of oxygen on the energy content of the bio-oil, in general the higher the oxygen content, the lower the HHV, hence one of the reasons upgrading bio-oil generally involves deoxygenation in one form or another. In addition to lowering energy content (HHV), oxygen is also the main cause of corrosiveness and thermal instability.86 After emulsification, the oxygen content of the crude bio-oil was reduced by about 1/3 and the HHV increased to just slightly below the value of pure biodiesel. It is also worthwhile to note that while the crude bio-oil from Diacarbon had an HHV that was nearly 50% more than the crude bio-oil from VTT Finland, after emulsification, the difference in their HHV values was rather small (~5%). Given that the starting crude bio-oil samples and biodiesel sources are different, it is difficult to draw definitive conclusions as to why this would be so as there are quite a number of variables that come into play.         30 Table 3-8. Elemental analysis and calculated HHVs for crude and emulsified samples. Sample %C %H %N %O HHV* (MJ/kg) Diacarbon Tank 2, May 8, 2012 15.71 9.16 <0.3 66.41 6.55 Diacarbon Tank 3, May 8, 2012* 47.25 7.92 0.34 28.2 22.32 Diacarbon Vacuum pump, May 8, 2012 66.64 6.96 <0.3 26.69 27.77 Diacarbon Tank 3, June 13, 2012 70.36 7.08 0.81 22.79 29.90 Biodiesel-rich layer (40%bio-oil, 4vol% octanol) 74.58 10.19 0.46 14.36 37.34 Biodiesel-rich layer (40%bio-oil, no octanol) 75.49 10.58 <0.3 13.87 38.29 Biodiesel (canola-based, from Milligan Biotech) 74.80 11.76 <0.3 12.95 39.93 VTT Finland Crude Bio-oil61 39.96 7.74 0.11 52.19 15.27 Biodiesel-rich layer (VTT Finland Bio-oil/World Energy Biodiesel (4:6 ratio) emulsion, with 4 vol% octanol)61 66.48 11.89 0.13 21.5 35.76 *Note that there was a water leak when the Tank 3 sample was collected on May 8, 2012, the results from the analysis are presented only for completeness. 3.3.2  Lignin -rich Phase (Bottom L ayer) It was originally planned to characterize the lignin-rich layer from the emulsification in the same manner as the biodiesel-rich layer. However, the bio-oil from Diacarbon was so dark in colour that after the homogenizing step, it was impossible to distinguish the layers clearly, and impossible to identify how fast the layers were developing. The only way to tell the layers apart was by pouring out the upper layer and leaving behind the bottom; in comparison, an emulsification trial performed using bio-oil from Dynamotive Energy Systems (Richmond, BC, Canada) showed layers that were fairly easy to distinguish (see Figure 3-2). Separating out the layers turned out to be the easy part as the entire upper layer was nice and fluid while the bottom layer remained stuck at the bottom. Removing the bottom layer turned out to be almost impossible especially during the initial experiments where the mixture was allowed to separate in     31 a graduated cylinder; the only way to get the bottom layer out was to dissolve it with lots of methanol and/or 2-propanol then scrub quite hard to get the remaining residue out.  Moving the emulsification to wide-mouthed jars made the bottom layer more accessible after separating out the upper layer. The bottom layer turned out to be a very viscous, almost taffy-like substance that would just not completely dissolve in any of the solvents that were used for water content determination and total acid number. Warming up the sample in the covered jar to about 40°C made it more malleable and gave it a glossy wet sheen, but as soon as it came into contact with the air at room temperature, it cooled instantly and the glossy sheen turned dull (see Figure 3-3 below).  Figure 3-2. Distinct layers shown during emulsification using Dynamotive bio-oil.  Figure 3-3. Lignin-rich phase after emulsification of Diacarbon’s bio-oil. Since the lignin-rich layer would not dissolve and consequently could not be characterized using the methods for water content and total acid number, the sample was subjected to the water precipitation method87 to determine if lignin could be extracted. As a basis, the lignin- Biodiesel-rich Layer  Lignin-rich Layer      32 precipitation was first carried out using crude and emulsified bio-oil samples from Dynamotive Energy Systems (see Figure 3-4). Using the bio-oil from Diacarbon, a gel-like blob formed on the surface of the filter instead, making filtration impossible (see Figure 3-5), and for the Diacarbon lignin-rich sample from the emulsification, it was not possible to get through the first part of precipitating the sample in the water. A possible explanation for the difference could be the pyrolysis processes used by the companies – the Dynamotive plant uses fast pyrolysis while Diacarbon’s is closer to that of a slow pyrolysis method. Bio-oils produced using fast pyrolysis have significantly different physical and chemical properties compared to those from slow pyrolysis processes, which are more like tar.27 In addition, the presence of residual char particles in bio-oil can make filtration under pressure very difficult due to the complex interactions between the char and the pyrolytic lignin, which results in the formation of a gel-like phase that rapidly blocks the filter.27    Figure 3-4. Lignin extraction using bio-oil from Dynamotive. (a) Start of experiment (b) After 10 minutes (c) Lignin extracted from crude bio-oil (after drying) (d) Lignin extracted from lignin-rich phase of emulsification (after drying)     33  Figure 3-5. Diacarbon bio-oil on filter surface after water precipitation method.  3.3.3 Water Mass Balance In order to confirm that the majority of the water is retained in the lignin-rich phase and establish the division of water between the two phases after emulsification, a water balance was carried out. Since the lignin-rich phase from the emulsification using Diacarbon’s bio-oil could not be analyzed for water content for the aforementioned reasons in the previous section, the water balance was carried out on the emulsification of bio-oil obtained from Dynamotive. While there was some water loss (~9%) after emulsification, likely to the atmosphere or during the transfer between containers, nearly all of the remaining water (~97%) ended up in the lignin-rich phase, which was one of the main objectives. The distribution of the water among the different components pre- and post-emulsification is shown on the following page in Figure 3-6.The water content of the biodiesel-rich phase was lowered significantly compared to the crude bio-oil (0.51 vs. 24.45 wt.%), however based on ASTM standards, the maximum allowable water is 0.05 vol.% for both diesel fuels64 and pure biodiesel.65         34  Figure 3-6. Distribution of water among the components: (a) pre-emulsification; and (b) post-emulsification.   98.97% 0.96% 0.06% Bio-oil  Biodiesel Octanol (a) 2.93% 97.07% Biodiesel-rich layer Lignin-rich layer (b)     35 Chapter 4 Design and Construction of Catalytic Reactor Test Unit 4.1 Objectives and Overall Design Considerations The main objective for this phase of the project was to conduct the upgrading of the pyrolysis vapours in situ prior to the condensation stage through catalytic vapour cracking with the aim of improving the quality of the resulting bio-oil before further upgrading via emulsification. A catalytic reactor test unit was designed and built to draw out a slipstream from the company’s main vapour stream so that a portion of the pyrolysis vapours could be passed through a catalyst bed without affecting the rest of the main unit.  In order to obtain meaningful results from the experiments, the catalytic reactor test unit (hereafter referred to as the ‘test unit’) was designed based on preliminary information provided by the company along with several design considerations in mind. - Modularity – The test unit had to be designed, assembled then partially disassembled in the department workshop for transport and reassembled out on site so the different sections had to be able to be broken down and put back together fairly easily without compromising structural integrity. In addition, certain components needed to be easily swapped out if they became clogged or required replacing for some reason or another. Hence, a metal Unistrut® frame with movable shelves was utilized as the structural support system to provide the flexibility required. - Material compatibility – Due to the nature of the material flowing through and to keep variables to a minimum, stainless steel was used for the majority of the parts that came into contact with the vapour stream.  - Simplicity – It would be nearly impossible to foresee all the potential problems that could happen while out on site (which could be considered in the middle of nowhere) and we could only bring so many tools and spare parts along so changes in connection size and type were kept to a minimum and most parts were relatively interchangeable and easy to replace. The set-up also needed to be operable by one person if necessary (two, in the ideal case), so the orientation of the different components was also important.      36 - Safety – The entire setup needed to be structurally sound and precautions incorporated as much as possible into the design and the procedure as the test unit would be under vacuum, at high temperatures in certain sections and connected to the company’s main unit to draw out the hot vapour stream. 4.2 Process Design Details Feed supply and handling Diacarbon uses a standard mixture of spruce-pine-fir (SPF) and puts it through a dryer to get the moisture content down to about 10wt.% before pyrolyzing it. The feed supply to the test unit is a portion (the ‘s lipstream’) of their vapour stream drawn out as it exits their furnace on the way to the large condenser. From there, the slipstream is diverted towards the packed bed reactor (filled either with inert material or catalyst) then passed through a cooling coil into the sample collection flask and through the impinger train. Packed-bed reactor The packed-bed reactor is essentially a 442-mm long stainless steel (grade 316L ) tube having an inner diameter (ID) of 21 mm, encased in a single-zone, split-hinged, round tube furnace with 35-mm vestibules (refer to Figure B-1 for the schematic diagram), that is capable of operating at temperatures up to 1250°C in air. The furnace, manufactured by The Mellen Company Inc. (Concord, NH, USA) , is equipped with an integral digital temperature control system featuring a 126-segment, 31-program controller. The total heated length of both the furnace and tube is 305 mm, allowing the space time (g of catalyst-h/ g of reactant) as well as residence time to be adjusted to a certain degree based on how much catalyst and/or inert material is packed into the tube.  The test unit was designed in such a way as to allow for the catalyst tube to be removable so that the tube could be replaced relatively quickly in between experiments should there be a need to change the type or amount catalyst (and/or inert material), replace the spent catalyst or replace the clogged tube. Given that the experiment was being performed outdoors  without the benefit of the services and amenities of the department, pre-assembled and pre-filled tubes (see Figure 4-1) were prepared prior to each visit to the site. This had the added benefit of saving time,     37 minimizing the number of tools needed to be brought out to site and eliminating the possibility of forgetting to bring small but crucial parts like an extra ferrule or nut out to site . Since each tube would likely have to be wiggled into place (and back out after the experiment ), the bottom of each tube was fitted with a mesh screen held in place by two circlips to keep the catalyst and the inert filler material in place.   Figure 4-1. Pre -assembled tubes with nuts and ferrules attached (these tubes have already been used, hence the discoloration).  Cooling Coil Since the slipstream is running parallel to the main unit except with the addition of the catalyst bed section, once it exits, the main concern is the same –  to lower the temperature of the vapour stream such that it starts to condense into bio-oil as soon as possible to stop any side reactions and polymerization from taking place. Since some of the heavier compounds start to condense at about 300°C, a stainless steel cooling coil was fabricated to cool the vapour stream  to about 200°C, just en ough so that the stream travels smoothly  into the sample collection flask downstream. The cooling coil was made out of a 6.1-metre long, 10 mm-ID  stainless steel tube (calculations for cooling coil sizing can be found in Section C.3)  and after an initial run showed that relying on ambient air did not provide sufficient cooling to condense the vapour stream, it was outfitted with a large bucket that could be filled with ice and/or water.  The bucket was open on top and had a spout on the side so that the water could be continuously drained and refilled as needed.        38 Sampling Flask and Impinger Train The cooling coil is attached directly to the sample collection flask so that the semi-cooled down stream exits the coil straight into the bottom of the flask that is sitting in an ice bath at 0°C, where most of the bio-oil is ideally collected. However, as there is the possibility of incomplete condensation and aerosols remaining in the vapour stream, an impinger train, made up of a series of four flasks is set up downstream of the collection vessel to catch any remaining condensates and ultimately protect the vacuum pump. The impinger train setup was adapted from the modular sampling train for tar collection in CENT/TS 15439: Biomass gasification – Tar and particles in product gases – Sampling and analysis. The main differences were the use of four 1000-mL Erlenmeyer vacuum flasks instead of six 100-mL impinger bottles; isopropanol volumes were scaled accordingly and the last vacuum flask was filled with glass wool. The flasks were connected using PE tubing and a quick-disconnect acetal coupling was placed in-line between the sampling flask and the first flask in the impinger train to prevent any backflow into the sampling flask once the vacuum pump was turned off. Vacuum Pump The vacuum pump was sized to fulfill two main requirements: to overcome the vacuum pressure in the main reactor unit and to draw a sufficient amount of sample within a couple of hours. The pump chosen was a ChemStar Vacuum Pump, single-phase, 5.6 cfm and an ultimate vacuum of 1x10-4 Torr, the pump provided the additional advantage of being able to withstand any corrosive compounds or gases that were not caught in the impinger train beforehand.  Heating and Auxiliary Components Since the slipstream had to be kept in vapour phase until it passed through the packed-bed, the lines leading from the port had to be kept as short as possible then heat traced and wrapped with fibreglass insulation to prevent heat loss and consequently, premature vapour condensation.  As shown in the process flow diagram (Figure 4-3), pressure gauges and thermocouples were attached at various points to monitor the temperature and pressure of the system. Pressure gauges were installed before (P1) and after (P2) the reactor bed to monitor the pressure drop across the     39 bed and a thermocouple was inserted into the middle of the bed (T2) to monitor its temperature as the vapour stream flowed through.  Power and Water Supply 120-volt power and cooling water was supplied on site, but a portable generator (Honda EU3000isKC) borrowed from the department was also brought out to site in the event that additional power was needed.  4.3  Safety Considerations  The design and operation of the test unit incorporated safety features and precautions to address the following concerns: - Relatively high operating temperatures (up to 650°C)  - Corrosive and acidic nature of the incoming pyrolysis vapour stream and the resulting condensed oil stream - Potential leaks in the system - Electrical hazards - Pressure-related issues One of the primary concerns during the design phase was the relatively high operating temperature of the vapours coming out of the main unit, which would have to be maintained until after the vapours had exited the reactor tube into the cooling coil. The additional restriction brought about by the corrosive/acidic nature of the pyrolysis vapour stream basically limited the material choice to stainless steel (SS 316) for the heated portion of the unit. The combination of high temperature and corrosiveness/acidity also eliminated the potential to use any type of quick-connect couplings or connectors (to facilitate the quick removal and installation of the reactor tube) that utilized any type of standard polymer seals (O-rings or gaskets); the typical O-rings used in couplings are made of Fluoro-carbon FKM (trade name Viton®) which are only rated to a maximum temperature of ~200°C and more specialized ones like Kalrez® have an upper limit of 325°C. Hence, stainless steel Swagelok tube fittings with male NPT connectors had to be used on either ends of each of the reactor tubes instead, making the removal and installation of the     40 tubes a bit more tedious since the high operating temperature can cause the fitting threads to gall or seize and additional tools and manoeuvring are required.  Stainless steel was also used for the cooling coil, all the thermocouples and for the wetted parts of the pressure gauges. The sampling flasks were made of heavy-walled, vacuum-rated glass to avoid the risk of implosion. As the vapour stream was designed to condense into the sample collection flask, once the vapours had condensed (eliminating the high temperature issue) PE tubing (which was chemically compatible with the pyrolysis oil) could be used to connect the sample flask to the impinger train. The semi-opaque tubing was also used to interconnect the flasks making up the impinger train and provided the benefit of being able to visually monitor the bio-oil flowing through.  Since the unit was located outdoors and exposed to the elements, a makeshift plywood roof was installed to shield the majority of it from rainfall. The heating cables were well wrapped under the layers of insulation and duct tape and the vacuum pump was stored in a shed on-site when not in use. The plugs on the furnace and the heating tape were wrapped in plastic and the electrical outlets and power cords were covered and stored when the unit was not running. The temperature controller box was portable and was kept in the department when not in use.  As the system was under vacuum pressure, the presence of any leaks in the system would be noticeable as the pressure gauges would indicate the lack of vacuum pressure and the lack of vapour flowing through the system would be evident as there would be no (or decreased) bubbling through the isopropanol in the flasks that make up the impinger train. Nonetheless, the experimental setup procedure in Section B.1 includes a protocol to check for leaks prior to starting the run since identifying the source of the leak for a system under vacuum can be more difficult compared to one under positive pressure. In the event of any unforeseen incidents, an emergency shut down protocol has also been included in Section B.4 and a hazard and operability (HAZOP) table is also included in Section B.3.  4.4 Catalyst Selection and Incorporation Various studies have reported on the catalytic upgrading of pyrolysis vapours – utilizing metal oxides, mesoporous catalysts, natural and synthetic zeolites, with zeolites being the most widely     41 used.57,71,72 While the varying feedstock, operating parameters, pyrolysis processes, reactor types and other factors make comparisons quite difficult, most studies do seem to conclude that the most promising catalyst is the ZSM-5 zeolite-based catalyst because it balances activity, shape selectivity for <C12 hydrocarbons, limited deactivation by coke and hydrothermal stability.56,69,88,89 For these reasons, while other commercially available catalysts (shown in Table 1-2) were considered, it was decided to run experiments using the ZSM-5 catalyst first with the possibility of testing out other catalysts in the future.    To minimize the pressure drop in the packed bed, attempts were made to agglomerate the zeolite powder into particles sized between 2-3 mm. Attempts to make catalyst extrudates from a mixture of 50-80 wt.% ZSM-5 powder (CBV8014, Zeolyst International), 20-50 wt.% alumina binder (58 Angstroms, Alfa Aesar) and proportionate water90,91 were unsuccessful. The mixture would not remain in a malleable state long enough as it alternated between being either too wet or too dry for loading into a pelletizer. Pouring the wet mixture into a shallow container then allowing it to dry in the oven overnight at 120°C produced a hardened zeolite-binder sheet that could be shattered into pieces with a mortar and pestle then sieved into the desired particle size. However, even after a 3-hour calcination period at 600°C under nitrogen, the resulting particles could still be easily crushed by hand and had irregularly shaped pores. Using a hydraulic press (Carver Manual Pellet Press 4350) on the dry zeolite-alumina mixture alone resulted in irregularly shaped discs of varying thickness, which also broke easily. The process also proved rather tedious as it produced only one 6-mm disc at a time; hence, pre-formed ZSM-5 pellets (2 mm x 2mm x 10 mm cylindrical pellets, SiO2/Al2O3 molar ratio: 38, pore volume: !0.25 mL/g, surface area: !250 m2/g) were purchased instead (Figure 4-2). The pellets were already in H-form (HZSM-5) when received so no calcination was necessary prior to use, although they were conditioned first to drive off any water present by heating to 500°C in the tube furnace for about 30 minutes before the start of the experiments.  Silicon carbide (SiC) (Figure 4-2) was used as the inert spacer at the bottom of the tubes to hold the catalyst in place within the heated zone of the furnace and as the bed material for the blank run for the uncatalyzed experiments. While it has been shown that coke deposits can be removed     42 by heating the zeolite catalyst at 550°C in the presence of air for 8 hours,92 each catalyst and SiC bed was used only once and no in-situ regeneration of the catalyst was performed.           Figure 4-2. Pre-formed zeolite pellets (left) and SiC granules (right). 4.5  E xperimental Set- up **  The vapour stream from the main unit exits the furnace at about 550 ± 50°C then travels through about one metre of DN 150 mm black iron pipe (covered in ~50 mm of fibreglass insulation) before it reaches the port to which the test unit is attached, where it has to travel through another half a metre of 19-mm (ID) tubing (preheated to 500°C by a flexible high-temperature heating cable and covered in ~50 mm of fibreglass insulation) before it reaches the preheated catalytic reactor bed. Providing and maintaining sufficient heating to the vapour stream proved to be one of the main challenges that needed to be addressed especially with weather conditions being less than ideal at times. During the fall/winter season, temperatures out on site typically ranged between -4 and 10°C and during the cold snap, temperatures reached a low of about -7°C (before factoring in wind chill) in December 2013 and again in February 2014. Site visits were limited as the cold weather suspended operations at the pilot plant (e.g. furnace burner not starting, water cooling system for the condenser freezing up).                                                  ** Unless otherwise specified, all tubing and connections are made out of stainless steel (grade 316).     43  Initially, the port  had been connected to the reactor bed through ~1.5 metres of 38.1 -mm (ID) tubing but this gave the vapour stream a greater surface area for heat dissipation, causing it to cool down too much as it made its way to the reactor bed. In addition, the  original heating cable (312 Watts) used did not have sufficient power to counteract the heat being lost to the atmosphere through the surface of the pipe; preliminary experiments using 305-mm and 152-mm packed bed heights had proven futile largely due to insufficient heating. Using a higher-wattage heating cable (1248 Watts)  in combination with the shortened connection (from port to reactor bed) appeared to have sufficiently reduced the heat loss of the vapour stream, keeping it from pre-condensing and clogging up the packed bed.  In order to draw a portion of the vapour stream from the main unit, the  experiments were carried out under vacuum. The port connecting the main unit to the test unit was opened and closed by a knife gate valve and the vapour flow rate through the test unit was generally ~30 L/min;  run times varied between 15-30 minutes. Based on the preliminary runs, several pairs of tubes (one filled with SiC only, the other with HZSM -5) with different bed heights (305, 254, 203 and 152 mm) were prepared so that the bed height could be adjusted  by simply swapping out the tube if the pressure drop became too high.  The reactor bed was preheated to 500°C, which was considered as the experimental temperature and a thermocouple was inserted into the middle of the bed to monitor the temperature during the course of the experiment. Similar studies on catalytic upgrading of pyrolysis vapours using zeolites have generally been done at temperatures ranging from 300-500°C;69,77,88,93  at higher temperatures there is a marked increase in gas production at the expense of the bio-oil yield and the production of aromatic hydrocarbons begins to level off around 500°C.94  In addition, while some of the coke generated is burned off at higher temperatures, irreversible dealumination and loss of acid sites have been shown to occur at temperatures as low as 450°C in the presence of water, leading to decreased catalytic activity.69,95  In order to isolate the effect of the catalyst, the experiment was conducted first with a reactor tube filled only with SiC granules followed by a run with a tube filled with the HZ SM-5 catalyst. Due to time and operation constraints, only the 254 mm packed beds were tested successfully; gas hourly space velocities (GHSVs) of the pyrolysis vapours through the reactor tube were     44 similar for each run, at ~ 18970 hr-1 (i.e.,  a vapour residence time of ~0.20 sec) . After passing through the reactor bed, the vapours flowed through the cooling coil that was immersed in a bucket filled with water maintained at ~15°C.  The majority of the condensate was then collect ed directly into the sample collection flask and the remainder of the uncondensed vapour stream was directed through the impinger train to remove any remaining tars and aerosols before finally exiting through the vacuum pump and exhausted to the atmosphere . A diagram of the  process is shown in Figure 4-3 and the full experimental procedure can be found in Section B.1.  Photos of the test unit are included in Section B.5.        45  Figure 4-3. Process flow diagram for catalytic reactor unit    46 4.6 Experimental Results and Discussion During the experiments, there was a significant amount of vapour build-up inside the sample collection flask, as shown in Figure 4-4, which lingered for a while after the end of each run and had to be allowed to dissipate before the bio-oil samples could be transferred to glass jars for storage. The acetal quick-disconnect coupling connecting the sample flask to the impinger train also failed due to the breakdown of the rubber O-ring – a testament to the corrosiveness/acidity of bio-oil, fortunately this happened prior to the start of a run and the part was easily replaceable once the problem was identified.   Figure 4-4. Vapour build-up inside the collection flask during the experiment.   The bio-oil samples collected from both the non-catalytic and catalytic pyrolysis runs both exhibited the same phase separation, forming a distinct aqueous top layer and an organic bottom layer, as shown in Figure 4-5. However, after the samples were poured out from the collection flask, some solid residue invariably remained stuck to the bottom of the flask, as shown in Figure 4-6. The actual runtime without the catalyst was ~15 minutes, as it appeared that no more bio-oil     47 was being collected in the flask after that time. After removing the tube, some of the bio-oil had condensed onto the SiC granules, causing the packed bed to clog up. Approximately 600 mL of sample was collected during the 15-minute run, with the organic layer making up less than 20% of the total volume. In comparison, ~1500 mL of sample was collected over a 30-minute period for the experiment with the HZSM-5 catalyst, with the proportion of the organic layer ~20 vol.%. After removing the catalyst tube, the catalyst remained firmly stuck inside the tube – possibly due to the bio-oil that had condensed and dried up causing the pellets to clump together and onto the walls of the tube. From the few catalyst pieces that could be shaken loose, it seemed that there was a notable amount of coking on the surface.    Figure 4-5. Samples collected from the non-catalytic (left) and catalytic (right) experiments shown in 600-mL jars.      48  Figure 4-6. Residue left behind in collection flask after the contents were transferred.   The separation between the aqueous and organic layers was distinct enough such that each layer could be stirred and samples extracted out without disturbing the other. Samples from each layer were extracted out and characterized separately on the basis of viscosity, density and water content††. In a similar experiment involving the cracking of oak-derived pyrolytic vapours, Mihalcik et al. observed the same distinct immiscible layers in the product collected from some of their runs and for those instances, the oil layer could be separated for direct analysis.96 Initially, the extracted layers from each experiment looked fairly similar, as shown in Figures 4-7 and 4-8, although for the uncatalyzed experiment, the aqueous layer was slightly darker in colour and the organic layer had what appears to be the beginning of a thin aqueous layer forming on the surface. In addition, after the layers were separated, the acrid, smoky smell distinct to bio-oil seemed to have been concentrated in the bottom layer while the top layer was significantly less pungent.                                                  †† As of this writing, issues were being encountered with the pre-installed MET method for TAN determination on the Metrohm 794 Basic Titrino and have yet to be resolved.      49  Figure 4-7. Samples extracted from the aqueous (top) and organic (bottom) layers of the non-catalytic (SiC only) experiment collected into 118-mL jars.    Figure 4-8. Samples extracted from the aqueous (top) and organic (bottom) layers of the catalytic experiment (with HZSM-5) collected into 118-mL jars.      50  However, about 5  minutes after extraction, the organic layer from the non-catalytic experiment (hereafter referred to as SiC bottom) started to exhibit some stratification, and after 10 minutes, a somewhat hazy-looking aqueous layer (with suspended oil droplets) was formed, sandwiched between a thin organic layer on top and a thicker organic layer at the bottom; in comparison, the one from the catalytic experiment (hereafter referred to as HZSM -5 bottom)  remained completely homogenous, as shown in Figure 4-9  below.   Figure 4-9 . The organic (bottom) layers from the non-catalytic (left) and catalytic (right) experiments, ~10 minutes after being extracted from their respective sample mixtures. Vigorously shaking the SiC bottom mixture results in a homogenous solution once again, which lasts for about 5 minutes, after which the re-stratification occurs; this re-stratification is reminiscent of the behaviour of the Diacarbon Tank 2 bio-oil sample analyzed in Chapter 3.  This could be attributed to a higher than usual water content which has been shown to cause separation once it exceeds ~30 wt.%; 37 analysis of the Diacarbon Tank 2 sample in Section 3.1 showed that it contained 71.5 ± 1. 5  wt.% wate r, while the SiC bottom mixture contained 31.9 ± 1.3 wt.% .      51 Table 4-1. Comparison of the water content for the top and bottom layers from the non-catalytic and catalytic experiments; the error range indicates the standard deviation for quadruplicate measurements of the same sample.  Water Content (wt.%)  Layer SiC HZSM -5 Top 64.2 ± 0.3 63.1 ± 0.6 Bottom 31.9 ± 1.3  11.4 ± 0.2  As shown in Table 4-1, the water content of the top layers from both experiments were fairly similar; however, the HZSM -5 bottom sample contained less than half as much water as the SiC bottom sample. Since the test unit was not set up in a way to permit a mass balance closure (as only the liquid product was collected), it is difficult to comment on the effect of the catalyst on the product distribution. However, the effect of the catalyst on the water content, and as shown in Table 4-2 below, on the viscosity of the organic layer is quite significant when compared to the one obtained from the non-catalytic run. Since deoxygenation via catalytic cracking of pyrolysis vapours over acidic zeolite catalysts such as HZSM -5 occurs through simultaneous dehydration-decarboxylation reactions where oxygen is rejected in the form of water, CO 2 and CO ,97 it has been proposed that the water content of the bio-oil may be used as a probable index for the degree of deoxygenation.95  Table 4-2. Summary of density and dynamic and kinematic viscosities for the top and bottom layers obtained from the non-catalytic and catalytic experiments; the error range indicates the standard deviation for triplicate viscosity measurements and at least triplicate density measurements of the same sample. Sample Spindle Speed (RPM) Dynamic Viscosity at 40°C (cP) Density (g/L) Kinematic Viscosity at 40°C (mm2/s) SiC Top 100 1.319 ± 0.012 864.3 ± 0.8 1.526 SiC Bottom‡‡  1.0 210.2 ± 13.4 855.2 ± 4.8 245.8 HZSM -5 Top 100 1.313 ± 0.014 865.4 ± 2.3 1.517 HZSM -5 Bottom 30 12.46 ± 0.14 884.7 ± 2.5 14.09                                                 ‡‡  The homogeneity of this sample inside the viscosity chamber was suspect during testing, as the readings fluctuated more than usual during and between each trial, as shown by the higher standard deviation.      52 Chapter 5  Conclusion s and Recommendations  The objectives of this thesis were twofold: (1) to evaluate the effect of extraction location on crude bio-oil quality prior to  its emulsification with biodiesel and to characterize the resulting biodiesel- and lignin-rich phases;  and (2) to design and build a catalytic test unit to conduct in situ vapour cracking of slow pyrolysis vapours.  The experimental results have shown that the extraction location does affect the quality of the crude bio-oil. The effect of the surfactant on the emulsification was minimal as the resulting biodiesel-rich layer from the emulsification without the surfactant showed similar improvements in terms of water content, viscosity, TAN and HHV . Although the lignin -rich phase from Diacarbon’s emulsified bio -oil sample could not be analyzed, a water mass balance carried out on the layers from the emulsified bio-oil sample from Dynamotive confirmed that the majority of the water (~97%) is retained in the lignin -rich phase after emulsification.  This is significant in that it shows that the solvency of biodiesel can be utilized to upgrade bio-oils by selectively extracting its desirable fuel components into a biodiesel-rich phase, which can then be easily separated from the lignin-rich phase where the higher molecular weight compounds, such as pyrolytic lignin, as well as the majority of the water , have been left behind. Given bio -oil’s complex composition, extracting out its desirable fuel components to improve the fuel quality of biodiesel and simultaneously leaving behind its lignin -rich components creates the potential to divert would-be waste into more than two valuable product co-streams. The bio-oil samples obtained from the non-catalytic and catalytic vapour cracking experiments separated into two distinct layers –  aqueous and organic layer s. While the aqueous layers were fairly similar in nature, the organic layer from the catalytic experiment showed a significant decrease in viscosity (94.3% less), water content (64.3% less) and remained homogeneous after extraction. The organic layer from the non-catalytic run exhibited further stratification almost immediately after being extracted from the mixture, splitting into a hazy aqueous layer (with suspended oil droplets) sandwiched between a thin organic layer on top and a thicker organic layer at the bottom. Due to the time and material constraints in this project , the experiments were only conducted under a very limited selection of operating conditions and the bio-oil yields were insufficient to perform the emulsification with biodiesel. Nonetheless, this work has shown the     53 viability of the catalytic cracking  of pyrolysis vapours as a prospective ‘co -upgrading’ step to improve the quali ty of the crude bio-oil, which c ould potentially make subsequent upgrading steps, easier and more economical . In addition, the process of designing and building the catalytic test unit itself provided valuable insights into the technical challenges of bridging the gap between industry and academia. E ven a relatively simple catalytic unit such as this proved that there is no substitute for hands-on experience as there are things t hat are difficult to foresee (and thus, be able to plan ahead for) during the planning stages  (e.g., setup times, temperature fluctuations,  operational delays) and some that are simply beyond our control (e.g., weather, accidents, mechanical failures, etc.)  especially when operating beyond the confines of a controlled laboratory environment.  For pyrolysis companies, the addition of a catalytic vapour cracking unit has the po tential to add value to their product line, without the need for a complete overhaul of their existing equipment or infrastructure. It is recommended that future bio -oil upgrading research utilize catalytic vapour cracking in conjunction with  biodiesel emulsification and investigate just how much further bio-oil’s  fuel properties can be improved  or how much closer they can get to the ASTM standards set for diesel and biodiesel fuels. From the emulsification results obtained for this project, it would be exp ected that emulsifying the catalytically cracked bio-oil would yield further improvements in fuel quality. However, further o ptimization of the catalytic process conditions to improve bio -oil yields,  as well as the possibility of scaling up may also be worth looking into especially if emulsification is to be carried out afterward.  Overall , the driving force behind  this work and other research on biomass-based fuels is to reduce our reliance on petroleum-based fuels in order to support the three key aspects of energy security, energy equity and environmental sustainability. Given the proliferation of biomass from Canada’s forestry and agricultural sectors, the development of technologies utilizing biomass-based alternative fuels would enable us to transform what was supposed to be a waste stream into an energy stream instead. This not only creates the opportunity to reduce waste (and net CO 2 emissions) but also produces an alternative and renewable source of fuel (as well as p otential valu e-added chemicals).    54 References 1.  Herrmann L, Dunphy E, Copus J. Oil and Gas for Beginners. A Guide to the Oil & Gas Industry. Deutsche Bank AG/London; 2010.  2.  National Research Council. Hidden Costs of Energy: Unpriced Consequences of Energy Production and Use. Washington, D.C.: The National Academies Press; 2010.  3.  International Center for Technology Assessment (CTA). The Real Price of Gasoline. Washington, D.C.: International Center for Technology Assessment; 1998. Report No: 3.  4.  Hughes JD. Energy: A Reality Check on the Shale Revolution. Nature.  2013;494(7437):307–8.  5.  Vidic RD, Brantley SL, Vandenbossche JM, Yoxtheimer D, Abad JD. Impact of Shale Gas Development on Regional Water Quality [Internet]. Science. 2013 [cited 2014 May 12];340(6134). Available from: http://dx.doi.org/10.1126/science.1235009 6.  Gordalla BC, Ewers U, Frimmel FH. Hydraulic Fracturing: A Toxicological Threat for Groundwater and Drinking-water? Environ Earth Sci. 2013;70(8):3875–93.  7.  Davis S, Diegel S, Boundy R. Transportation Energy Data Book: Edition 32. Oak Ridge, Tennessee: Oak Ridge National Laboratory; 2013. Report No: ORNL-6989.  8.  World Energy Council. World Energy Trilemma: Time to get Real - The Case for Sustainable Energy Investment. London, United Kingdom: World Energy Council; 2013.  9.  U.S. Energy Information Administration. International Energy Outlook 2013 [Internet]. U.S. Department of Energy; 2013. Report No: DOE/EIA-0484(2013). [cited 2014 Mar 7]. Available from: www.eia.gov/forecasts/ieo/ pdf/0484(2013).pdf 10.  International Energy Agency. World Energy Outlook 2013. OECD Publishing; 2013.  11.  Kantarelis E, Yang W, Blasiak W. Biomass Pyrolysis for Energy and Fuel Production. In: Dahlquist E, editor. Technologies for Converting Biomass to Useful Energy: Combustion, Gasification, Pyrolysis, Torrefaction and Fermentation. CRC Press; 2013. p. 245–78.  12.  Venderbosch R, Prins W. Fast Pyrolysis. In: Brown RC, editor. Thermochemical Processing of Biomass. John Wiley & Sons, Ltd; 2011. p. 124–56 13.  McKendry P. Energy Production from Biomass (Part 1): Overview of Biomass. Bioresour Technol. 2002;83(1):37–46.  14.  Vakkilainen E, Kuparinen K, Heinimö J. Large Industrial Users of Energy Biomass [Internet]. IEA Bioenergy; 2013. [cited 2014 Mar 7]. Available from: http://www.bioenergytrade.org/downloads/t40-large-industrial-biomass-users.pdf     55 15.  Vassilev SV, Baxter D, Andersen LK, Vassileva CG. An Overview of the Chemical Composition of Biomass. Fuel. 2010;89(5):913–33.  16.  Brown RC. Biorenewable Resources: Engineering New Products from Agriculture. Ames, Iowa: Iowa State Press; 2003.  17.  Mohan D, Pittman, Steele PH. Pyrolysis of Wood/Biomass for Bio-oil:! A Critical Review. Energy Fuels. 2006;20(3):848–89.  18.  Ringer M, Putsche V, Scahill J. Large-Scale Pyrolysis Oil Production: A Technology Assessment and Economic Analysis [Internet]. Golden, Colorado: National Renewable Energy Laboratory; 2006. Report No: NREL/TP-510-37779. [cited 2013 Oct 21]. Available from: http://www.nrel.gov/docs/fy07osti/37779.pdf 19.  Sinha S, Jhalani A, Ravi MR, Ray A. Modelling of Pyrolysis in Wood: A Review. J Sol Energy Soc India SESI. 2000;10(1):41–62.  20.  Lange J-P. Lignocellulose Conversion: An Introduction to Chemistry, Process and Economics. Biofuels Bioprod Biorefining. 2007;1(1):39–48.  21.  Bridgwater AV. Upgrading Fast Pyrolysis Liquids. In: Brown RC, editor. Thermochemical Processing of Biomass. John Wiley & Sons, Ltd; 2011. p. 157–99.  22.  Beenackers AACM, Bridgwater AV. Gasification and Pyrolysis of Biomass in Europe. Pyrolysis and Gasification. Elsevier Applied Sciences; 1989. p. 129–55.  23.  Bridgwater AV, Bridge SA. A Review of Biomass Pyrolysis and Pyrolysis Technologies. In: Bridgwater AV, Grassi G, editors. Biomass Pyrolysis Liquids Upgrading and Utilization. Springer Netherlands; 1991. p. 11–92.  24.  Rosendahl L, editor. Biomass Combustion Science, Technology and Engineering. Sawston, Cambridge, UK: Woodhead Publishing Limited; 2013.  25.  Lehmann J, Joseph S. Biochar for Environmental Management: Science and Technology. Earthscan Publications Ltd.; 2009.  26.  Laird DA, Brown RC, Amonette JE, Lehmann J. Review of the Pyrolysis Platform for Coproducing Bio-oil and Biochar. Biofuels Bioprod Biorefining. 2009;3(5):547–62.  27.  Bridgwater A, Czernik S, Piskorz J. An Overview of Fast Pyrolysis. Progress in Thermochemical Biomass Conversion. Blackwell Science Ltd; 2001. p. 977–97.  28.  Suib SL, editor. New and Future Developments in Catalysis!: Catalytic Biomass Conversion. Amsterdam, NLD: Elsevier Science & Technology; 2013.  29.  Bridgwater AV. Review of Fast Pyrolysis of Biomass and Product Upgrading. Biomass Bioenergy. 2012;38:68–94.      56 30.  Butler E, Devlin G, Meier D, McDonnell K. A Review of Recent Laboratory Research and Commercial Developments in Fast Pyrolysis and Upgrading. Renew Sustain Energy Rev. 2011;15(8):4171–86.  31.  Venderbosch R, Prins W. Fast Pyrolysis Technology Development. Biofuels Bioprod Biorefining. 2010;4(2):178–208.  32.  Aston University Bioenergy Research Group. Biomass Pyrolysis - A Guide to UK Capabilities. Aston University; 2011.  33.  Oasmaa A, Kuoppala E, Gust S, Solantausta Y. Fast Pyrolysis of Forestry Residue. 1. Effect of Extractives on Phase Separation of Pyrolysis Liquids. Energy Fuels. 2003;17(1):1–12.  34.  Agblevor FA, Besler S. Inorganic Compounds in Biomass Feedstocks. 1. Effect on the Quality of Fast Pyrolysis Oils. Energy Fuels. 1996;10(2):293–8.  35.  Garcìa-Pérez M, Chaala A, Pakdel H, Kretschmer D, Rodrigue D, Roy C. Multiphase Structure of Bio-oils. Energy Fuels. 2005;20(1):364–75.  36.  Piskorz J., Scott D. S., Radlein D. Composition of Oils Obtained by Fast Pyrolysis of Different Woods. Pyrolysis Oils from Biomass. American Chemical Society; 1988. p. 167–78.  37.  Lehto J, Oasmaa A, Solantausta Y, Kytö M, Chiaramonti D. Fuel Oil Quality and Combustion of Fast Pyrolysis Bio-oils [Internet]. Espoo, VTT: VTT Technology; [cited 2014 Mar 17]. Available from: http://www.vtt.fi/inf/pdf/technology/2013/T87.pdf 38.  Czernik S, Bridgwater AV. Overview of Applications of Biomass Fast Pyrolysis Oil. Energy Fuels. 2004;18(2):590–8.  39.  Zhang Q, Chang J, Wang T, Xu Y. Review of Biomass Pyrolysis Oil Properties and Upgrading Research. Energy Convers Manag. 2007;48(1):87–92.  40.  Jain KK, Asthana RB. Automobile Engineering. West Patel Nagar, New Delhi: Tata McGraw-Hill Education; 2002.  41.  Diebold JP. A Review of the Chemical and Physical Mechanisms of the Storage Stability of Fast Pyrolysis Bio-Oils [Internet]. NREL; 2000 Jan. Report No: NREL/SR-570-27613. [cited 2013 Oct 21]. Available from: http://www.nrel.gov/docs/fy00osti/27613.pdf 42.  Oasmaa A, Peacocke C. Properties and Fuel Use of Biomass-derived Fast Pyrolysis Liquids [Internet]. VTT; 2010. Report No: P731. [cited 2014 Apr 22]. Available from: http://www.vtt.fi/inf/pdf/publications/2010/P731.pdf 43.  Solantausta Y, Oasmaa A, Sipilä K, Lindfors C, Lehto J, Autio J, et al. Bio-oil Production from Biomass: Steps toward Demonstration. Energy Fuels. 2012;26(1):233–40.      57 44.  Wagenaar B, Venderbosch R, Prins W, Pennicks F. Bio-oil as a Coal Substitute in a 600 MWe Power Station. Amsterdam, NLD; 2002.  45.  Bridgwater AV, editor. Pyrolysis and Gasification of Biomass and Waste: Proceedings of an Expert Meeting, Strasbourg, France, 30 September-1 October 2002. Newbury, UK: CPL Press; 2003.  46.  Corporate Communications FC. Fortum’s bio-oil plant commissioned in Joensuu - first of its kind in the world [Press Release]. The Wall Street Journal [Internet]. 2013; [cited 2014 Mar 17]. Available from: http://online.wsj.com/article/PR-CO-20131129-905979.html 47.  Effendi A, Gerhauser H, Bridgwater AV. Production of Renewable Phenolic Resins by Thermochemical Conversion of Biomass: A Review. Renew Sustain Energy Rev. 2008;12(8):2092–116.  48.  Chiaramonti D, Bonini M, Fratini E, Tondi G, Gartner K, Bridgwater AV, et al. Development of Emulsions from Biomass Pyrolysis Liquid and Diesel and their Use in Engines—Part 1: Emulsion Production. Biomass Bioenergy. 2003;25(1):85–99.  49.  Chiaramonti D, Bonini M, Fratini E, Tondi G, Gartner K, Bridgwater AV, et al. Development of Emulsions from Biomass Pyrolysis Liquid and Diesel and their Use in Engines—Part 2: Tests in Diesel Engines. Biomass Bioenergy. 2003;25(1):101–11.  50.  Diebold JP, Czernik S. Additives to Lower and Stabilize the Viscosity of Pyrolysis Oils during Storage. Energy Fuels. 1997;11(5):1081–91.  51.  Ikura M, Stanciulescu M, Hogan E. Emulsification of Pyrolysis Derived Bio-oil in Diesel Fuel. Biomass Bioenergy. 2003;24(3):221–32.  52.  Baglioni P, Chiaramonti D, Bonini M, Soldaini I, Tondi G. Bio-Crude-Oil/Diesel Oil Emulsification: Main Achievements of the Emulsification Process and Preliminary Results of Tests on Diesel Engine. In: Bridgwater AV, editor. Progress in Thermochemical Biomass Conversion. Blackwell Science Ltd; 2008. p. 1525–39.  53.  Churin E. Upgrading of Pyrolysis Oils by Hydrotreatment. In: Bridgwater AV, Grassi G, editors. Biomass Pyrolysis Liquids Upgrading and Utilization. Springer Netherlands; 1991. p. 103–17.  54.  Conti L, Scano G, Boufala J, Mascia S. Bio-Crude Oil Hydrotreating in a Continuous Bench-Scale Plant. In: Bridgwater AV, Boocock DGB, editors. Developments in Thermochemical Biomass Conversion. Springer Netherlands; 1997. p. 622–32.  55.  Wildschut J, Mahfud FH, Venderbosch RH, Heeres HJ. Hydrotreatment of Fast Pyrolysis Oil Using Heterogeneous Noble-Metal Catalysts. Ind Eng Chem Res. 2009;48(23):10324–34.      58 56.  French R, Czernik S. Catalytic Pyrolysis of Biomass for Biofuels Production. Fuel Process Technol. 2010;91(1):25–32.  57.  Czernik S. Catalytic Pyrolysis of Biomass. In: Lee JW, editor. Advanced Biofuels and Bioproducts [Internet]. Springer New York; 2013 [cited 2013 Mar 6]. p. 119–27. Available from: http://link.springer.com.ezproxy.library.ubc.ca/chapter/10.1007/978-1-4614-3348-4_9 58.  Xu J, Jiang J, Lv W, Dai W, Sun Y. Rice husk bio-oil upgrading by means of phase separation and the production of esters from the water phase, and novolac resins from the insoluble phase. Biomass Bioenergy. 2010;34(7):1059–63.  59.  Maggi RE, Elliott DC. Upgrading Overview. In: Bridgwater AV, Boocock DGB, editors. Developments in Thermochemical Biomass Conversion [Internet]. Springer Netherlands;  1997. [cited 2014 Mar 11]. p. 575–88. Available from: http://dx.doi.org/10.1007/978-94-009-1559-6_45 60.  Jacobson K, Maheria KC, Kumar Dalai A. Bio-oil Valorization: A Review. Renew Sustain Energy Rev. 2013;23:91–106.  61.  Jiang X, Ellis N. Upgrading Bio-oil through Emulsification with Biodiesel: Mixture Production. Energy Fuels. 2010;24(2):1358–64.  62.  Jiang X, Ellis N, Zhong Z. Fuel Properties of Bio-oil/bio-diesel Mixture Characterized by TG, FTIR and 1H NMR. Korean J Chem Eng. 2010;28:133–7.  63.  US Department of Energy. Alternative Fuels Data Center: Biodiesel Blends [Internet]. [cited 2014 Mar 19]. Available from: http://www.afdc.energy.gov/fuels/biodiesel_blends.html 64.  ASTM Standard D975, 2011b, “Specification for Diesel Fuel Oils.” West Conshohocken, PA: ASTM International; 2011.  65.  ASTM Standard D6751, 2011b, “Specification for Biodiesel Fuel Blend Stock (B100) for Middle Distillate Fuels.” West Conshohocken, PA: ASTM International; 2011.  66.  Vamvuka D. Bio-oil, Solid and Gaseous Biofuels from Biomass Pyrolysis Processes—An Overview. Int J Energy Res. 2011;35(10):835–62.  67.  Elliott DC. Historical Developments in Hydroprocessing Bio-oils. Energy Fuels. 2007;21(3):1792–815.  68.  Bulushev DA, Ross JRH. Catalysis for Conversion of Biomass to Fuels via Pyrolysis and Gasification: A Review. Catal Today. 2011;171(1):1–13.      59 69.  Gayubo AG, Aguayo AT, Atutxa A, Prieto R, Bilbao J. Deactivation of a HZSM-5 Zeolite Catalyst in the Transformation of the Aqueous Fraction of Biomass Pyrolysis Oil into Hydrocarbons. Energy Fuels. 2004;18(6):1640–7.  70.  Lappas AA, Iliopoulou EF, Kalogiannis KG. Catalysts in Biomass Pyrolysis. In: Crocker M, editor. Thermochemical Conversion of Biomass to Liquid Fuels and Chemicals. Royal Society of Chemistry; 2010. p. 263–83.  71.  Aho A, Kumar N, Eränen K, Salmi T, Hupa M, Murzin DY. Catalytic Pyrolysis of Biomass in a Fluidized Bed Reactor: Influence of the Acidity of H-Beta Zeolite. Process Saf Environ Prot. 2007;85(5):473–80.  72.  Arabiourrutia M, Olazar M, Aguado R, López G, Barona A, Bilbao J. HZSM-5 and HY Zeolite Catalyst Performance in the Pyrolysis of Tires in a Conical Spouted Bed Reactor. Ind Eng Chem Res. 2008;47(20):7600–9.  73.  Pattiya A, Titiloye JO, Bridgwater AV. Fast Pyrolysis of Cassava Rhizome in the Presence of Catalysts. J Anal Appl Pyrolysis. 2008;81(1):72–9.  74.  Pattiya A, Titiloye JO, Bridgwater AV. Evaluation of Catalytic Pyrolysis of Cassava Rhizome by Principal Component Analysis. Fuel. 2010;89(1):244–53.  75.  Myrén C, Hörnell C, Björnbom E, Sjöström K. Catalytic Tar Decomposition of Biomass Pyrolysis Gas with a Combination of Dolomite and Silica. Biomass Bioenergy. 2002;23(3):217–27.  76.  Kuhn JN, Zhao Z, Felix LG, Slimane RB, Choi CW, Ozkan US. Olivine Catalysts for Methane- and Tar-steam Reforming. Appl Catal B Environ. 2008;81(1–2):14–26.  77.  Adjaye JD, Bakhshi NN. Production of Hydrocarbons by Catalytic Upgrading of a Fast Pyrolysis Bio-oil. Part I: Conversion over Various Catalysts. Fuel Process Technol. 1995;45(3):161–83.  78.  Gayubo AG, Aguayo AT, Atutxa A, Aguado R, Bilbao J. Transformation of Oxygenate Components of Biomass Pyrolysis Oil on a HZSM-5 Zeolite. I. Alcohols and Phenols. Ind Eng Chem Res. 2004;43(11):2610–8.  79.  Gayubo AG, Aguayo AT, Atutxa A, Aguado R, Olazar M, Bilbao J. Transformation of Oxygenate Components of Biomass Pyrolysis Oil on a HZSM-5 Zeolite. II. Aldehydes, Ketones, and Acids. Ind Eng Chem Res. 2004;43(11):2619–26.  80.  Jiang X, Ellis N. Upgrading Bio-oil through Emulsification with Biodiesel: Thermal Stability. Energy Fuels. 2010;24(4):2699–706.  81.  ASTM Standard E203, 2008, “Test Method for Water Using Volumetric Karl Fischer Titration.” West Conshohocken, PA: ASTM International; 2008.      60 82.  ASTM Standard D664, 2011a, “Test Method for Acid Number of Petroleum Products by Potentiometric Titration.” ASTM International; 2011.  83.  Xu Y, Hu X, Li W, Shi Y. Preparation and Characterization of Bio-oil from Biomass. In: Shaukat S, editor. Progress in Biomass and Bioenergy Production [Internet]. InTech; 2011 [cited 2014 Feb 17]. Available from: http://www.intechopen.com/books/howtoreference/progress-in-biomass-and-bioenergy-production/preparation-and-characterization-of-bio-oil-from-biomass 84.  ASTM Standard D2709, 2011e1, “Test Method for Water and Sediment in Middle Distillate Fuels by Centrifuge.” West Conshohocken, PA: ASTM International; 2011.  85.  DynaMotive BioOil Book.pdf [Internet]. [cited 2012 Jun 22]. Available from: http://www.sustainablenorthwest.org/resources/biomass/DynaMotive%20BioOil%20Book.pdf 86.  Elliott DC, Hart TR, Neuenschwander GG, Rotness LJ, Zacher AH. Catalytic Hydroprocessing of Biomass Fast Pyrolysis Bio-oil to Produce Hydrocarbon Products. Environ Prog Sustain Energy. 2009;28(3):441–9.  87.  Scholze B, Meier D. Characterization of the water-insoluble fraction from pyrolysis oil (pyrolytic lignin). Part I. PY–GC/MS, FTIR, and functional groups. J Anal Appl Pyrolysis. 2001;60(1):41–54.  88.  Park HJ, Jeon J-K, Suh DJ, Suh Y-W, Heo HS, Park Y-K. Catalytic Vapor Cracking for Improvement of Bio-Oil Qual ity. Catal Surv Asia. 2011;15(3):161–80.  89.  Carlson TR, Tompsett GA, Conner WC, Huber GW. Aromatic Production from Catalytic Fast Pyrolysis of Biomass-Derived Feedstocks. Top Catal. 2009;52(3):241–52.  90.  Auerbach SM, Carrado KA, Dutta PK, editors. Handbook of Zeolite Science and Technology. CRC Press; 2003.  91.  Ramaswamy S, Huang H, Ramarao B, editors. Separation and Purification Technologies in Biorefineries. New York: John Wiley & Sons; 2013.  92.  Williams PT, Nugranad N. Comparison of Products from the Pyrolysis and Catalytic Pyrolysis of Rice Husks. Energy. 2000;25(6):493–513.  93.  Vitolo S, Seggiani M, Frediani P, Ambrosini G, Politi L. Catalytic Upgrading of Pyrolytic Oils to Fuel over Different Zeolites. Fuel. 1999;78(10):1147–59.  94.  Williams PT, Brindle AJ. Catalytic Pyrolysis of Tyres: Influence of Catalyst Temperature. Fuel. 2002;81(18):2425–34.      61 95.  Park HJ, Dong J-I, Jeon J-K, Yoo K-S, Yim J-H, Sohn JM, et al. Conversion of the Pyrolytic Vapor of Radiata Pine over Zeolites. J Ind Eng Chem. 2007;13(2):182–9.  96.  Mihalcik DJ, Boateng AA, Mullen CA, Goldberg NM. Packed-Bed Catalytic Cracking of Oak-Derived Pyrolytic Vapors. Ind Eng Chem Res. 2011;50(23):13304–12.  97.  Stuart PR, El-Halwagi MM. Integrated Biorefineries: Design, Analysis, and Optimization. CRC Press; 2012.        62 Appendix A:  Experimental Data and Results A.1 Sample Data and Results for Biodiesel Emulsification Experiments§§ Table A1. Total acid number for Diacarbon Tank 3 (collected June 13, 2012). Trial # Mass of Sample (g) End point (mL) Acid # (mg KOH/g) 1 0.1115 1.257 58.05 2 0.1037 1.048 51.11 3 0.1161 1.164 51.26   Average 53.47 ± 3.97   Table A2. Total acid number for Diacarbon Tank 3 (collected May 8, 2012). Trial # Mass of Sample (g) End point (mL) Acid # (mg KOH/g) 1 0.1440 1.310 47.01 2 0.1418 1.295 47.15 3 0.1272 1.191 47.97 4 0.1203 1.147 48.67   Average 47.70 ± 0.77  Table A3. Total acid number for Diacarbon Tank 2 (collected May 8, 2012). Trial # Mass of Sample (g) End point (mL) Acid # (mg KOH/g) 1 0.1116 1.659 70.82 2 0.1030 1.532 69.81 3 0.1023 1.559 71.77 4 0.1142 1.713 71.86   Average 71.07 ± 0.96                                                 §§ For Tables A1-7, to calculate the TAN for each trial, 0.103 mL (the average end point for the triplicate blank runs with only isopropanol) was subtracted from each endpoint.      63 Table A4. Total acid number for Diacarbon Vacuum Pump (collected May 8, 2012). Trial # Mass of Sample (g) End point (mL) Acid # (mg KOH/g) 1 0.1229 0.999 40.9 2 0.1058 0.919 43.3 3 0.1008 0.900 44.3   Average 42.8 ± 1.8  Table A5. Total acid number for biodiesel from Milligan Biotech. Trial # Mass of Sample (g) End point (mL) Acid # (mg KOH/g) 1 0.1827 0.200 6.14 2 0.2282 0.222 5.46 3 0.1818 0.200 6.17   Average 5.92 ± 0.40  Table A6. Total acid number for biodiesel-rich layer (40%bio-oil from Tank 3 - collected June 13, 2012, with 4vol% octanol). Trial # Mass of Sample (g) End point (mL) Acid # (mg KOH/g) 1 1.0022 3.097 16.76 2 1.0203 3.095 16.45 3 1.0200 3.096 16.46   Average 16.56 ± 0.18          64 Table A7. Total acid number for Biodiesel-rich layer (40%bio-oil from Tank 3 - collected June 13, 2012, no octanol). Trial # Mass of Sample (g) End point (mL) Acid # (mg KOH/g) 1 1.0101 2.690 13.55 2 1.0007 2.665 13.54 3 1.0005 2.848 14.57 4 1.0090 2.596 13.04   Average 14.50 ± 0.64  Table A8. Water content measurements for Diacarbon Tank 3 (collected June 13, 2012).  Trial # Mass of Sample (g) Water Content (wt.%)  1 0.0568 7.49 2 0.0843 7.40 3 0.0767 7.31  Average 7.40 ± 0.09  Table A9. Water content measurements for Diacarbon Tank 3 (collected May 8, 2012). Trial # Mass of Sample (g) Water Content (wt.%)  1 0.0371 82.7 2 0.0199 83.0 3 0.0179 84.0  Average 83.2 ± 0.7          65  Table A10. Water content measurements for Diacarbon Tank 2 (collected May 8 2012). Trial #  Mass of Sample (g) Water Content (wt.%)  1 0.0138 71.9  2 0.0127 72.7 3 0.0990  69. 9   Average 71.5 ± 1. 5   Table A11. Water content measurements for Diacarbon Vacuum Pump (collected May 8 , 2012). Trial #  Mass of Sample (g) Water Content (wt.%)  1 0.0526  22.1 2 0.053 0 23.5  3 0.2205  23.15   Average 22.9 ± 0.7   Table A12. Water content measurements for biodiesel from Milligan Biotech.  Trial #  Mass of Sample (g) Water Content (wt.%)  1 0.1883 0.21 2 0.2164  0.22 3 0.2294  0.21  Average 0.21 ± 0.01   Table A13. Water content measurements for Biodiesel -rich layer (40%bio -oil from Tank 3 - collected June 13, 2012, with 4vol% octanol) . Trial #  Mass of Sample (g) Water Content (wt.%)  1 0.4603  1.34 2 0.4245  1.28 3 0.3910  1.29   Average 1.30 ± 0.03       66 Table A14 . Water content measurements for Biodiesel -rich layer (40%bio -oil from Tank 3 - collected June 13, 2012, no octanol). Trial #  Mass of Sample (g)  Water Content (wt.%)  1 0.4304 0.95 2 0.5127 0.94 3 0.7523 0.94  Average  0.94 ± 0.01  Table A15 . Density measurements for Diacarbon Tank 2 (collected May 8 , 2012). Trial#  Density (g/L)  1 1085.7 2 1084.3 3 1088.8 4 1089.0 5 1085.7 A verage 1086.7 ± 1.9  Table A16 . Density measurements for Diacarbon Vacuum Pump (collected May 8, 2012).  Trial#  Density (g/L)  1 1031.8 2 1035.5 3 1037.9 A verage 1035.1 ± 3.1           67  Table A17 . Density measurements for Diacarbon Tank 3 (collected June 13, 2012).  Trial#  Density (g/ L) 1 984.3  2 989.1  3 990.3  4 988.8  Average 988.1  ± 2.6   Table A1 8 . Density measurements for Biodiesel-rich layer (40%bio -oil from Tank 3 - collected June 13, 2012, with 4vol% octanol) . Trial#  Density (g/ L) 1 931.8  2 931.9  3 944.4  4 944.4  5 931.6  Average 936.8 ± 6.9    Table A1 9 . Density measurements for Biodiesel-rich layer (40%bio -oil from Tank 3 - collected June 13, 2012, no octanol). Trial#  Density (g/L)  1 924.2  2 923.2  3 923.0  4 923.0  5 924.0  Average 923.5 ± 0.6        68  Table A20 . D ynamic viscosity measurements (at 40°C)  for Diacarbon T ank 2 (collected May 8 , 2012).     Trial #1  Trial #2  Trial #3  Speed Shear Rate  Dial Reading  Viscosity  Dial Reading  Viscosity  Dial Reading  Viscosity  RPM  (s-1) %Torque  cP  %Torque  cP  %Torque  cP  60 73.38  18.2  1.82  18.7  1.87  18.7  1.87  18.2  1.82  19.8  1.88  18.7  1.87  18.2  1.82  18.7  1.87  18.7  1.87  18.2  1.82  18.8  1.99  18.7  1.87  18.1  1.81  19.7  1.87  18.7  1.87  18.1  1.81  18.8  1.99  18.7  1.87  100 122.3 30.6 1.836  31.2 1.872  31.2 1.872  30.5 1.830  31.2 1.872  31.2 1.872  30.5 1.830  31.3 1.878  31.2 1.872  30.4 1.824  31.3 1.878  31.2 1.872  30.5 1.830  31.3 1.878  31.2 1.872  30.5 1.830  31.2 1.872  31.2 1.872                 69  Table A21 . Dynamic viscosity measurements (at 40 °C)  for Diacarbon Vacuum Pump (collected May 8, 2012).      Trial #1  Trial #2  Trial #3  Speed Shear Rate  Dial Reading  Viscosity Dial Reading  Viscosity Dial Reading  Viscosity RPM  (s-1) %Torque  cP %Torque  cP %Torque  cP 1.0 1.223  61.4  368.3  64.1  384.5  68.2  409.1  61.1  366.5  64.0  383.9  68.2  409.1  61.3  367.7  64.1  384.5  68.3  409.7  61.3  367.7  63.9  383.3  68.4  410.3 61.6  369.5  64.2  385.1  68.4  410.3 61.3  367.7  64.1  384.5  68.5  410.9  61.3  367.7  64.4  386.3  68.5  410.9  61.6  369.5  64.1  384.5  68.5  410.9  61.4  368.3  64.2  385.1  68.6  411.5  61.3  367.7  64.4  386.3  68.6  411.5  61.3  367.7  64.5  386.9  68.7  412.1  1.5  1.835  92.5  369.9  96.1  384.3  94.7  393.1  92.4  369.5  96.3  385.1  94.6  394.3  92.6  370.3  96.2  384.7  94.7  393.9  92.5  369.9  96.0  383.9  94.7  394.7  92.5  369.9  96.3  385.1  94.6  395.1  92.5  369.9  96.1  384.3  94.6  395.9  92.6  370.3  96.2  384.7  94.6  395.9  92.4  369.5  96.2  384.7  94.6  395.9  92.5  369.9  96.1  384.3  94.6  396.3  92.5  369.9  96.3  385.1  94.7  396.3  92.5  369.0  96.1  384.3  94.6  397.1       70 Table A22. Dynamic viscosity measurements (at 40°C)  for Diacarbon Tank 3 (collected June 13, 2012).     Trial #1  Trial #2  Trial #3  Speed Shear Rate  Dial Reading  Viscosity  Dial Reading  Viscosity  Dial Reading  Viscosity  RPM  (s-1) %Torque  cP  %Torque  cP  %Torque  cP  2.0 2.446 76.0 228.0  75.8  227.4 76.0 228.0  75.8  227.4 75.7 227.1 76.3 228.9  75.6 226.8  75.8  227.4 76.1 228.3  75.7 227.1 75.7 227.1 75.9 227.7 75.5 226.5 75.8  227.4 75.8  227.4 75.8  227.4 75.6 226.8  75.7 227.1 75.8  227.4 75.7 227.1 75.6 226.8  75.7 227.1 75.8  227.4 75.6 226.8  75.8  227.4 75.9 227.7 75.6 226.8  75.5 226.5 75.9 227.7 75.7 227.1 75.7 227.1 75.9 227.7 75.7 227.1 2.5 3.058  94.6 227.0 94.7 227.2 94.5 226.8  94.5 226.8  94.6 227.0 94.5 226.8  94.5 226.8  94.7 227.2 94.5 226.8  94.5 226.8  94.7 227.2 94.7 227.2 94.5 226.8  94.6 227.0 94.6 227.0 94.5 226.8  94.6 227.0 94.6 227.0 94.4 226.5 94.6 227.0 94.5 226.8  94.4 226.5 94.6 227.0 94.6 227.0 94.5 227.0 94.6 227.0 94.7 227.2 94.5 226.8  94.7 227.2 94.7 227.2 94.5 226.8  94.6 227.0 94.6 227.0      71 Table A23. Dynamic viscosity measurements (at 40°C) for Biodiesel-rich layer (40%bio-oil from Tank 3 - collected June 13, 2012, with 4vol% octanol).     Trial #1 Trial #2 Trial #3 Speed Shear Rate Dial Reading Viscosity Dial Reading Viscosity Dial Reading Viscosity RPM (s-1) %Torque cP %Torque cP %Torque cP 50 61.15 76.6 9.19 76.3 9.15 76.3 9.15 76.6 9.19 76.2 9.14 76.4 9.17 76.5 9.18 76,3 9.15 76.4 9.17 76.6 9.19 76.3 9.15 76.4 9.17 76.6 9.19 76.2 9.14 76.4 9.17 76.6 9.19 76.3 9.15 76.4 9.17 60 73.38 92.1 9.21 91.8 9.18 91.8 9.18 92.2 9.22 91.8 9.18 91.8 9.18 92.1 9.21 91.8 9.18 91.8 9.18 92.2 9.22 91.8 9.18 91.9 9.19 92.1 9.21 91.8 9.18 91.9 9.19 92.1 9.21 91.8 9.18 91.9 9.19                72  Table A24. Dynamic viscosity measurements (at 40°C) for Biodiesel-rich layer (40%bio -oil from Tank 3 - collected June 13, 2012, no octanol).     Trial #1  Trial #2  Trial #3  Speed Shear Rate  Dial Reading  Viscosity Dial Reading  Viscosity Dial Reading  Viscosity RPM  (s-1) %Torque  cP %Torque  cP %Torque  cP 50 61.15 70.5  8.46 76.3  9.15 76.3  9.15 70.5  8.46 76.2  9.14 76.4  9.17  70.4  8.45 76,3  9.15 76.4  9.17  70.4  8.45 76.3  9.15 76.4  9.17  70.5  8.46 76.2  9.14 76.4  9.17  70.4  8.45 76.3  9.15 76.4  9.17  60 73.38  84.6 8.46 91.8 9.18 91.8 9.18 84.6 8.46 91.8 9.18 91.8 9.18 84.6 8.46 91.8 9.18 91.8 9.18 84.7  8.47  91.8 9.18 91.9 9.19 84.6 8.46 91.8 9.18 91.9 9.19 84.6 8.46 91.8 9.18 91.9 9.19  Table A25. Stability measurements after emulsification of bio-oil with biodiesel at a 4:6 ratio with 4 vol.% octanol - total mixture volume of 150.0 mL. Trial #  Biodiesel-rich (upper) layer Lignin-rich (lower) layer Volume Ratio of Upper to Lower layer Stability (S) Volume (mL) Volume (mL) 1 125.0 25.0 83:17  0.447  2 126.0 24.0 84:16 0.458 3 126.0 24.0 84:16 0.458 Average 125.7 ± 0.6  24.3 ± 0.6  21:4 0.454 ± 0.007       73 Table A26. Stability measurements after emulsification of bio-oil with biodiesel at a 4:6 ratio with no octanol - total mixture volume of 150.0 mL.  Trial # Biodiesel-rich (upper) layer Lignin-rich (lower) layer Volume Ratio of Upper to Lower layer Stability (S) Volume (mL) Volume (mL) 1 133.0 17.0 89:11  0.478 2 132.0 18.0 88:12  0.467 3 133.0 17.0 89:11  0.478 Average 132.7 ± 0.6 17.3 ± 0.6 89:11  0.474 ± 0.006  Table A27. Mass of water in the different components prior to emulsification; the error range indicates the standard deviation for triplicate measurements of the same sample.  Sample Volume (mL) Water content (wt.%)  Density (g/mL) Mass of water (g) Bio-oil (Dynamotive) 57.6 24.45 ± 0.04 1.1606 ± 0.0062 16.3 Biodiesel (Milligan) 86.4 0.21 ± 0.05 0.8641 ± 0.0018 0.16 Octanol 6.00 0.21 ± 0.01 0.8270 ± 0.0013 0.010    Total 16.5  Table A28. Mass of water in the lignin-rich and biodiesel-rich phases after emulsification; the error range indicates the standard deviation for triplicate measurements of the same sample. Sample Volume (mL) Water content (wt.%) Density (g/mL) Mass of water (g) Lignin-rich Phase 50.0 24.86 ± 0.68 1.1733 ± 0.0006 14.6 Biodiesel-rich Phase 99.0 0.51 ± 0.01 0.8705 ± 0.0051 0.44     Total 15.0     74 A.2 Sample Data and Results for Catalytic Reactor Experiments Table A2 9. Water content measurements for Diacarbon  bio-oil from test unit (SiC top) (collected March 14, 2014).  Trial #  Sample Mass (g)  Water Content (wt.%)  1 0.0174 64.0 2 0.0169 64.1 3 0.0185 64.7 4 0.0183 64.1  Average  64.2 ± 0.3   Table A 30. Water content measurements for Diacarbon  bio-oil from test unit (SiC bottom) (collected March 14, 2014).  Trial #  Sample Mass (g)  Water Content (wt.%)  1 0.0323 30.6 2 0.0370 32.8 3 0.0381 31.0 4 0.0370 33.2  Average  31.9 ± 1.3   Table A 31. Water content measurements for Diacarbon  bio-oil from test unit (HZSM -5 top) (collected March 14, 2014).  Trial #  Sample Mass (g)  Water Content (wt.%)  1 0.0103 62.3 2 0.0186 63.0 3 0.0930 63.4 4 0.0255 63.6  Average  63.1 ± 0.6      75 Table A32. Water content measurements for Diacarbon bio-oil from test unit (HZSM -5 bottom) (collected March 14, 2014).  Trial #  Sample Mass (g) Water Content (wt.%)  1 0.0770 11.3 2 0.0520 11.2 3 0.0570 11.6 4 0.0560 11.6  Average 11.4 ± 0.2   Table A33. Density measurements for Diacarbon bio-oil from test unit (SiC top) (collected March 14, 2014).  Trial#  Density (g/L)  1 863.7  2 863.6  3 865.0  Average 864.3 ± 0.8   Table A34. Density measurements for Diacarbon bio-oil from test unit (SiC bottom) (collected March 14, 2014).  Trial#  Density (g/L)  1 850.9  2 859.8  3 851.2  4 859.0  Average 855.2 ± 4.8       76  Table A35. Density measurements for Diacarbon bio-oil from test unit (HZSM -5 top) (collected March 14, 2014).  Trial#  Density (g/L)  1  861.4  2  864.6  3 864.6  4  867.0  Average 865.4 ± 2.3   Table A36.  Density measurements for Diacarbon bio-oil from test unit (HZSM -5 bottom) (collected March 14, 2014).  Trial#  Density (g/L)  1  874.2  2  887.4  3 884.0  4  882.6  Average 884.7 ± 2.5              77  Table A37. Dynamic viscosity measurements (at 40°C)  for Diacarbon bio -oil from test unit (SiC top) (collected March 14, 2014).      Trial #1  Trial #2  Trial #3  Speed Shear Rate  Dial Reading  Viscosity Dial Reading  Viscosity Dial Reading  Viscosity RPM  (s-1) %Torque  cP %Torque  cP %Torque  cP 60  73.38  10.3 1.03 10.4 1.04 10.7  1.07  10.4 1.04 10.4 1.04 10.6  1.06  10.3 1.03 10.4 1.04 10.6  1.06  10.3 1.03 10.5  1.05  10.6  1.06  10.3 1.03 10.3 1.03 10.6  1.06  10.3 1.03 10.4 1.04 10.6  1.06  100 122.3  22.0  1.320  21.9  1.314 22.1  1.326  22.2  1.332  22.4  1.344 21.7  1.302  22.0  1.320  22.0  1.320  21.9  1.314 22.3  1.338  21.7  1.302  21.7  1.302  22.0  1.320  22.1  1.326  22.1  1.326  22.1  1.326  21.8  1.308  21.7  1.302                78  Table A38 . Dynamic viscosity measurements (at 40 °C)  for Diacarbon bio -oil from test unit (SiC bottom) (collected March 14, 2014).      Trial #1  Trial #2  Trial #3  Speed Shear Rate  Dial Reading  Viscosity Dial Reading  Viscosity Dial Reading  Viscosity RPM  (s-1) %Torque  cP %Torque  cP %Torque  cP 1.5  1.8345  32.1  128.4  30.6  122.4  31.8  127.2  30.0 120.0  30.4 121.6  34.8  138.9  33.2  132.8  30.6  122.4  27.6  110.4 33.4 133.6  30.7  122.8  28.6  114.4 33.2  132.8  30.7  122.8  29.7  118.8  30.8  123.2  30.4 121.6  32.7  132.8  1.0 1.223  32.5  195.0  29.8  178.8  40.2  241.1  26.2  157.2  29.2  175.2  41.8  250.7  34.5  207.0  30.0 180.0  42.3  253.7  34.6  207.6  30.0 180.0  40.0 239.9  34.7  208.2  31.6  189.6  43.5  260.9  35.2  211.2  30.9  185.4  43.7  262.1                79 Table A39. Dynamic viscosity measurements (at 40°C)  for Diacarbon bio-oil from test unit (HZSM -5 top) (collected March 14, 2014).      Trial #1  Trial #2  Trial #3  Speed Shear Rate  Dial Reading  Viscosity  Dial Reading  Viscosity  Dial Reading  Viscosity  RPM  (s-1) %Torque  cP  %Torque  cP  %Torque  cP  60 73.38  10.3 1.03 10.5 1.05 10.7 1.07 10.4 1.04 10.6 1.06 10.7 1.07 10.4 1.04 10.5 1.05 10.7 1.07 10.4 1.04 10.5 1.05 10.7 1.07 10.4 1.04 10.5 1.05 10.6 1.06 10.3 1.03 10.5 1.05 10.7 1.07 100 122.3 21.6 1.296 22.2 1.332 21.7 1.302 21.6 1.296 22.0 1.320 21.8  1.308  21.8  1.308  21.9 1.314 22.2 1.332 22.1 1.326 21.9 1.314 22.3 1.338  21.7 1.302 22.2 1.332 21.9 1.314 21.5 1.290 21.6 1.296 21.8  1.308                80 Table A40. Dynamic viscosity measurements (at 40°C) for Diacarbon bio-oil from test unit (HZSM -5 bottom) (collected March 14, 2014).      Trial #1 Trial #2 Trial #3 Speed Shear Rate Dial Reading Viscosity Dial Reading Viscosity Dial Reading Viscosity RPM (s-1) %Torque cP %Torque cP %Torque cP 20 24.46 49.1 14.73 40.7 12.21 45.1 13.53 47.8 14.34 41.3 12.39 45.2 13.56 47.1 14.13 41.0 12.30 44.8 13.40 46.9 14.07 41.1 12.33 44.5 13.35 46.4 13.92 41.0 12.30 44.3 13.29 46.2 13.86 41.0 12.30 44.0 13.20 30 36.69 64.1 12.82 59.0 11.80 62.7 12.54 63.4 12.68 59.9 11.98 62.4 12.48 63.4 12.68 58.8 11.76 62.9 12.58 63.7 12.74 59.9 11.98 61.9 12.38 63.7 12.74 58.8 11.76 62.0 12.40 63.6 12.72 59.7 11.94 61.8 12.36  Table A41. Temperature and pressure readings recorded for catalytic test unit run with SiC only (254-mm packed bed height). time T1 T2 T3 T4 T5 P1 P2 P3 (min) °C °C °C °C °C (in Hg)  (in Hg)  (in Hg)  0 496.4 507 632 94.7 1 -7 -11 -15 5 499.8 498 505 105.3 7.9 -1 -9 -19.5 10 500.0 340 507 265 8.8 -1 -9 -19.5 15 501.2 325 498 303 4.7 0 -5.5 -18      81  Table A42. Temperature and pressure readings recorded for catalytic test unit run with HZSM -5 catalyst (254 -mm packed bed height).  time T1  T2  T3  T4 T5  P1  P2  P3  (min)  °C  °C  °C  °C  °C  (in Hg)  (in Hg)  (in Hg)  0  499.9 570  502  132.6  7 -7 -14  -21  5  499.9 322  501  264  9 -1  -9 -19.5  10  499.3  326  503  285  4.4 -1  -9 -19.5  15  499.8 333  503  299  4.7 -1  -9 -19.5  20  500.2  340  501  307  5.6  -0.5  -8 -20  25  500 .0  344  506  314  8.6 -0.5  -9 -20  30  500.1  344  499 321  4.8 0  -9 -20.5      82 Appendix B :  Catalytic Reactor Design Supplementary Documentation  B.1  Experimental Procedure   As the unit was built and assembled in the department, taken apart then re-assembled out on site, the start-up procedure is a bit more involved.  Refer to Figure 4 -3 in Section 4.5 for the process flow diagram.  B.1.1 Start - up 1. Check that unit has been properly assembled  and aligned, that everything is connected. (If starting up a run after the reactor tube has been replaced, ensure that the Swagelok fittings on both ends of the tube are properly tightened and the fitting ends sit outside of the furnace vestibules.) Ensure that gate valve (V1) connecting to the main reactor unit is closed. Note: To check for leaks , open V2, run vacuum pump until pressure reading at P3 reads at least -2” Hg. Close V2, note the pressure readings at P1 and P2 and check that they stay reasonably constant for about an hour or so (open V3 to relieve vacuum  pump during this waiting time).  2. Open valve V3, then set heating tape controller (T I C1) and tube furnac e controller (TI C2) temperature to 500°C.  3. While waiting for the system to come to temperature, set up ice baths for the impinger train - A2 and A3:  a. For A2, fill container with a 50:50 mixture of ice and water (desired temperature is 0°C), insert dial therm ometer (T6) to monitor temperature. b. For the first half of A3, fill container with pure water and check that the temperature stays between 10° and 20°C, insert dial thermometer (T7) to monitor temperature c. For the latter half of A3, fill container with a 1:3  ratio of salt to ice and check that the temperature reaches about -20°C, insert dial thermometer (T8) to monitor temperature     83 d. For impinger flask set-up: (allow at least 30 minutes to cool the impinger bottles from ambient temperature down to -20°C) i. Fill Flask 1 with 600 mL of isopropanol (if succeeding runs are <1 hour long, volume may be reduced to 500 mL)  ii. Fill Flasks 2 and 3 with 500 mL of isopropanol  iii. Keep Flask 4 empty and loosely fill about halfway with glass wool.  4. Once system has rea ched the desired temperature (TIC1 AND TI1  reads ~500°C for T1 and T2, respectively, and TIC2 reads 500°C for T3), close valve V3 then turn on vacuum pump (A4). 5. When the pressure reading on P3 is lesser than or equal to -0.5 in Hg (-7” H2O or -0.25 psig), open up gate valve (V1) and valve on main reactor unit, adjust the flow rate on the rotameter (R1) using V2 to !5.6 cfm using (V3).  6. Record the pressure reading at P1 and P2 after vacuum is turned on and after gate valve is opened. 7. Check temperature at T1 (should be around 500°C), T4 (between 450° and 500°C) and T5 (around 200°C) as vapour flows through the system and starts condensing.  B.1.2 Experimental 1. Check consistency of bio-oil sample collecting in flask and ensure that after the second or third flask in the impinger train, there are no more condensates on the tubing heading towards the vacuum pump. 2. Record temperature at T2 and T4 every 5-10 minutes, note the time interval and increase/decrease depending on how much the temperature changes over time.  3. Record pressure r eading at P1, P2 and P3 every 5-10 minutes, note the time intervals and adjust accordingly depending on how much the readings change over time. If the pressure reading at P3 starts increasing, close V1 slightly until pressure decreases to ! -0.5 in Hg.      84 4. Monitor the flow rate reading from the rotameter (R1). 5. Check ice bath temperature for A2 (0°C) and A3 (first half: 10-20°C and latter half: -20°C) every 15 minutes, drain and replace water and/or add more ice/salt as necessary. 6. Run may be terminated once the bio-oil sample volume in A2 has reached ~2 L or when no more sample is being collected, whichever comes first. B.1.3 Shut Down 1. Close gate valve (V1) – be sure to wear PPE (high T gloves, hard hat, eye protection and coveralls) 2. Turn off vacuum pump (A4), simultaneously disconnecting the acetal quick-disconnect coupling between the sample collection flask and the first impinger flask, then fully open valve V3 to open system to atmosphere.  3. Turn off heating tape controller (TIC1) and tube furnace controller (TIC2) and allow system to cool down. 4. Disconnect PE tubing connections and transfer the collected samples (from A2 and A3) to sample jars and store in ice chest.  5. Rinse lines and flasks with isopropanol into a labelled waste container jar. 6. If equipment is to be left outdoors for additional experiments on another day, bring back glassware (for washing and storage) and cover unit with tarp (if necessary) after it has cooled down. Wrap electrical plugs with plastic and put vacuum pump into storage. Unit may be disassembled once experiments have been completed.    85 B.2 Legend for Process Flow Diagram  Label Description  Line Size  Material  Supplier/ Manufacturer  Model/Part No.  V1 Bellows Sealed Gate Valve, 1 -1/2" Female NPT Ends 1-1/2"  Stainless Steel/Graphite  AR Thomson/Dixon Eagle H8C22SE-150 TIC1 Temperature Control Box (with built-in indicator) N/A  ---  Assembled by UBC workshop, parts from Omega   --- T1 K -type thermocouple 1/8"  Stainless Steel Omega  KQXL -18U-12 T2 K -type thermocouple 1/8"  Stainless Steel Omega  KQXL -18U-24 P1 4" Dial Size Pressure Gauge ( -30HG /0 KPA Dual Scale Range) 304SS, Liquid Filled Case  1/4" NPT Lower Connection 316 Stainless Steel Wetted Parts Corix/Wika WIKA 233.54/ 7297509 TIC2 Temperature Controller (Built into Furnace), SSD output N/A  Plastic Case Mellen Company Inc./Omega  Omega CN96211TR  T3 K -type thermocouple (supplied with Furnace) connected to TC N/A  Stainless Steel Mellen Company Inc. --- A1 Tube furnace housing reactor tube 1"  Stainless Steel, Ceramic Mellen Company Inc. MTSC12.5R-1.25x12.00 TI1 Double Input Type K switchable 0.1 or 1° resolution handheld digital thermometer (for T2 and T4) N/A  N/A  Omega  HH12B P2 4" Dial Size Pressure Gauge ( -30HG /0 KPA Dual Scale Range) 304SS, Liquid Filled Case  1/4" NPT Lower Connection 316 Stainless Steel Wetted Parts Corix/Wika WIKA 233.54/ 7297509 T4 K -type thermocouple 1/8"  Stainless Steel Omega  KQXL -18U-10     86 C1 1/2" Cooling Coil 1/2" Stainless Steel Unified Alloys ---  T5 K-type thermocouple 1/8" Stainless Steel Omega KQXL-18U-10 TI2 Double Input Type K switchable 0.1 or 1° resolution handheld digital thermometer (for T2 and T4) N/A N/A Omega HH12B T6 Clip-on dial thermometer, range -10° to 110° C 3/16" stem 304 stainless steel McMaster-Carr 4101K31 A2 Vacuum Flask Collection Vessel 1/2" Glass Fisher Scientific   T7 Clip-on dial thermometer, Range -10° to 110° C 3/16" stem 304 stainless steel McMaster-Carr 4101K31 T8 Clip-on dial thermometer, Range -40° to 160° F 3/16" stem 304 stainless steel McMaster-Carr 4101K31 A3 Tar Impinger Train - 4 Vacuum Filter Flasks in Series 1/2" Glass Fisher Scientific  --- V2 High-Pressure needle valve, 1/2" NPT Female Ends 1/2" 316 Stainless Steel McMaster-Carr 4644K34 R1 OEM-style Acrylic Rotameter (Air Range: 14.5 SCFM), 1/2" FNPT 1/2" 316 SS float and guide rod Omega FL7313 V3 Needle valve to adjust vacuum pressure 1/2" 316 Stainless Steel McMaster-Carr 4644K34 P3 Liquid-Filled SS Vacuum & Compound Gauge, 2-1/2" Dial  1/4" NPT Lower Connection 304/316 Stainless Steel McMaster-Carr 38605K1 A4 Chemstar Vacuum Pump (for pumping corrosive gases) 115 V, 60Hz, 1ph, 5.6 cfm, 1x10-4 Torr 13/16" tube ID --- Fisher/Welch Welch 1402N-01        87  B.3 HAZOPS Parameter/ Process Variable Deviation Consequences Causes Existing Safeguard Action Items/ Recommendations Temperature at T1 Low 1. Vapour condensation in pipe if T1 < 400°C  Insufficient Insulation Additional insulation available Stop run, add insulation (and heat trace if necessary) 2. Temperature would drop too much below 500°C before it reaches reactor Heating tape can be cranked up as high as 760°C if necessary  Temperature difference between T2 and T3 High  Depending on the temperature deviation, results may be affected since the temperature is not as it was set to be.  Furnace thermocouple signal relay may be faulty or faulty thermocouple Temperature controller, extra thermocouples Replace thermocouple. Adjust temperature set point on furnace temperature controller   Endothermic reaction in catalyst bed Temperature at T2 and T3 Runaway Overheating, experiment failure  Electrical failure Thermocouple, tube furnace controller Kill heater, shut down unit Vacuum Pressure at P1 Low (i.e., too positive) 1. No/low vapour flow, no sample collected Insufficient vacuum, leaks Pressure relief needle valve Check system for leaks Catalyst bed blockage or too much pressure drop Removable reactor tube, extra reactor tube Adjust needle valve opening Vacuum Pressure at P2, P3 Too Low (i.e., too positive) 1. No vapour flow, no sample collected Insufficient vacuum, leaks Pressure relief needle valve Adjust needle valve opening     88  High (i.e. , too negative) 2. Glassware may implode  Catalyst bed blockage, clogging or kink in tubing, vacuum may be low on oil or oil may be contaminated Pressure relief needle valve, duct tape on glassware, extra tubing, extra glassware, extra oil Adjust needle valve opening, shut down pump and refill/change the oil  Temperature at T5 High  1. Ice bath and impinger train may not cool the vapour down fast enough for complete condensation Outside air temperature may not be cold enough Salt, ice Adjust vapour flow rate; drain and replace water in bucket containing the cooling coil 2. Loss of sample due to incomplete condensation Cooling coil area may be insufficient Extra cooling coil available, chiller bucket  Add more ice and/or salt to lower ice bath temperature 3. Clogging of vacuum pump Flow rate may be too high Rotameter  Add in extra cooling coil Low  1. Condensation in cooling coil  Flow rate may be too low Rotameter  Increase flow rate  2. Fouling and blockage of coil Cooling coil area may be too much Flow High  1. Vapours may not have time to react in catalyst bed Too much pull from vacuum Rotameter  Decrease flow rate 2. Incomplete condensation Low  1. Vapour may condense too early and clog up the coil Leaks  Increase flow rate, check for system leaks 2. Sample collection will take longer Catalyst bed blockage       89 B.4 Emergency Shut Down  1. Kill power source (power bar or generator) or unplug or turn off vacuum pump, tube furnace and heating tape controller.  2. Close bellows gate valve (V1) leading to main reactor unit.  3. Fully open needle valve, V3, to open system to atmosphere. 4. Disconnect PE tubing connection to vacuum pump. B.5 Test Unit Diagrams and Photos   Figure B-1. Schematic diagram of packed bed reactor.  90  Figure B-2. Diagram of tube furnace (courtesy of The Mellen Company, Inc., all units in mm).     91  Figure B-3. Diagram of catalytic test unit with spatial dimensions (after the cooling coil, the distances are relatively more flexible with the exception of V2, R1, V3 and P3, which are fixed onto the same frame holding the packed bed reactor in place).     92   Figure B-4. Close-up of catalytic reactor.     93  Figure B-5. Catalytic reactor test unit attached to Diacarbon’s pilot plant.     94 Appendix C: Sample Calculations  C.1 Total Acid Number  Record the mass of the sample used for testing then note the volume of the titrant at the endpoint as indicated on the display of the Metrohm 794 Basic Titrino, TAN = (EP*C01*C02)/C00  where: TAN  = total acid number (mg KOH/ g of sample)    EP  = volume of titrant at the endpoint  (mL)   C01  = concentration of titrant , usually a constant (0.1 M KOH)   C02  = molecular weight of titrant  (MWK OH = 56.1056 g/mol)   C00  = mass of sample  (g) Example: Using data from Table A1, Trial #1 , and subtracting 0.103 mL (the endpoint average of the triplicate blank runs with isopropanol only) TAN  = (1.257 mL  –  0.103 mL)*(0.1 M)*(56.1056 g/mol)/(0.1115 g)  = 58.05 mg KOH/g of sample   C.2 Stability  For a 150.0-mL mixture of bio-oil and biodiesel at a 4:6 ratio, resulting in a biodiesel-rich layer of 133.0 mL and a lignin-rich layer of 17.0 mL, the stability (S) can be calculated by: S = (volume of biodiesel -rich layer –  original volume of biodiesel)/ (volume of biodiesel) S = (133.0 –  90.0)/ 90.0 = 0.478   For a 150.0-mL mixture of bio-oil and biodiesel at a 4:6 ratio and 4 vol.% octanol, resulting in a biodiesel-rich layer of 125.0 mL and a lignin-rich layer of 25.0 mL, the stability (S) can be calculated by: S = (125.0 –  86.4)/86.4 = 0.447      95 C.3 Calculations for Cooling Coil Sizing Fluid1 volumetric flow rate, V1 = 160 L/min   Density of pyrolysis vapour, !1 = 1.08 kg/m3 (assumed from density of pyrolysis oil) Fluid1 mass flow rate, m1 = 10.4 kg/hr V1*60* !1/1000 Fluid1 temp. in, T1in = 500 °C   Fluid1 temp. out, T1out = 200 °C   Fluid1 sp. heat, Cp1 = 3.8 kJ/kg-°C http://www.vtt.fi/inf/pdf/publications/2001/P450.pdf Outside Air or Cooling Water Temperature, T2 15 °C   Overall heat transfer coefficient, estimated, U = 50 J/sec-m2-°C http://www.engineeringpage.com/technology/thermal/transfer.html Heat Transfer Rate, Q = 11819.5 kJ/hr m1* Cp1*( T1in-T1out) Log Mean Temperature Diff, DTlm = 311.3 °C ((T1in-T2)-(T1out-T2))/ln((T1in-T2)/( T1out-T2)) Heat Transfer Area for 1” SS Tee = 0.009 m2 Dimensions: L = 97 mm, ID = 29.7 mm, OD = 36.3 mm Thermal conductivity of SS316L, k (at 500°C) 21.4 W/m-K http://www.lenntech.com/stainless-steel-316l.htm Heat flux through 1” SS Tee, q = 0.36 kW k*(T1in-T2)/(OD-ID)/1000*heat transfer area for 1” SS Tee Calculated Area required, A = 0.187 m2 ((Q/3600)-q)*1000/U/DTlm Length of "” SS tubing required = 6.096 m Dimensions: OD = 13mm, ID =10mm; A/(#*ID)         96  C.4 Power Calculations for Heating Requirements C.4.1 Power Requirements for Heat Loss in Larger Pipe Section (from main unit furnace to port) Pipe Size, OD 6.625 in   Specific heat of cast iron 0.12 BTU/lb-F Source: http://www.ramacorporation.com/engineer/Engineerweb.pdf   p. 12, Table 7 Density of cast iron 0.2598  lb/in3 Volume of empty tube 200.93  in3             0.00286  m3  Pipe ID = 6.065”, length = 36”  Mass of cast iron pipe 52.2 lbs 20.63  kg (volume of empty tube*density of cast iron)  Heat up time desired (estimated) 0.5 hr      Volume of space inside tube 1040.05 in3 0.017 m3   Mass of vapour stream 0.0184 kg 0.0405 lbs   Specific heat of fluid 3.8 kJ/kg -oC 0.908  BTU/lb-F  http://www.vtt.fi/inf/pdf/publications/2001/P450.pdf Temperature difference between cold pipe surface and desired vapour stream temp. 891  oF Assume outside temp. to be = ~5  oC, desired temperature of vapour stream = 50 0oC Thermal conductivity of fiberglass insulation 0.5 (BTU-in/ft2-F-hr) http://www.engineeringtoolbox.com/fiberglas-insulation-k-values-d_1172.html    Step 1. Calculate power required to heat the equipment (i.e. the pipe) in contact with the material to be heated (i.e., wattage required for heating at start-up) = weight of material*specific heat*temp. difference/[(3.412 BTU/Whr)*Time allowed for heat up]   Power needed to heat pipe: 3272.08 Watts http://www.ramacorporation.com/engineer/Engineerweb.pdf      97 Step 2. Power needed to heat the added material (i.e. the vapour stream)  = Weight of material*specific heat*temp. difference/[(3.412 BTU/Whr)*Time allowed for heat up]  Time allowed to heat up material  0.00028 hr  (Assumed that vapour flowing through needs to be heated as fast as possible, i.e., per second) Power needed to heat vapour stream:  34566.8 Watts     Step 3. Power loss from surfaces =wattage loss/ft*surface area*time of heat loss  Surface area 818.213 in2 5.68 ft2     Insulation thickness 1 inch    Wattage loss/ft (for 1" insulation , dT=900  oF ) 1 Watts/in 2 144 Watts/ft 2 < ---- heat loss factor for insulated metal surface from graph, http://www.ramacorporation.com/engineer/Engineerweb.pdf) Wattage loss/ft (for 2" insulation dT=900  oF ) 0.5 Watts/in 2 72 Watts/ft 2 Wat tage loss/ft (no insulation, dTave=445 F)  300 Watts/in 2 43200 Watts/ft 2 T ime of heat loss 0.5 hr   Conductive heat loss (1" insulation)  409.11 Watt -hrs < --- using heat loss factors Conductive heat loss (2" insulation)  204.55 Watt -hrs Uninsulated  122731.96 Watt -hrs  Step 4. Start-up power required  =  [step 1 + (2/3)*step 3/exposure time]*1.15   With 1" insulation  With 2" insulation      Safety factor 15%      Exposure time  0.50 hr 0.50 hr     Start-up power =  4390.19 Watts  4076.55 Watts          98 Step 5. Operating (or Maintenance) Power=[(step 2 + (step 3/cycle time)]*1.15  With 1" insulation With 2" insulation   Cycle time 0.75 hr 0.75 hr     Operating power = 40379.1 Watts 40065.4 Watts                   Total power (start up + operating power) = 44769.26 Watts 44141.96 Watts      C.4.2  Power Requirements for Heat Loss between Port and Packed Bed Reactor Part/section OD (inch) Body length (inch) ID (inch) Volume of material (in3) Volume inside (in3)  Tapered neck of pipe reducer 2.375 4 2.067 17.72 13.42 http://www.mcnichols.com/viewer.htm?pageCode=pipedims ! ” connector section 0.75 5 0.701 2.21 1.93  ! ” stainless steel cross 1.25 6.84 0.75 8.39 3.02    Specific heat of iron/stainless steel 0.12 BTU/lb-F Source: http://www.ramacorporation.com/engineer/Engineerweb.pdf; p. 12, Table 7, for these calculations, the specific heat and material densities for iron and SS are assumed to be the same… Density of iron/stainless steel 0.2598 lb/in3 Total volume of the diff. parts  28.32 in3             0.00046  m3 Sum of all the inside volumes of the diff. parts Total mass of connecting parts 8.2 lbs 3.72  kg (Total material volume of parts*density of material) Heat up time desired (estimated) 0.5 hr      Volume of space inside tube 18.37 in3 0.0003 m3       99 Mass of vapour stream 0.00033 kg 0.00072 lbs   Specific heat of fluid 3.8 kJ/kg -oC 0.908 BTU/lb-oF http://www.vtt.fi/inf/pdf/publications/2001/P450.pdf Temperature difference between pipe surface and vapour stream 891 oF Assume outside temp. to be = ~5  oC, desired temperature of vapour stream = 500oC Thermal conductivity of fiberglass insulation 0.5 (BTU-in/ft2-oF-hr) http://www.engineeringtoolbox.com/fiberglas-insulation-k-values-d_1172.html   Step 1. Calculate power required to heat the equipment (i.e. the pipe) in contact with the material to be heated (i.e., wattage required for heating at start-up) = weight of material*specific heat*temp. difference/[(3.412 BTU/Whr)*Time allowed for heat up]   Power needed to heat pipe: 514.78 Watts http://www.ramacorporation.com/engineer/Engineerweb.pdf   Step 2. Power needed to heat the added material (i.e. the vapour stream)  = Weight of material*specific heat*temp. difference/[(3.412 BTU/W hr)*Time allowed for heat up]  Time allowed to heat up material 0.00028 hr (Assumed that vapour flowing through needs to be heated as fast as possible, i.e., per second) Power needed to heat vapour stream:  610.67 Watts    Step 3. Power loss from surfaces =wattage loss/ft*surface area*time of heat loss  Surface area 80.68 in2 0.516 ft2   Insulation thickness 1 inch  Wattage loss/ft (for 1" insulation, dT=900  oF) 1 Watts/in2 144 Watts/ft2 <---- heat loss factor for insulated metal surface from graph, http://www.ramacorporation.com/engineer/Engineerweb.pdf) Wattage loss/ft (for 2" insulation dT=900  oF) 0.5 Watts/in2 72 Watts/ft2 Wattage loss/ft (no insulation, dTave=445 F)  300 Watts/in2 43200 Watts/ft2     100 Time of heat loss 0.5 hr   Conductive heat loss (1" insulation)  40.34 Watt-hrs < --- using heat loss factors  Conductive heat loss (2" insulation)  20.17 Watt-hrs Without insulation 12102.74 Watt-hrs   Step 4. Start-up power required = [step 1 + (2/3)*step 3/exposure time]*1.15   With 1" insulation With 2" insulation      Safety factor 15%     Exposure time  0.50 hr 0.50 hr     Start-up power =  653.866 Watts 622.93 Watts      Step 5. Operating (or Maintenance) Power=[(step 2 + (step 3/cycle time)]*1.15       With 1" insulation  With 2" insulation   Cycle time  0.75 hr 0.75 hr     Operating power =  764.13 Watts 733.2 Watts                   Total power (start up + operating power) =  1417.9 Watts 1356.13 Watts        

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0167466/manifest

Comment

Related Items