UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Ethanol production from bio-oil Chan, Jacky 2009

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

Item Metadata

Download

Media
24-ubc_2009_spring_chan_jacky.pdf [ 2.53MB ]
Metadata
JSON: 24-1.0058763.json
JSON-LD: 24-1.0058763-ld.json
RDF/XML (Pretty): 24-1.0058763-rdf.xml
RDF/JSON: 24-1.0058763-rdf.json
Turtle: 24-1.0058763-turtle.txt
N-Triples: 24-1.0058763-rdf-ntriples.txt
Original Record: 24-1.0058763-source.json
Full Text
24-1.0058763-fulltext.txt
Citation
24-1.0058763.ris

Full Text

ETHANOL PRODUCTION FROM BIO-OIL  by  JACKY CHAN  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Chemical and Biological Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) APRIL 2009 © Jacky Chan, 2009  Abstract Bioethanol is often viewed as one of the solutions to the tight gasoline supplies in North America. Although bioethanol is already available in the market, there are a number of problems associated with the current processes for the production of bioethanol. The current bioethanol production processes are often referred as first generation bioethanol production processes. For these first generation processes, the feedstocks for production are usually energy crops. The most common energy crops in North America are corn and wheat. The use of these energy crops has triggered debates on the problems associated with using food sources to create energy and the uptake of agricultural land to produce energy. In this project, an alternative feedstock for bioethanol is investigated. The feedstock used in the project is bio-oil, which can be derived from any biomass waste. An advantage of using bio-oil is that it is not derived from food crops but instead waste material is being converted into energy. The objective of this study was to determine the technical viability of producing bio ethanol using bio-oil as a substrate for fermentation. In order to maximize the ethanol yield, the extraction of levoglucosan with water was optimized and a number of detoxification techniques for inhibitor removal were evaluated. This report provides a technical overview of conditions evaluated for extracting levoglucosan from bio-oil, and methods used for improving the fermentability of bio-oil hydrolysate by detoxification. The techniques used in an attempt to improve the fermentability of bio-oil hydrolysate include: adsorption, overliming, solvent extraction, and hydrogenation. In addition, a biological approach called adaptive evolution was used to aid the yeast to adapt to the inhibitory environment of bio-oil hydrolysate in order to increase their resistance to inhibitors. The optimal condition for aqueous extraction of levoglucosan from bio-oil was found to be 1:1 (mass water to mass bio-oil). It was found that the temperatures examined (25°C and 80°C) had minimal effect on the amount of levoglucosan extracted. Among the  11  detoxification techniques tested, it was found that overliming and solvent extraction were able to improve the fermentability of bio-oil hydrolysates. Overliming was able to increase the yield of ethanol from bio-oil hydrolysate by 0.19±0.01 (g ethanol/g glucose) at 50% strength hydrolysate and 0.45±0.05 (g ethanol/g glucose) at 40% strength hydrolysate. A number of extractants were examined and the three best solvents were 25% volume of tri-n-octylamine with co-solvent 1-octanol, 50% volume of alamine 336 with co-solvent 1-octanol and oleyl alcohol. These three solvents were able to selectively remove at least 84  —  93% of acetic acid, which was the targeted inhibitor in bio-oil  hydrolysate. In addition, a technique called adaptive evolution of yeasts was applied, which was capable of increasing the ethanol yield by at least 6% when compared with the unadapted parental strains.  111  Table of Contents Abstract  .  Table of Contents  ii iv  List of Tables  viii  List of Figures  x  Acknowledgements  xiii  Chapter 1 Introduction 1.1 Background  I  1.2 Ethanol  2  1.2.1 Overview of ethanol production  6  1.2.2 Substrates for bioethanol  7  1.2.3 Pyrolysis and other substrate pretreatments methods  7  1.3 Characteristics of bio—oil  10  1.3.1 Compositions of bio-oil  11  1.3.2 Phase behavior and stability of bio-oil  12  1.3.3 Anhydrosugars in bio-oil  15  1.3.4 Extraction of levoglucosan from bio-oil  16  1.3.5 Hydrolysis of levoglucosan and other anhydrosugars  18  1.4 Yeasts  18  1.4.1 Glucose metabolism  20  1.4.2 Glycolysis mechanism  20  1.4.3 Utilization of pyruvate under aerobic conditions and the tricarboxylic acid cycle  22  iv  1.4.4 Utilization of pyruvate under anaerobic conditions  22  1.4.5 Comparison of aerobic and anaerobic conditions  24  1.4.6 Parameters affecting fermentation  26  1.4.7 Inhibition of fermentation  26  1.5 Improving the fermentability of bio-oil hydrolysates  28  1.5.1 Air stripping  28  1.5.2 Adsorption on activated carbon  29  1.5.3 Overliming  29  1.5.4 Coagulation and flocculation  30  1.5.5 Solvent extraction  30  1.5.6 Hydrogenation  31  1.5.7 Adaptive evolution  32  Chapter 2 Research Objectives  33  Chapter 3 Materials and Methods  34  3.1 Process overview  34  3.2 Procedure for extraction of levoglucosan  34  3.2.1 Storage and preparation of bio-oil  35  3.2.2 Titration of bio-oil to determine the critical amount of water required to induce phase separation  35  3.2.3 Levoglucosan extraction from bio-oil  37  3.3 Dilute acid hydrolysis of the aqueous fraction of bio-oil  39  3.4 Procedure for fermentation of bio-oil hydrolysate  40  3.4.1 Preparation of yeast inoculum for fermentation  V  40  3.4.2 Fermentation of bio-oil hydrolysate 3.5 Procedure for detoxification of bio-oil hydrolysates  42 43  3.5.1 Air stripping  43  3.5.2 Adsorption on activated carbon  44  3.5.3 Overliming  44  3.5.4 Coagulation and flocculation  45  3.5.5 Solvent extraction for detoxification of bio-oil hydrolysate  46  3.5.6 Hydrogenation for detoxification of bio-oil hydrolysate  47  3.5.7 Adaptive evolution  50  3.6 Analytical methods  52  3.6.1 Determining yeast concentration  52  3.6.2 Analysis of ethanol and levoglucosan by gas chromatography  53  3.6.3 Analysis of sugars by high pressure liquid chromatography  54  Chapter 4 Levoglucosan extraction from bio-oil  56  4.1 Bio-oil titration to determine phase behavior  56  4.2 Suspended solids in aqueous phase  56  4.3 Effect of temperature on levoglucosan extraction  57  4.4 Effect of the order of solvent addition on levoglucosan extraction  57  4.5 Effect of mixing on levoglucosan extraction  58  4.6 Optimal water-to-bio-oil ratio for levoglucosan extraction  59  4.7 Acetic acid in aqueous phase  61  4.8 Other compounds in aqueous phase  62  Chapter 5 Methods for improving the fermentability of bio-oil hydrolysate  vi  64  5.1 Effect of hydrolysate concentration on fermentation by S. cerevisiae  64  5.2 Detoxification of artificial aqueous phase and hydrolysate by hydrogenation  65  5.3 Detoxification using activated carbon, air stripping and overliming  68  5.4 Detoxification by coagulation and flocculation  69  5.5 Detoxification by extraction with organic solvents  70  5.5.1 Solvent extraction for acetic acid removal from artificial bio-oil hydrolysate.71 5.5.2 Evaluation of the toxicity of the solvents towards fermentation  73  5.5.3 Fermentation of bio-oil hydrolysate treated with solvent extraction  75  5.6 Adaptive evolution of yeast  78  Chapter 6 Conclusions  81  Chapter 7 Recommendations for future work  83  References  84  Appendix  89  vii  List of Tables Table 1 Ethanol producers in Canada [311  5  Table 2 Typical product yields (dry wood basis) for pyrolysis [10]  10  Table 3 Typical compounds in pyrolysis oil [11]  12  Table 4 Typical composition of bio-oil [11]  13  Table 5 Parameters for levoglucosan extraction with water  37  Table 6 Extraction ratios examined to determine the optimal ratios  38  Table 7 Parameters for activated carbon adsorption  44  Table 8 Summary of the volume of coagulant and flocculant added  45  Table 9 Experimental design for solvent extraction using model hydrolysate  47  Table 10 Concentration of levoglucosan, glucose and acetic acid in the artificial solutions 50 Table 11 Growth conditions tested for adaptive evolution of yeast  51  Table 12 Summary of gas chromatography methods used for ethanol and levoglucosan analysis  54  Table 13 Operating conditions for high pressure liquid chromatography analysis of carbohydrate  55  Table 14 Specifications of nanopure water used in high pressure liquid chromatography 55  Table 15 Effect of hydrolysate concentration on ethanol yield  65  Table 16 Concentration of levoglucosan, glucose and acetic acid in artificial solution A and B after hydrogenation  66  Table 17 Materials removed by treatment with coagulation and flocculation  70  viii  Table 18 List of identified compounds in the aqueous phase for pine wood derived bio-oil [46]  89  ix  List of Figures Figure 1 A timeline for gasoline price from 2001 —2008, gasoline price (light colored line), crude oil price (dark colored line) [2]  2  Figure 2 Historic U.S. fuel ethanol production (1980-2007) [3]  3  Figure 3 Summary of bio-oil applications [10]  11  Figure 4 Change in viscosity and molecular weight of bio-oil due to aging at 40°C [12114 Figure 5 Rate of viscosity change at different temperatures, Czernik et a!. [66] (diamond), Oasmaa and Sipila [65] (square), Diebold and Czernik [67] (triangle), Oasmaa et al. [11] (cross)  14  Figure 6 Structure of levoglucosan  15  Figure 7 Phase diagram of bio-oil and water system at room temperature[l5]  17  Figure 8 Glycolysis pathway [8, 57]  21  Figure 9 Tricarboxylic acid cycle  22  Figure 10 Major reaction pathways for fermentation [57]  23  Figure 11 Aerobic energy pathway for S. cerevisiae [18]  24  Figure 12 Alcoholic fermentation energy pathway for S. cerevisiae [18]  25  Figure 13 Process overview for ethanol production from bio-oil  34  Figure 14 Bio-oil titration equipment  36  Figure 15 Phase separation of bio-oil  36  Figure 16 Schematic for extraction of levoglucosan  38  Figure 17 Hydrolysate in 5OmL falcon tubes  40  Figure 18 Yeast preparation for use as inoculum  41  Figure 19 Fermentation medium in serum vials  43  x  Figure 20 Reactor from Autoclave Engineers used for hydrogenation of artificial hydrolysates  48  Figure 21 Temperature control (bottom unit) and stirrer control (top unit) for Autoclave Engineers reactor  48  Figure 22 PARR reactor used for hydrogenation of artificial hydrolysate  49  Figure 23 Typical yeast growth curve for S. cerevisiae strain T2  51  Figure 24 Optical density calibration curve for S. cerevisiae strain T2  53  Figure 25 Comparison of levoglucosan extracted between the addition of water to bio-oil (light colored bars) and bio-oil to water (dark colored bars)  58  Figure 26 Comparison of levoglucosan extraction between handheld mixer (light colored bars) and vortex mixer (dark colored bars)  59  Figure 27 Effect of water addition on levoglucosan extraction (100  —  2000 wt% water-to  bio-oil)  60  Figure 28 Effect of water addition on levoglucosan extraction (10  —  50 wt% water-to-bio  oil)  61  Figure 29 Weight percentage of acetic acid extracted from bio-oil (10-2000 wt% water to-bio-oil)  62  Figure 30 Mass transferred from bio-oil to aqueous phase (10-2000 wt% water-to-bio-oil) 63 Figure 31 Ethanol produced in various strengths of hydrolysate  64  Figure 32 First order reaction kinetics for hydrogenation of levoglucosan, glucose and acetic acid  67  xi  Figure 33 Acetic acid removals, co-solvents: 1-octanol (light colored bars) and kerosene (dark colored bars), solvents: TBP alamine 336, JMT  tributyl phosphate, ALT  primene JMT, TOA  =  =  aliquot 336, ALA  tri-n-octylamine  72  Figure 34 Percentage of glucose retained after extraction, co-solvents: 1 -octanol (light colored bars) and kerosene (dark colored bars), solvents: TBP =  aliquot 336, ALA  =  alamine 336, JMT  =  =  primene JMT, TOA  tributyl phosphate, ALl =  tri-n-octylamine  73  Figure 35 Comparison of yeast concentration after 48 hours fermentation, co-solvents: 1octanol (light colored bars) and kerosene (dark colored bars), solvents: TBP phosphate, ALT  =  aliquot 336, ALA  =  alamine 336, JMT  =  =  tributyl  primene JMT, TOA  =  tn-n  octylamine  74  Figure 36 Comparison of ethanol concentration after 48 hr fermentation (alamine 336 and tri-n-octylamine in 1-octanol)  75  Figure 37 Comparison of acetic acid removed (dark colored bars) and glucose retained (light colored bars) between 25% TOA with 1-octanol, oleyl alcohol and oleic acid  76  Figure 38 Comparison of ethanol concentration after 48 hr fermentation (tri-n-octylamine, oleyl alcohol and oleic acid)  77  Figure 39 Comparison of ethanol produced after 49 hours of fermentation (yeast adapted in aerobic and low hydrolysate concentration, yeast adapted in micro-aerophilic and low hydrolysate concentration)  79  Figure 40 Effect of hydrolysate concentration and culture conditions on ethanol yield after 49 hours fermentation  80  xii  Acknowledgements This research project was funded by Dr. Sheldon Duff of the Department of Chemical and Biological Engineering at the University of British Columbia. I would like to acknowledge and express my gratitude to Dr. Duff for his support, guidance and contribution during the course of this research project. I would like to extend my gratitude to all of my friends and all my peers in the Biofuels Research Lab and all the staff members in the department for their support. I would also like to thank my thesis committee member Dr. Sheldon Duff, Dr. Kevin Smith and Dr. Naoko Ellis for their feedback on my thesis. A special thank to Dr. Kevin Smith and Dr. Yun Quan Yang, for their help on the catalytic reactor. Finally, I would like to thank my family and Clara Leung for their love and support.  xlii  Chapter 1 Introduction 1.1 Background The purpose of this research is to produce ethanol in an economically and environmentally sustainable way, to relieve the dependency on petroleum fuel in North America. As an alternative source of energy, ethanol can be used as a fuel additive in gasoline and as fuel in vehicles [1]. The addition of ethanol to gasoline can reduce emissions of greenhouse gases, and reduce the consumption of gasoline. These benefits have led to an increasing demand for ethanol [2]. According to the Renewable Fuels Association (RFA) of Canada, the production of ethanol in the past 5 years has nearly tripled, and it is predicted that the production will continue to grow [3]. It is widely known that ethanol fuel can be produced by energy crops such as corn, wheat and sugar cane. While it is a cleaner alternative than petroleum, it poses many problems, such as: the negative net energy balance for the production of ethanol, the uptake of agricultural lands and the increasing demand and price of the energy crops [1]. In this project, the production of ethanol from bio-oil is being investigated. Bio-oil is a product of processing biomass in a process called fast pyrolysis. The biomass used to produce bio-oil can be any biomass wastes such as wood wastes or agricultural wastes [4]. Wood wastes like sawdust from pulp mills or beetle-killed lodgepole pine can be used to produce bio-oil in British Columbia [3]. This option has an additional benefit of reducing the forest fire hazard [3]. The feasibility of producing ethanol from bio-oil has been proven by previous researchers here at UBC and other places in the world who worked on similar projects [5, 6, 7]. The production of ethanol from bio-oil involves three stages: extraction, hydrolysis and fermentation. In the first stage, the anhydrosugars, mainly levoglucosan, are extracted using water. This process takes advantage of the hydrophilic nature of levoglucosan. In addition, water is a clean and inexpensive solvent, which reduces the risk associated with the use of organic solvents. In the second stage, the extraction product is hydrolyzed into  fermentable sugars using dilute sulfuric acid. At this stage, the yield of the hydrolysate from levoglucosan will exceed 100% due to the hydrolysis of other anhydrosugars in the bio-oil extract [6]. This will improve the fermentation by providing more free sugars to ferment into ethanol, thus increasing the production [6]. In the final stage of the process, the hydrolysate is fermented into ethanol using yeast, specifically Saccharomyces cervisiae strain T2. This yeast is used in the fermentation process due to its effectiveness and resistance to inhibitors.  1.2 Ethanol Ethanol is a volatile, flammable, and a colorless liquid at room temperature. It has a strong alcohol smell and burns with a smokeless blue flame. Ethanol is also a versatile solvent, miscible with water and with many organic solvents. The discovery of ethanol can be dated back to the  th 15  century [1]. Ethanol has a long history as a source of energy.  In the early 1800s, ethanol was the major illuminating fuel for lamps in the United States [1]. In the modern world, ethanol is used commonly as motor fuel and fuel additive.  and Rita  Hurricanes  Tight gasoline supplies in the  140 120 100  60  c-) 40 20 0 —  c)  -.  ID.  0Ic)  N -  ID.  <  N  N  C”  0 0 0J)(.)  0  0  0  —  ID.  ID.  <<<<  0  <  0  0  0 ID.  0 0 0i)U a.  <<<<  0 -  ID.  IN 0  IN 0 a  00 0 ID  <<<  Figure 1 A timeline for gasoline price from 2000 2008, gasoline price (light colored line), crude oil price (dark colored line) 12] —  2  Gasoline has dominated the fuel market for over a century. Since 2001, the retail price of gasoline has increased dramatically due to the fear of tight gasoline supply in the United States [2]. From January 2001 to May 2008, the retail price of gasoline has tripled [2]. There are a number of explanations for the increase in the price of gasoline, but the major reason is the dependency on gasoline as transportational fuel. Thus, any events that could potentially upset the supply of gasoline can drive the gasoline price up. Figure 1 has recorded the timeline of gasoline price and events in the Unites States that led to the increase in the retail price of gasoline. The jump in the gasoline price, and an increase in the environmental awareness, have sparked opportunities for alternative energy sources. Some of the vehicles that use alternative energy sources are hydrogen vehicle, biodiesel vehicle, fuel cell vehicle, and electric/gasoline hybrid vehicle. While ethanol can also be an alternative energy source for vehicles, its major use is as a fuel additive. In the past decade, ethanol has gained importance as a fuel additive in the United States. A graph of ethanol produced in the United States from 1990 to 2008 is shown in Figure 2.  10000 8000  j  6000 4000 2000 0 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 Year Figure 2 Historic U.S. fuel ethanol production (1990-2008) [3]  3  As an oxygenated fuel additive, ethanol can reduce the emission of pollutants, including particulates, and reduce the consumption of gasoline. The reduced harmful pollutants include carbon monoxide, carbon dioxide, formaldehyde and other hydrocarbons [8]. In addition, the higher octane number of ethanol  Q— 110) increases the octane number of the  mixture, thus reducing the need for toxic, octane-enhancing additives [8]. These benefits have led to the increasing demand for ethanol. The Energy Information Association (EIA) of the Unites States estimated that the gasoline market in the United State will grow from 142 billion gallons per year in 2007 to 149 billion gallons of gasoline by 2015 [1]. With this in mind, a ten percent additive in the gasoline would require 14.2 billion gallons of ethanol per year in 2007 to 14.9 billion gallons per year by 2015 [1]. The estimate was based on ten percent because most cars can be run on a mixture of 90% gasoline and 10% ethanol without adjusting their engines [9]. Newer model cars can even use mixtures of gasoline and ethanol that contain up to 20% ethanol. Furthermore, there are new engines that are designed to run on pure ethanol [8]. The new flexible fuel vehicles take advantage of the capabilities of this technology and are able to use mixtures of 0  —  85%  ethanol in gasoline. This is commonly known as the E85 vehicles [9]. Ethanol can also be mixed with diesel fuel in compression ignition engines with the help of an emulsifier [9]. Canadian production of ethanol has increased dramatically in recent years as well. In 2003, ethanol production was only 212 million litres per year. By the end of 2007, production had risen to approximately a billion litres per year. Currently there are four plants under construction and are projected to be commissioned in 2009 [10]. Table 1 shows a list of all the major bioethanol producers in Canada and their production capacity [10]. When the plants are in operation, the production of ethanol in Canada is expected to reach over a billion and a half litres of ethanol (Table 1) [8, 10]. With the growing awareness of environmental issues, ethanol is attracting a lot of interest in the international and domestic market due to the environmental benefits associated with its use [11]. In addition, Canada has introduced a five percent ethanol mandate by 2010, which requires 5% ethanol to be blended in gasoline. This mandate will provide further incentive to the production of ethanol in Canada.  4  Table 1 Ethanol producers in Canada 1101 Ethanol producers  GreenField Ethanol Pound Maker Agventures_Ltd. Permolex International,_L.P GreenField Ethanol logen Corporation  City  Province  Feedstock  Capacity (million litre per year)  Tiverton  ON  Corn  26  1989  Laniga  SK  Wheat  12  1991  Red Deer  AB  Wheat  40  1996  Chatham  ON  Corn  150  1996  Ottawa  ON  Straw from wheat, barley and oats  2  2004  Start Date  NorAmera BioEnergy_Corp. Enerkem Inc.  Weybury  SK  Wheat  25  2005  Westbury  QC  Wood waste  5  2005  Suncor Energy  Sarnia  ON  Corn  200  2006  Husky Energy Collingwood Ethanol_LP Husky Energy GreenField Ethanol GreenField Ethanol  Lloydminster  SK  Wheat  130  2006  Collingwood  ON  Corn  50  2007  Minnedosa  MB  Wheat and corn  130  2007  Varennes  QC  Corn  120  2007  Johnstown  ON  Corn  200  2008  IGPC Ethanol Inc.  Aylmer  ON  Corn  150  2008  Belle Plaine  SK  Wheat  150  2008  Edmonton  AB  Municipal landfill waste  36  Under construction  Unity  SK  Wheat  25  Hensall  ON  Corn  200  Havelock  ON  Corn  80  Terra Grain Fuels Inc. GreenField Ethanol Enerkem Inc. North West BioEnergy GreenField Ethanol -  Kawartha Ethanol Total  1731  5  Under construction Under construction Under construction  1.2.1 Overview of ethanol production  A number of different methods can be used to produce ethanol. The most common ones are hydration of ethylene and fermentation of sugars. Ethanol that is used as industrial feedstock is most often made from petrochemical feed stocks, typically by the acidcatalyzed hydration of ethylene:  4 H 2 C  +  0 2 H  —*  CH 3 CH O 2  The most commonly-used catalyst is phosphoric acid, adsorbed onto a porous support such as diatomaceous earth or charcoal [121. The reaction is carried out with high pressure steam at 300°C. In an earlier process, ethylene was hydrated indirectly by reacting with concentrated sulfuric acid to produce ethyl sulfate, which was then hydrolyzed to yield ethanol and sulfuric acid [13, 14]:  4 H 2 C  +  CO 3 CH S 2 H 4 H  S0 2 H 4  —  CO 3 CH S 2 H 4 H  0 2 H  —÷  CH 3 CH O 2  +  +  S0 2 H 4  Ethanol produced for alcoholic beverages and fuel, is mostly produced by fermentation. Yeast such as Saccharomyces cerevisiae can metabolize sugar in an oxygen limiting environment to produce ethanol and carbon dioxide [15]. The chemical equation below summarizes ethanol fermentation, in which one hexose molecule is converted into two ethanol molecules and two carbon dioxide molecules:  12 H C 6 0  —  2C OH +2 CO 5 H 2 2  Yeast will perform the reactions only if oxygen is excluded or limited from the environment. Otherwise, yeast will oxidize pyruvate completely to carbon dioxide and water [15].  6  1.2.2 Substrates for bioethanol  In the past several decades, a number of feedstocks have been studied as potential substrates for the fermentation of ethanol [5, 16, 17]. In most bioethanol plants, the feedstock for production is an energy crop [3, 11]. The most common energy crops in North America are corn and wheat [3]. The use of energy crops to produce ethanol has received a lot of attention. This is because there are debates about whether the net energy return from these energy crops is positive. In light of this, the U.S. Department of Energy (DOE) has studied the energy return from the energy crops and reported a negative energy return regarding ethanol production using corn [17]. Several other studies have also suggested that ethanol production from corn does not provide a positive energy return and its production process contributes to global warming [18, 19]. Furthermore, the increased demand for corn to produce ethanol drives corn prices up, which can translate into higher meat, milk and egg prices for consumers [18]. Despite the studies, producing ethanol from corn is a mature technology and, with the help of government subsidies, companies are currently employing this method to produce ethanol. However, the drawbacks of the energy crops also inspired the study of alternative feedstocks  for bioethanol  production.  Potentially  more  attractive  options  are  lignocellulosic waste materials that contain sugars [4]. Ethanol produced from biomass wastes is often referred as second generation bio-fuel, with the first generation of bio fuels being that produced from energy crops.  1.2.3 Pyrolysis and other substrate pretreatments methods  In order to produce ethanol from lignocellulosic materials, single or multistep processes must be carried out to produce fermentable sugars. These processes are used for the removal of lignin, which allows easier access to the cellulose and hemicellulose in enzymatic or acid hydrolysis process. This helps to increase the sugar yield in the subsequent hydrolysis step. A number of pretreatment methods have been proposed for  7  the removal of lignin. They can be categorized into chemical, thermochemical and biological pretreatments or a combination of these pretreatment processes. Prior to the pretreatment methods, lignocellulosic materials are often pulverized by a combination of chipping, grinding, and milling [20]. This is to reduce the physical size and to reduce cellulose crystallinity of the lignocellulosic materials [20].  The chemical pretreatments include ozonolysis and the organosolv process. Ozonolysis is the addition of ozone to degrade lignin while keeping the degradation of cellulose to a minimum [20]. This process has the advantage that it does not produce additional toxic chemicals, but the process requires the use of ozone which makes the process more expensive [20]. In the organosolv process, organic solvents such as methanol, ethanol, acetone, ethylene glycol, triethylene glycol and tertrahydrofurfurl alcohol are used to break the internal bonds of lignin and hemicellulose [21, 22]. This process can be catalyzed by the addition of organic acids such as oxalic or salicylic acid [20]. The recovery and removal of the solvents is essential to reduce the operating cost and minimize the inhibition to the organisms in the enzymatic hydrolysis and fermentation process [20].  Steam explosion is one of the most common thermochemical pretreatments of lignocellulosic materials [20]. In steam explosion, the materials are treated with high pressure saturated steam and then the pressure is swiftly reduced to cause an explosive decompression [20, 23, 24]. This process has been proven to be a good pretreatment for enzymatic hydrolysis of hardwoods. Ninety percent efficiency of enzymatic hydrolysis had been achieved in 24 hours for poplar chips that were treated by steam explosion, compared with only 15% efficiency with untreated poplar chips [25, 26].  Oxygen delignification and ammonia fibre explosion (AFEX) are examples of a combination of chemical and thermochemical processes. The AFEX process is similar to steam explosion. The lignocellulosic materials are exposed to ammonia at high temperature and the pressure is quickly reduced to cause an explosive decompression [20]. Oxygen delignification is a process that is widely used in the pulping industry. In  8  oxygen delignification, the materials are treated with high pressure oxygen at 100  —  150°C under alkaline conditions [27]. This triggers a series of reactions that lead to the degradation of lignin [27]. However, this process is reported to be ineffective for biomass with high lignin content (>18%) [20].  Biological pretreatments are not as commonly used as the thermochemical and chemical methods due to the long residence time and low efficiency [28]. However, it has some advantages over the other pretreatment methods, such as the low energy requirement for the process and the mild reaction conditions. These advantages attract interest from researchers trying to improve the process and overcome its weakness. The white rot fungi are used for the biological pretreatment because they are capable of selectively degrading lignin. The most common fungi used to remove lignin from plant cell walls are white rot fungi [28]. A study by Fackler et al. has reported a 1.7 3.2% lignin removal in 10 weeks -  on spruce wood veneer with various white rot fungi strains [28].  Among these pretreatment methods, pyrolysis is selected because the technology is mature and it can be applied to most lignocellulosic waste without any adjustment to the process [29]. In this project, the lignocellulosic material (pine) was pyrolyzed using a process called fast pyrolysis to produce a liquid called bio-oil. In fast pyrolysis, the biomass wastes are treated at high temperature (>300°C), resulting in a rapid decomposition of biomass into liquid, char and gas [4]. The pyrolysis liquid will contain most of the sugars (in a dehydrated or “anhydro” form) from the raw material and can be used for ethanol fermentation. The actual sugar content will depend on the raw material used in the pyrolysis process, as well as the processing conditions. These anhydrosugars can easily be hydrolyzed to produce fermentable sugars.  9  1.3 Characteristics of bio—oil Bio-oil is one of the products of processing biomass using the process called pyrolysis. Hence, bio-oil is also called pyrolysis oil. Pyrolysis is a process of thermal decomposition of biomass occurring in the absence of oxygen. It is also the first step in combustion and gasification processes where it is followed by total or partial oxidation of the primary products [4, 30]. The process of pyrolysis can vary, but generally it can be categorized into three groups fast pyrolysis, carbonization and gasification. The main variables for —  the pyrolysis processes are reaction temperature and residence time. Table 2 summarizes the different conditions for the pyrolysis processes and their resultant product distribution in liquid, char and gas. Fast pyrolysis for liquid bio-oil production is of particular interest because it can produce up to 75% of the original mass of wood as liquid bio-oil [4]. The remaining fractions are 12% char and 13% gas [4]. The char and gas can be recycled in the fast pyrolysis process to produce heat, thus making the process more energy efficient. Table 2 Typical product yields (dry wood basis) for pyrolysis [4] Process  Conditions Moderate temperature, short residence Fast pyrolysis time Low temperature, very long residence Carbonisation time Gasification High temperature, long residence time  Liquid  Char  Gas  75%  12%  13%  30% 5%  35% 10%  35% 85%  There are many potential applications for bio-oil, such as transportation fuel, and fuel for boilers, furnaces, and turbines [4]. In this project, one fraction of bio-oil is used to produce ethanol through a series of process such as extraction, hydrolysis, detoxification and fermentation. Several researchers have reported their work on similar project. Yu and Zhang [7] have reported that they were able to hydrolyze cotton pyrolysate, and after subsequent detoxification, obtained 14.2 g/L of ethanol from 31.6 g/L of glucose after 24 hours of fermentation [7]. Another study by Bennett [6] has reported that by diluting the bio-oil hydrolysate (2% v/v), an ethanol yield of 51% (g ethanol/g glucose) was found [6]. Figure 3 summarizes the different applications for bio-oil.  10  Recycle process heat  Extraction  Chemicals  Conversion  Transportation pgrading fuels  Electricth  Recycle process heat  Figure 3 Summary of bio-oil applications [101  1.3.1 Compositions of bio-oil  Bio-oil is usually dark brown with a distinctive smoke-like smell.  The elemental  composition of bio-oil is similar to that of the biomass before pyrolysis [30]. Bio-oil contains a complex mixture of acetic acid, formic acid, methanol, formaldehyde, aldehydes, ketones, phenols, hydroxymethyl, furans, furanones, pyranones, anhydrosugars, oligosaccharides, lignin and water (Table 19) [30, 31]. Table 3 shows the different groups of compounds that are found in bio-oil [30]. Some suspended solids may also be present in the bio-oil.  11  Table 3 Typical compounds in pyrolysis oil  [30]  Composition range (wt% of bio-oil)  Hydrophilicity (arbitrary scale, 4 is highest)  Cl compounds (formic acid, methanol and formaldehyde, C0 ) 2  5-10  4  C2 C4 compounds ( linear hydroxyl- and oxo-substituted aldehydes and ketones)  15-35  4  C5 C6 compounds (hydroxyl-, hydroxymethyl- and/or oxo substituted furans, furanones and pyranones)  10-20  3  C6 compounds (anhydrosugars, anhydro oligosaccharides)  5-30  4  Water-soluble carbohydrate derived oligomeric and polymeric materials  5-10  4  Monomeric methoxyl substituted phenols  6-15  2  Pyrolytic lignin  15-30  Compounds  -  -  I 32 Phase behavior and stability of bio-oil  Typically bio-oil contains around 15  —  30 wt% of water depending on how it was  produced and collected [4, 30, 32]. The water molecules are chemically dissolved in bio oil. In this case, the bio-oil is in single phase due to the presence of polar carboxyl and hydroxyl compounds [30]. However, phase separation can occur with an increase in the concentration of water or lignin-derived material to a critical level [4, 7, 33]. Thus, phase separation can be induced when a critical concentration of water is added, usually 10—25 wt% [33]. The typical composition of bio-oil is provided in Table 4.  12  Table 4 Typical composition of bio-oil 1301 Component Weight percentage (wt%) Water 21.1  Solids  0.008  Ash  0.030  Nitrogen  0.060  Carbon  43.5  Hydrogen  7.09  Oxygen  28.2  It has been reported that bio-oil is not a product of thermodynamic equilibrium [4, 30]. Instead, it is produced with short residence time in the reactor and rapid quenching from the pyrolysis temperature. This produces liquid bio-oil that is not at thermodynamic equilibrium under normal storage conditions. Thus, the chemical composition of the bio oil will tend to change towards thermodynamic equilibrium during storage. This results in changes in viscosity, molecular weight, chemistry and co-solubility of its many compounds [31].  The aging of the bio-oil can be monitored by its viscosity [311. The change in viscosity is caused by the loss in volatile components in bio-oil and/or the changing of the chemical composition of bio-oil as it moves toward thermodynamic equilibrium [31]. The relationship between the change in viscosity and molecular weight of bio-oil due to aging is shown on Figure 4 [31]. According to Figure 5, the viscosity of the bio-oil used for this project had increased by 182.5 cp over the past 5 years at 4°C [30].  13  1000 900 800 700 600  MW=  500  278.3  400 300 200 100 0 100  150  200  250  300  350  400  Viscosity Figure 4 Change in viscosity and molecular weight of bio-oil due to aging at 40°C [311 It is reported that the rate of viscosity change is slower at lower temperatures. This has been confirmed by a number of researchers {30, 34, 35, 36]. A graph summarizing different results from four different researchers is shown in Figure 5.  -20 C.? C,)  0.01 -.  Temperature (°C) Figure 5 Rate of viscosity change at different temperatures, Czernik et al. [35] (diamond), Oasmaa and Sipila [34] (square), Diebold and Czernik [36] (triangle), Oasmaa et al. [301 (cross) 14  1.3.3 Anhydrosugars in bio-oil As can be seen in Table 3, bio-oil is comprised of a significant fraction of anhydrosugars. Levoglucosan (1, 6—anhydro—3—D—glucopyranose, Figure 6) is a major component in bio-oil and the compound of interest in the project. There are a number of studies on the levoglucosan yield from pyrolysis oil. Typically, wood derived bio-oil contains about 5  —  30% of levoglucosan [4, 30]. In another study, Li and Zhang [5] has reported levoglucosan yield of 18.66% from the pyrolysis of newspaper and 33.06% from the pyrolysis of waste cotton [5]. Generally, the amount of levoglucosan depends on the types of feedstock, the pretreatment of feedstock and the pyrolysis process conditions [30]. As an anhydrosugar, levoglucosan can be hydrolyzed into glucose by acid hydrolysis, which can then be fermented into ethanol [5, 7, 37]. Thus, bio-oil provides a potentially rapid and efficient route to the supply of substrate for bioethanol production [5, 7]. Other uses for levoglucosan include the synthesis of oligosaccharides and  pharmaceuticals materials [5].  Figure 6 Structure of Jevoglucosan Cellobiosan, another anhydrosugar produced during wood pyrolysis can also be hydrolyzed to glucose by acid hydrolysis [38]. However, unlike levoglucosan hydrolysis, cellobiosan is first hydrolyzed into levoglucosan and cellobiose. Then, levoglucosan and cellobiose are further hydrolyzed to glucose [38]. The exact concentration of cellobiosan in wood derived bio-oil is uncertain, but it has been reported by Helle et al. [38] that cellobiosan yield can be as high as 30% of levoglucosan yield (1 .5 yield) at a pyrolysis temperature of 500°C [38]. 15  —  9% cellobiosan  1.34 Extraction of levoglucosan from bio-oil In order to utilize the levoglucosan for fermentation, solvents are required to extract it out of the bio-oil via liquid—liquid extraction. Water is selected as the solvent for the extraction because levoglucosan is hydrophilic. There are other advantages in using water as a solvent, such as availability and cost. However, other compounds with similar hydrophilicity are also extracted along with levoglucosan from the bio-oil. These compounds are listed in Table 3 [30]. A phase diagram for bio-oil and water system is given in Figure 7. The phase diagram correlates the weight percentage of water and organic compounds. In the diagram, the linear segment represents the bio-oil as a single phase liquid. Once the water content has exceeded 25 wt%, phase separation can occur for this particular bio-oil [33]. However, this phase diagram exhibits only the general trend. Other factors, such as the age and the raw materials from which the bio-oil was made have a significant effect on the water content and the organic content in the bio-oil [4]. This ultimately affects the critical amount of water required to induce phase separation.  After phase separation, the top phase is often referred as the aqueous phase because it is a homogenous solution that behaves like water. The aqueous phase contains most of the levoglucosan and other carbohydrate-derived compounds [30]. The bottom phase is often called the organic phase. In the organic phase, most of the lignin-derived compounds in the bio-oil are retained [30]. The organic phase is darker in color, behaves like a single phase and flows like a viscous fluid.  16  ‘Water  iii  system (wt%)  ioo 30  80  70  60  50  40  30  20  1  10  30  10  80  20  70  30  60  40  50  50  40  60  30  70  213  80  10  90  —  0 0  0  —100  ....—  0  10  30  20  40  50  60  70  80  90  Organics in system (wt%) Figure 7 Phase diagram of bio-oil and water system at room temperature  133]  The extracted levoglucosan can then be hydrolyzed into fermentable sugar. The sugar solution after hydrolysis, called hydrolysate, can be used subsequently for fermentation with Saccharoniyces cervisiae to produce ethanol. Since typical concentrations of levoglucosan range from 5  —  30% in the bio-oil, producing ethanol from bio-oil is not  economically feasible [4. 30]. In order to increase the economic feasibility, producing ethanol with bio-oil in an integrated pyrolysis combined cycle (IPCC) power plant has been proposed and studied by Sandvig et al. [39]. The concept of IPCC power plant is that the bio-oil serves both as a fuel for a gas turbine cycle and as a feedstock for valueadded chemicals production [39]. This increases the efficiency and economic feasibility of using pyrolysis oil due to the co-generation of products and recycling of process and pyrolysis heat [6, 39].  17  1.3.5 Hydrolysis of levoglucosan and other anhydrosugars Hydrolysis is a chemical process in which a molecule is cleaved into two parts by the addition of a molecule of water. One of the molecules gains a hydrogen ion (H) from the water molecule. The other part of the molecules collects the remaining hydroxyl group (0H). In bioethanol production, the hydrolysis step is performed to break down complex sugars, hemicellulose and cellulose or in this case levoglucosan, to monomer sugars like glucose. This hydrolysis step can be performed with different methods such as using concentrated acid, dilute acid or enzymes. Each method has its advantages and disadvantages. For instance, concentrated acid hydrolysis will introduce more inhibitory compounds in the system than the other dilute acid and enzymatic hydrolysis, while enzymatic hydrolysis is more costly than acid hydrolysis [6, 40, 41, 42, 43]. In this project, sulfuric acid was used for hydrolysis because it had demonstrated to be effective and inexpensive [6]. Several researchers had reported that a glucose yield of greater than 100% was observed using dilute acid hydrolysis to hydrolyze bio-oil extract [6, 7, 38]. Helle et al. [38] had reported that the extra glucose came from the hydrolysis of cellobiosan in bio-oil [38]. The intermediate products of cellobiosan hydrolysis were cellobiose and levoglucosan [38]. These two anhydrosugars were then further reacted into glucose [38]. Thus, dilute acid hydrolysis of bio-oil extract would often result in greater than 100% theoretical glucose yield, when the calculation was based on initial concentration of levoglucosan [381.  1.4 Yeasts Yeasts are unicellular microorganisms classified in the kingdom fungi, which are common in the natural environment. They can be found in terrestrial, aerial and aquatic environments, where the success of their colonization is closely related to their adaptability to highly variable environments [44]. Yeasts can be classified with respect to the types of energy-generating process involved in sugar metabolism, namely facultative  18  fermentative, non-facultative-fermentative and obi igate-fermentative yeasts [44]. The non-facultative fermentative yeasts are only capable of the respiratory metabolism and are not capable of alcoholic fermentation from glucose. An example of this type of yeast is Rhodotorula giutinis [44]. On the other hand, the obligate fermentative yeasts are only capable of metabolizing glucose through alcoholic fermentation. An example of this yeast is Kazachstania sloofjIae  [441.  Most of the identified yeasts are facultative  fermentative. These yeasts may display either a fully respiratory or a fermentative metabolism or even both in a mixed respiratory-fermentative metabolism [44]. The type of metabolism the facultative fermentative yeasts experience depends on the growth conditions, such as the types of sugars present, the concentration of sugars and the availability of oxygen [44]. S. cerevisiae, the yeast used in the project, is an example of facultative fermentative yeast.  The most common microorganisms that are suitable for our application are Z mobilis and S. cerevisiae [45]. However, many other microorganisms are also capable of fermenting sugars into ethanol [45], such as engineered Escherichia coli [46], Monilia sp. [47], Paeciiomyces sp. [48], Closiridium ihermoceilurn [49] and Neurospora crassa [47]. The latter four microbes are used in single stage cellulose-to-ethanol biotransformation. In this biotransformation, these microbes ferment lignocellulosic waste directly to ethanol without any pretreatment or preparation stages. However, studies on the fermentation process that utilizes these microorganisms have shown that the processes are slow and the ethanol yields are low [45]. These cellulose-to-bioethanol biotransformation processes have been found to take up to 12 days to complete [45]. Another disadvantage is that these microorganisms have lower resistance to high concentrations of ethyl alcohol than yeast. In the case of bacterial fermentation, the process produces by-products such as acetic and lactic acids, which can inhibit the fermentation reactions [45]. There is not a lot of research on the use of engineered E. coil due to drawbacks such as the narrow and neutral pH growth range [45]. Furthermore, E. coil are less robust when compared with yeast cultures [45].  19  1.4.1 Glucose metabolism The two glucose metabolism pathways of interest are aerobic respiration and anaerobic metabolism. Glucose metabolism under both aerobic and anaerobic conditions begins with the process called glycolysis, which breaks down a glucose molecule into two pyruvate molecules. Under aerobic conditions, glycolysis is followed by tricarboxylic acid (TCA) cycle and oxidative phosphorylation for ATP generation and biosynthesis [15]. Under anaerobic conditions, the resulting pyruvate molecules are then converted into ethanol through a series of reactions [50].  1.4.2 Glycolysis mechanism Glycolysis is the first step in both aerobic and anaerobic respiration and is the process by which glucose is broken down into pyruvate. The reaction sequence begins with a glucose molecule being phosphorylated by reaction with ATP and the enzyme hexokinase to yield glucose 6-phosphate (G-6P) [15, 50]. Then, G-6P is isomerized to fructose 6-phosphate with the help of phosphoglucose isomerase and subsequently converted to fructose-i, 6-bisphosphate by the phosphofrutokinase-catalyzed reaction [15, 50]. This gives us an intermediate molecule that is cleaved by aldolase [15, 50] to produce glyceraldehyde-3 -phosphate (GA-3 P) and dihydroxyacetone phosphate (DHAP) [15, 50]. GA-3P and DHAP are in equilibrium with each other but as GA-3P is oxidized and phosphorylated to yield 3-phospoglyceryl phosphate in the subsequent step, DHAP is converted to GA-3P due to a shift in the equilibrium [15, 50].  A phosphate in the  carboxyl group of 3-phosphoglycerate (3 P-GA) is transferred to ADP generating an ATP molecule and reducing 3-phosphoglyceryl phosphate to 3P-GA [15, 50]. Then, 3P-GA is converted to 2-phosphoglycerate by phosphoglyceromutase followed by a dehydration step which yields phosphoenolpyruvate (PEP) [15, 50]. In the final step of glycolysis, a phosphate group is transferred to ADP to generate ATP and yielding the final product, pyruvate [15, 50]. Pyruvate is a key metabolite; it can be converted to ethanol under  20  anaerobic conditions. Under aerobic conditions, pyruvate can be converted to CO 2 and NADH through the TCA cycle. A summary of the metabolic pathways of glycolysis is constructed in Figure 8.  3 O 2 CH PO  HOCH  CH:OPO*  Fructose 6-phosphate  H  OOJ  Li < ATP ADP Hexokinase  Glucose  OH  Glucose phosphate isomerase  OH  Glucose 6-phosphate  Phosphofrucw icirase  Fructose I ;6bSphSphtC 3 -Phosphoelveervi phosphate  3 -Phospholvcerate 0  4/  O  C OH 2 3 CHOPO  Aldolase  1 OPO ADP iTP’ c\ -r Ph1yen inas  L  H—c—OH OPO 1 CH 5  4 -JJHi-i y-j  GA-SF ddrese  H(O  3-phosphate 04/0  0  04/0  C  c  20 p 0 H_C_ 3 CH O 2 I-I  2-Phosphogivcirate  Eno1a  C0  i_opo  CH  2 CH  Phosphoenolpvruvate Figure 8 Glycolysis pathway 115, 501  21  I  ‘OPOCH:CHCH 4+ O3POCHCCHOH  Glvceraldehvde  ipbc1ycwmutase  ‘V  Dihvdroxyacetone phosphate  1.4.3 Utilization of pyruvate under aerobic conditions and the tricarboxylic acid cycle  The major roles of the tricarboxylic cycle are to supply carbon skeletons for amino acid synthesis, and to provide electrons to the electron transport chain, thereby generating energy for biosynthesis [15]. In order for pyruvate to go through the TCA cycle, it has to undergo a series of transformations to acetyl CoA. The TCA cycle is followed by oxidative phosphorylation where NAD molecules are regenerated for glycolysis and ATPs are produced for biosynthesis. A diagram of major reactions and functions of the product is shown in Figure 9.  acetyl CoA  citrate -synthesis of fatty acids  oxaloacetate -synthesis of asparate. purmes  a —ketoglutarate -  synthesis of glutarnine  succinyl CoA -synthesis of herne. chlorophyll  Figure 9 Tricarboxylic acid cycle  1.4.4 Utilization of pyruvate under anaerobic conditions Fermentation is a process where yeasts derive energy from the oxidization of carbohydrates using an endogenous electron acceptor, which is an organic electron acceptor synthesized by the cell itself [51]. There are different species of yeasts that can  22  metabolize sugars and convert them into ethanol. In this project, sugar is being converted to ethanol by the yeast Saccharomyces cerevisiae. Under anaerobic conditions, glucose metabolism begins with glycolysis, which breaks down a glucose molecule into two pyruvate molecules. Then the pyruvate molecules are converted to acetaldehyde by decarboxylation [50]. Acetaldehyde molecules are further reduced to ethanol by the enzyme called alcohol dehydrogenase [50]. The reactions require NADH as cofactor and regenerate NAD to be used in glycolysis [50]. In this cycle, the organism is growing without using the electron transport chain to generate energy, which is a characteristic of fermentation [15]. A flow diagram of the major reaction pathways for fermentation is shown in Figure 10.  glucose  ADP  NAD  AlP  NADH  0  4-, C CE0  pyruvate  NADH  pyruvate  decarboxylase HC = 0  2 Co  3 CH  \  NAIY 7[  alcohol  dehydrogenase acetaldehyde  C 2 H 3 CH  ethanol  Figure 10 Major reaction pathways for fermentation [50J  23  —  OH  1.4.5 Comparison of aerobic and anaerobic conditions Yeasts were grown under two types of growth conditions during this project: aerobic and micro-aerophilic conditions. Since the yeasts used in the project are facultative fermentative, they are able to perform a mix of both respiratory and fermentative metabolisms. Under aerobic respiration, yeasts can generate more ATP for biosynthesis. However, the production of ethanol is not favorable under aerobic conditions [6]. Thus, the yeasts are grown under aerobic conditions for culturing (producing inoculum) while anaerobic conditions are used for ethanol production.  02  Glucose  Glucose  NAD-f  NADH \  NADH,/ NAD+  2 CO  -  -  ATP  ATP  —  I  —-  /  ADP  ADP \\  /  ,/  Biomass  0 2 H  N / 2 CO  +  0 2 H  Figure 11 Aerobic energy pathway for S. cerevisiae 1511  Under aerobic conditions or respiratory growth, NADH produced by glycolysis can be re oxidized to yield an additional ATP molecule [44, 51]. This is called oxidative phosphorylation, a metabolic pathway that uses energy released by the oxidation of glucose to produce AlP [51]. A flow diagram of this energy pathway is shown in Figure 11. In aerobic conditions, the reactions yield a total of sixteen ATP molecules from one glucose molecule [44].  24  In micro-aerophilic fermentation, substrate level phosphorylation in glycolysis is the major source of ATP in S. cerevisiae, because oxidative phosphorylation is limited with limited oxygen [51]. This reaction gives a net yield of two ATP molecules for each molecule of glucose [44]. A flow diagram of this energy pathway is shown in Figure 12.  Glucose  Glucose  Glucose  NAD+  N Co  /  NADH / —7 ----‘  All’  ATP. ‘N —  /  /  ADP  Biomass  Glycerol  ADP  Ethanol  Figure 12 Alcoholic fermentation energy pathway for S. cerevisiae [51]  By comparing the aerobic conditions with the micro-aerophilic conditions, yeasts growing under aerobic conditions produce eight times more ATP. Thus, a much higher yeast growth rate is observed. The typical yield of biomass with aerobic fermentation is 0.5 g/g glucose, while the typical yield for micro-aerophilic fermentation is 0.1 g/g glucose [51].  25  1.4.6 Parameters affecting fermentation The physical factors that can affect the production of ethanol are temperature, mixing speed and residence time. Temperature can determine the ability of the yeasts to grow and affect the growth rate of the yeasts. The optimal temperature range of the yeast Saccharomyces cerevisiae is reported to be 28 40°C [521. Within this temperature range, the growth rate of yeasts is a linear function of temperature and is modeled using the —  Arrhenius equation [53]. At temperatures below the optimal growth range, yeast growth rate decreases steadily, while zero growth occurs at temperature above 40°C and below 5°C [52, 53]. The rapid decrease in growth rate is due to the decrease in the resistance to stress at temperatures outside of the optimal growth range [53]. Yeasts like Saccharomyces cerevisiae can tolerate a wide range of pH. They are known to be able to grow at a pH range from 2.4 8.6 [52]. However, the optimal pH range is reported to be 4.5 5.5 [53]. The actual optimal pH range for the yeasts is dependent on —  —  the type of acid dissociating in the medium. For instance, acetic acid was found to be more inhibitory than lactic and citric acids [53]. In addition, the optimal pH range is also dependent on the temperature and the presence of other inhibitory compounds [53].  1.4.7 Inhibition of fermentation When bio-oil is extracted with water, levoglucosan as well as a number of other water soluble chemical compounds are extracted to the aqueous phase. Some of the compounds extracted are inhibitors in the fermentation process. Some of the inhibitors identified in bio-oil are dissolved solids, acetic and formic acids [52]. Acetic acid inhibition is mainly caused by the undissociated form of acetic acid [52]. Undissociated weak acids are liposoluble and can diffuse across the plasma membrane by passive diffusion [20, 52]. Upon entering the cell’s cytosol, dissociation of the acids occurs due to the neutral intracellular pH, thus decreasing the internal pH [20]. At a concentration of 3 gIL acetic acid, the internal pH of S. cerevisiae decreases from 7.0 to 5.6, which result in a 20%  26  decrease in the fermentation rate [52]. A study by Larsson et al. [54] has reported that the ethanol yield is reduced from 0.45 (g ethanol/g glucose) to 0.35 (g ethanol/g glucose) at an acetic acid concentration of 29.7 gIL [54]. In addition, when other acids such as formic or levulinic acid are present, the ethanol yield is reduced to 0.33 (g ethanol/g glucose) and 0.29 (g ethanol/g glucose), respectively [54]. Other factors that can influence the toxicity of acetic acid include membrane permeability, pH gradient across the cell membrane, and buffering of the cytoplasm [52]. As dissociated acetic acids build up inside the cell, acetic acid will begin to buffer the cytoplasm thereby partially protecting the cell from additional acetic acid [52]. The increase in buffer capacity is mainly responsible for the acetic acid adaptation. In addition to the internal pH, short-chain fatty acids can also affect the membrane transport of phosphate, resulting in an increase in the maintenance energy for the cells [52]. Also, formic acid is more toxic than acetic acid, due to a lower pKa than acetic acid [52, 54]. The combined effect of these toxic compounds can result in a more severe fermentation inhibition than a single toxic compound [40]. This was observed in Nigam’s [55] experiment where wheat straw-derived hydrolysate and synthetic mediums were compared. Both of the solutions contained the same concentration of acetic acid, furfural and lignin derivatives. However, the ethanol yield (g ethanol /g glucose) and biomass productivity (g biomass/L medium hour) were 74.4% and 83% lower in the wheat straw .  medium than in the synthetic medium [55]. This researcher also tested the three inhibitors individually and found that the reduction in ethanol yield and biomass productivity were 30.2% and 59.6% for the medium with acetic acid, 7.0% and 12.8% for the medium with furfural and 14.0% and 44.7% for the medium with lignin derivatives [55]. From these observations, the effects of these inhibitory compounds are greater when they are combined.  27  1.5 Improving the fermentability of bio-oil hydrolysates Bio-oil  contains  approximately  162  compounds  that  are  detectable  by  gas  chromatography-mass spectrometry (GC-MS) [30, 32]. Some of the compounds are identified as inhibitors in the bio-oil, and the synergistic effect of the collective compounds in the bio-oil is unknown. However, despite the unknown compounds and their synergistic effect, there are ways to approach and solve the inhibition problems in the fermentation of bio-oil hydrolysate. There are four approaches to reduce and minimize the effect of inhibition of fermentation medium. The first approach is prevention. In this case, the objective is to avoid the formation of inhibitors prior to the fermentation process [40]. The second approach is to detoxify the hydrolysate before fermentation, which is also a major focus in this project.  The third approach is to  develop a species of microorganism capable of resisting inhibitors [40]. This can be achieved by genetically engineering microorganisms or by a technique called adaptive evolution. The fourth approach is to convert the toxic compounds into products that do not interfere with the metabolism of the microorganism [40].  1.5.1 Air stripping Air stripping is a common chemical engineering technique in which volatile organic compounds (VOCs) are extracted from a solution into air. In this project, air stripping is employed to remove the volatile acids such as acetic and formic acids that are in bio-oil hydrolysate. The removal of VOCs by air stripping is governed by Henry’s law. Thus, a larger Henry’s constant will results in greater removal of VOCs by air stripping. Consequently, temperature has an influence on air stripping since Henry’s constant increases with increasing temperature. Other factors that can affect the removal efficiency of air stripping include pH and contact time. It has been reported by Helle et al. [56] that air stripping at low pH (pH not specified) could remove volatile acids while at high pH (pH 10) ammonium could also be removed [56].  28  1.5.2 Adsorption on activated carbon Treatment with activated carbon is a common detoxification technique used for removing toxicants from liquids. The parameters which typically control adsorption are carbon dosage, temperature and contact time [40]. In a study using straw hemicellulose hydrolysate, it was found, that temperature affects how well the activated carbon can remove lignin degradation products [57]. It was found that the removal rate of the toxic compounds increased six fold when the temperature rose from 20 to 40°C [57]. These same researchers examined effect of pH on adsorption on activated carbon and found that a low pH (pH 2) favored the adsorption of the lignin degradation products, particularly the phenolic compounds [57]. The effect of contact time was also well documented by researchers. Parajo et al. [58] found that the removal of lignin-derived compounds was at maximum at around 20 minutes and the adsorption of the compounds reached equilibrium after 20 minutes [58]. Other researchers had found that using 1% (w/w) of activated carbon to treat sugarcane bagasse resulted in the removal of 94% phenolic compounds with a sugar loss of only 0.47% [59]. Increasing the amount of activated carbon to 30% (w/w) enhanced the removal of phenolic compounds slightly but the sugar loss increased to 31.3%, which was an undesirable result [59].  1.5.3 Overliming Overliming is regarded as a very effective method for detoxification of hydrolysates prior to fermentation [60]. Martinez et al. [60] conducted experiments on overliming of hemicellulose hydrolysate and reported removal of up to 51% of furans and 41% of phenolics, with minimal (9%) sugar loss [60]. A study by Yu et al. [61] has reported that using overliming to detoxify cotton-derived bio-oil hydrolysate, resulted in an ethanol yield of 0.39 g ethanol! g glucose [61]. In addition, Yu et al. [61] reported that combining overliming with other adsorbent materials such as diatomite, bentonite and zeolite can increase the ethanol yield further by 5  —  13% [61]. However, another study on treating  the hydrolysate with the overliming technique reported up to 20% loss of sugars [62]. The sugar loss was postulated to be due to calcium ion-catalyzed sugar degradation under  29  alkaline conditions [62]. In addition to the sugar loss, the formation of formic and acetic acid was observed [62]. At a temperature of 80°C and a treatment time of 3 hours, as much as a 2 to 3 fold increase in the concentration of acetic and formic acid was reported [62]. Despite the formation of formic and acetic acid, the removal of other inhibitors such as furfurals and phenols was effective [62]. Thus, the evaluation of this method and for other detoxification methods should be based upon fermentability and the yield of ethanol.  1.5.4 Coagulation and flocculation  The aim of using coagulation and flocculation to treat bio-oil is to remove the dissolved and suspended particles in the bio-oil. Coagulation and flocculation are commonly used in water treatment processes such as recovery of processed water, and treating municipal waste water. The addition of coagulants can destabilize the forces that keep the particles apart, so that formation of larger particles can occur. Then, flocculant is added to increase the cluster size of the particles, thus allowing the materials to settle.  1.5.5 Solvent extraction Solvent extraction is a method aimed at selectively removing inhibitory compounds such as acetic acid and formic acid from the fermentation medium. It has been shown that C8— dO saturated aliphatic tertiary amines dissolved in organic solvents are effective extractants for carboxylic acids [63]. Several aliphatic amines, for instance, alamine 336 and aliquat 336 with co-solvents such as kerosene and l-octanol, have been used successfully to extract acetic acid [64]. It has been confirmed by Yang et al. [64] that a polar diluent can increase the extracting efficiency of a non-polar amine by providing additional solvating power that allows higher levels of polar acid-amine complexes to stay in the organic phase [64]. In the design of an amine-diluent system for acid removal, there are three factors that can influence the extraction of carboxylic acids and should be  30  considered. They are: the nature of the acids, the concentration of the acids and the concentration of the arnines [63]. Other solvents such as oleic acid and oleyl alcohol were also evaluated in the extraction of the bio-oil hydrolysates. These two solvents were examined by researchers in the Netherlands [65]. They experimented with oleic acid and oleyl alcohol along with other solvents (hexane, octane, decane, dodecane, hexadecane, hexanol, octanol and decanol) for the feasibility of extracting inhibitory compounds from lignocellulosic hydrolysates. Zausten et a!. [65] found that there was a two-fold increase in ethanol productivity by extracting the fermentation medium with oleyl alcohol [65]. In addition, oleic acid and oleyl alcohol were found to have a biocompatibility that exceeds 100% [65]. The biocompatibilities were defined by measuring the change in carbon dioxide production from the yeast after the addition of 5% solvent during the fermentation processes [65].  1.5.6 Hydrogenation Hydrogenation is a reaction that results in the addition of a pair of hydrogen atoms to the reactants. The process has been investigated by a number of researchers to upgrade and stabilize bio-oil for transportation and storage [66, 67, 68]. The hydrogenation process requires metal catalysis as well as high temperature and pressure. One study found that two catalysts, Re 7 and 2 S 2 Re were able to hydrogenate dilute acetic acid solution into 7 0 ethanol at 230°C and 12.2 MPa of hydrogen pressure [69]. The reaction took about six hours, the ethanol yield was found to be 68.6% and the acetic acid reduction was found to be 76.9% for hydrogenation with the Re S-, catalysis [69]. Under the same conditions, the 2 ethanol yield is reported to be 41.1% and the acetic acid reduction was found to be 62.6% for experiments with 7 SO catalysis [69]. Other studies that are related to 2 Re hydrogenation reactions and reaction kinetics are focused on phenolic and furanic groups in the bio-oil [67]. Moreover, the hydrogenation experiments on bio-oil were focused on stabilizing and upgrading of bio-oil. Thus, very little information can be found regarding the removal of inhibitors from bio-oil hydrolysate with hydrogenation.  31  1.5.7 Adaptive evolution  Adaptive evolution is, by definition, a set of mutations that occur in response to a specific challenges or changes in the environment and is advantageous to the cells under these conditions [70]. This technique can potentially be used to develop an adapted yeast strain that can resist the inhibitory compounds in the fermentation broth. Thus, the adapted yeast strain can show an increase in the ethanol yield and biomass productivity without the need of any additional detoxification stage [70]. The stresses that can cause yeasts to function poorly include: temperature, pH, inhibitory compounds and limiting substrate [52, 70]. The mechanism of adaptive evolution is still not fully known, but there are three models that can potentially explain the process. The three models are: the directed mutation model, the hypermuation model and the cryptic growth model [70, 71]. In the directed mutation model, it is hypothesized that mutations are targeted toward genes in which a mutation directly relieves the stresses caused by the environment [70, 7!]. In the hypermutation model, it is postulated that genome—wide mutation rates increase such that both adaptive and non-adaptive mutations are stimulated [70, 71]. In this situation, most of the mutated genes are unhelpful in the adaptation of the stress but some mutations are useful, thus able to survive and out-compete other mutants. In the cryptic growth model, the mutation rate is hypothesized to be unaffected by the environmental conditions, hut mutations that occurred randomly appear to be adaptive due to  the survival of some mutants [70. 71].  The adaptive evolution models can be applied to yeast cultures. When a yeast population is propagated asexually under continuous long term stress, it undergoes a series of adaptive shifts. The clones that are well adapted to the environment successively replace one another and eventually replace the entire population of yeast [71]. Such changes have been observed to occur in every 50 generations [71]. The evolution can be identified by the appearance of mutants in the population, as each new and fitter mutant outgrows the population of the progenitor genotype [711.  .5  Chapter 2 Research Objectives The objective of this project is to determine the technical viability of producing bio ethanol using bio-oil as a substrate for fermentation. The major tasks involve in the project include:  1. Optimize the extraction of levoglucosan from bio-oil 2. Screen and evaluate techniques that could remove inhibitors and/or enhance the yield of ethanol a. Air stripping b. Adsorption on activated carbon c. Overliming d. Coagulation and flocculation e. Solvent extraction f.  Hydrogenation  g. Adaptive evolution of yeast  33  Chapter 3 Materials and Methods 3.1 Process overview Production of ethanol from bio-oil began with the extraction of levoglucosan with water. Adding a sufficient amount of water to bio-oil causes the bio-oil to separate into an aqueous phase and an organic phase. The aqueous phase would contain most of the levoglucosan. It can then be hydrolyzed with dilute sulfuric acid. The product of hydrolysis, the hydrolysate, is used as fermentation medium to produce ethanol [6, 7]. However, it was found that bio-oil hydrolysate contained fermentation-inhibiting compounds, thus the hydrolysates must first be detoxified [6]. After the detoxification stage, the detoxified hydrolysates were then used as fermentation medium for Saccharomyces cerevisiae, strain T2 to produce ethanol.  extraction hydrolysis  detoxification  fermentation  Figure 13 Process overview for ethanol production from bio-oil  3.2 Procedure for extraction of levoglucosan The objective of this part of the experiment was to optimize the extraction of levoglucosan and other anhydrosugars from bio-oil. The experiments were designed to 34  find the critical concentration of water required to induce phase separation and to determine the optimal mass ratio of water-to-bio-oil for levoglucosan extraction.  3.2.1 Storage and preparation of bio-oil The bio-oil used in the project came from the Technical Institute of Finland. It was stored in the refrigerator at about 3°C, to slow the aging of the bio-oil. The bio-oil used in this project has been stored for approximately 5 years. Before any experiment with the bio-oil, the oil was placed in an environmental shaker at 20°C and 120 rpm for approximately 20 minutes. This decreased the viscosity of the bio-oil and allowed the bio-oil to be poured out of the container easily.  3.2.2 Titration of bio-oil to determine the critical amount of water required to induce phase separation Water is soluble in bio-oil and the bio-oil will behave like a single phase until a critical amount of water is reached. As soon as the critical amount of water is added, phase separation will occur. In order to determine the critical point, a titration experiment with water and bio-oil was conducted. In the experiment, water was added to the bio-oil slowly until phase separation could be observed. When phase separation was observed, the volume of the water added to bio-oil was recorded. A photograph of the experimental apparatus is shown in Figure 14.  35  Figure 14 Bio-oil titration equipment  It is important to note that addition of a few drops of water might give the illusion of phase separation if there is inadequate mixing. Therefore, it was vital to provide sufficient mixing during the titration experiment. A photograph of the bio-oil at the point of phase separation is shown in Figure 15.  Figure 15 Phase separation of bio-oiJ  36  3.23 Levoglucosan extraction from bio-oil A preliminary experiment was conducted to test whether it was beneficial to add water to bio-oil or bio-oil to water. The ratios tested were 50 and 100 wt% of water added to bio oil. Before the experiment, both the bio-oil and water were weighed and measured according to the predetermined mass ratios as shown in Table 5. Then, the water was slowly added to the bio-oil (or vice versa) in a 50 mL beaker with a magnetic stirrer. During the extraction experiment, it was necessary to provide sufficient mixing time for the two liquids, which was approximately 20 minutes [6]. After 20 minutes, the aqueous phase samples were decanted into falcon tubes for storage and analysis. Table 5 Parameters for levoglucosan extraction with water  50%  Water add to bio-oil Weight of Weight of Bio-oil (g) water (g) 10 5  Bio-oil add Weight of Bio-oil (g) 10  100%  10  10  Wt% of water added to bio-oil  10  to water Weight of water (g) 5  10  Two methods of mixing were examined to determine which method would provide optimal extraction: the mixer method and the vortex method. Schematics of both mixing methods are shown in Figure 16. In the mixer method, bio-oil was poured into the water and a handheld kitchen mixer (Braun, Multiquick MR405) with steel blades was used to mix the liquids and grind the tar like organic phase formed in the initial separation of water and bio-oil. In the vortex method, bio-oil droplets were dripped into a falcon tube, which was secured on a vortex mixer. The flow of the bio-oil droplet was controlled by a valve at the end of a buret. The bio-oil was slowly dropped into the swirling liquid to enhance contact time and mixing between the liquids.  37  I Figure 16 Schematic for extraction of levoglucosan  In finding the optimal extraction ratios, the bio-oil and water were weighed and measured according to the predetermined ratios shown in Table 6. The extraction ratios examined ranged from 10  2000 wt% of water added into bio-oil. In this set of experiments, the mixer method was used and bio-oil was added to water. These were chosen based on the —  results from previous set of experiments on levoglucosan extraction. Table 6 Extraction ratios examined to determine the optimal ratios Wt% of water added to bio-oil  Weight of Bio-oil (g)  Weight of water (g)  10%  10  1  20%  10  2  40%  10  4  50%  10  5  100%  10  10  200%  10  20  400%  10  40  600%  10  60  800%  10  80  1000%  10  100  2000%  10  200  38  Small aliquots (1 mL) from the aqueous phase from the extraction experiments were taken with a pipette and analyzed by gas chromatography (GC). Since previous works had shown that temperature played a major role in the properties of bio-oil, it was beneficial to determine if temperature had any effect in the extraction of levoglucosan [31]. The two temperatures tested in levoglucosan extraction were 25°C and 80°C. The extraction ratio tested for this experiment was 100 wt% of water added into bio-oil. Similar to other extraction experiments, when the bio-oil extracts were cooled to room temperature, small aliquots from the aqueous phase were taken and analyzed by GC.  3.3 Dilute acid hydrolysis of the aqueous fraction of bio-oil The purpose of this part of the experiment was to hydrolyze anhydrosugars, such as levoglucosan and cellobiosan into glucose for fermentation in later stages of the experiments. In the project, sulfuric acid was used for hydrolysis because it had demonstrated to be effective and inexpensive [6]. The procedure for hydrolysis of levoglucosan was based on published methods [6, 61]. Prior to hydrolysis, samples from the aqueous phase were analyzed with gas chromatography to determine the molar concentration of levoglucosan. Then, an equal molar ratio of sulfuric acid and levoglucosan was used in the hydrolysis [6]. The hydrolysis was carried out with an autoclave (Steris Amsco Century, SV-1262 prevac steam sterilizer) at 121°C for 20 minutes (excluding heating and cooling time). After hydrolysis, the hydrolysate was cooled in a cold water bath until room temperature was reached. Then, the hydrolysate was sampled and filtered with a syringe filter with a pore size of 20 tim. Finally, the filtered sample was injected to a high pressure liquid chromatography system for analysis. A picture of the hydrolysate in 50 mL falcon tubes is shown in Figure 17.  39  Figure 17 Hydrolysate in 5OmL falcon tubes  3.4 Procedure for fermentation of bio-oil hydrolysate Fermentation was one of the most important stages of the experiment. The optimal conditions for ethanol production were governed by a number of different factors such as temperature, nutrients concentration, oxygen concentration, pH, residence time, mixing speed, and the presence of bacteria and antimicrobial compounds [53, 73]. Some of these factors could interact and act synergistically to affect the well-being of yeasts, which made the realization of true optimal conditions more complex [53].  3.4.1 Preparation of yeast inoculum for fermentation The yeasts were stored in refrigerator at 4°C on agar plates. When preparing the yeast for fermentation, a loop full of yeast from the agar plate was transferred aseptically to YPG medium containing 1% (w/v) of yeast extract, 2% (w/v) of peptone and 2% (wlv) of glucose. YPG medium contains nutrients that provide for starter yeast cultures. Yeast extract and peptone are complex substances that contain vitamins, minerals and proteins needed for cell growth. Glucose is needed to provide carbon and energy for the yeasts. The YPG medium was sterilized by autoclaving for 20 minutes at 121°C. The yeast was grown in the YPG medium for 1 full day in an environmental shaker at 30°C and 150  40  rpm. Then, 2 mL of the starter culture broth was transferred to fresh YPG medium to grow for another day. After two days of culturing, the yeast cultures were ready to be used as fermentation inoculum. Yeast broths from second stage growth flasks were transferred aseptically to 50 mL falcon tubes and centrifuged for 10 minutes at 2285g (DamonIIEC division CU-5000). This caused the yeast broth to be separated into a yeast pellet and supernatant (YPG medium). The supernatant was then discarded and the yeast pellet was resuspended with 10 mL of sterile water. This process was repeated once to wash the yeast. The purpose of washing the yeast was to remove any trace amount of YPG medium that might still be present during the first separation of yeast and YPG medium. After the yeast pellet was washed, it was ready to be inoculated into the fermentation medium. The volume of the yeast inoculated to fermentation medium was 2 mL, which yielded a concentration of approximately 2.5 gIL (in 50 mL fermentation medium). A flow diagram of the yeast preparation procedure is shown in Figure 18.  1St clay transfer  2nd day transfer  wash x2 4 yeast pellets fermentaLion 10 mL to 1 tube sterile water  Figure 18 Yeast preparation for use as inoculum It was important that sterile water was used for resuspending the yeast. In addition, yeasts were transferred in an aseptic environment; otherwise contamination by other microorganisms might generate inaccurate results.  41  3.4.2 Fermentation of bio-oil hydrolysate In this work, the fermentations were carried out in 15 and 150 mL serum vials. The 15 mL serum vials were used for determining the ethanol yield after 48 hours of fermentation, whereas the 150 mL serum vials allowed more samples to be taken during the 48 hour fermentation period. Depending on the goal of the experiments, the volume of the fermentation medium such as YPG medium and bio-oil hydrolysate ranged from 10-100 mL.  As previously mentioned, yeasts under micro-aerophilic conditions favour the production of ethanol. In order to achieve this condition, the serum vials were capped with a rubber stopper and sealed tightly with aluminum caps so that the vials were hermetically sealed (Figure 19). An 18-gauge, 1.5 inch needle and 5 mL syringe filled with foam was inserted into the vial through the rubber stopper to allow a small amount of air to circulate in the vials, thus simulating the micro-aerophilic conditions. The vials were incubated in an environmental shaker for maintaining the mixing speed and temperature at optimal conditions. In this work, the rotating speed of the shaker was set at 150 rpm and the temperature was set at 30°C. Short term modifications to the rotating speed of the shaker were made from time to time because the yeast would adhere to the wall of the serum vial as the cultures grew. The residence time was usually 48 hours, and samples were usually withdrawn from the serum vials at 1, 2, 4, 8, 12, 24, and 48 hours via syringes. In all of the fermentation experiments for this project, the pH of the hydrolysate was adjusted to 5.5 prior to fermentation but, overtime, the pH typically decreased to 4.5 —4.7.  42  Figure 19 Fermentation medium in serum vials  3.5 Procedure for detoxification of bio-oil hydrolysates A detoxification stage was necessary in order to increase the fermentability of the bio-oil hydrolysates and improve the production of ethanol. The detoxification techniques examined include air stripping, overliming, hydrogenation, adsorption and solvent extraction. The selected methods were targeted at different inhibitory compounds in the bio-oil. The targeted inhibitors in the bio-oil were furfural, phenolics, acetic acid, formic acid, and dissolved solids. The evaluation of these detoxification methods was based upon the fermentability of the bio-oil hydrolysates. In addition, a technique called adaptive evolution was used to help the yeast to adapt to the bio-oil hydrolysate, thereby increasing the yield of ethanol.  3.5.1 Air stripping An experiment was conducted to test if air stripping could increase the fermentability of bio-oil hydrolysate. The purpose of air stripping was to remove the volatile organic components in the hydrolysate. The parameters tested for this experiment were temperature and time. In this experiment, air was bubbled through a nozzle into flasks that contain bio-oil hydrolysate. Therefore, factors, such as packing materials, size of the packing materials and size of the stripper tower, were not studied. Prior to air stripping,  43  the pH of the bio-oil hydrolysate was adjusted to 10 by adding NaOH (50% w/w). The experiments were conducted at 25°C, 50°C and 80°C for 30 minutes and 1 hour.  3.5.2 Adsorption on activated carbon The procedure for adsorption on activated carbon was to first determine the weight of the bio-oil hydrolysate. Then, activated carbon powder (Fisher Scientific, Fisherbrand AC 50-200 mesh, Diameter 0.074-0.297 mm) was measured according to the predetermined concentration (Table 7) with an analytical balance (Ohaus, APi lOS analytical plus). Afterward, the bio-oil hydrolysate and the activated carbon were placed into an environmental shaker to allow the adsorption to take place in controlled conditions (Table 7). Finally, the bio-oil hydrolysates were transferred to 50 mL falcon tubes and centrifuged at 2285g for 5 minutes (DamonlIEC division CU-5000). The bio-oil hydrolysates were decanted and used for fermentation in subsequent experiments. Table 7 summarizes the parameters for the adsorption on activated carbon in this study. The parameters were chosen based on journal articles that had similar experiments [40, 57, 58, 59].  Table 7 Parameters for activated carbon adsorption Parameter  Condition A  Condition B  Temperature  25°C  40°C  pH  2  2  Concentration  1% (w/v)  10% (w/v)  Contact time  30 mm  1 h.  3.5.3 Overhming  In overliming, Ca(OH) 2 was added to the bio-oil hydrolysate (with adequate mixing) until a pH of 10 was reached. The pH of the bio-oil hydrolysate was measured with a pH meter (Thermo Orion, model 710 pH meter). To ensure the accuracy of the equipment, the pH 44  electrode was calibrated with buffer solutions (pH 4, 7, 10). The hydrolysate was then mixed with a magnetic stirrer for 30 minutes. Afterward, the hydrolysate was centrifuged for 15 minutes at 2285g (DamonIIEC division CU-5 000). The supematant was decanted to be used in subsequent fermentation experiments.  35.4 Coagulation and flocculation The coagulant and flocculant used in the project is called Callaway C-4907 and Profloc 17ORR. These two chemicals can be used in low pH (pH 2) environments similar to that of bio-oil hydrolysate. Both substances are called cationic polyacrylamide copolymers in water-in-oil emulsions. The chemistries and compositions of these two chemicals are listed as petroleum distillate and surfactant, whereas the concentrations of the substances are confidential.  The procedure for treating bio-oil hydrolysate with coagulation and flocculation was to pipette the predetermined concentration of the coagulant and flocculant (Table 8) into the bio-oil hydrolysate. Then, the liquids were mixed with magnetic stirrer for 10 minutes and allowed 20 minutes for the materials to congregate. Finally, the congregated material was removed by filtration (Whatman filter paper, Grade 1, 11 tim).  Table 8 Summary of the volume of coagulant and flocculant added Volume of chemicals (iL) 10 20 50 10 20 50 10 20 50  Chemicals Callaway C-4907 CallawayC-4907 CallawayC-4907 Profloc 17ORR Profloc 17ORR Profloc I7ORR Callaway C-4907&Profloc 17ORR Callaway C-4907&Profloc I7ORR Callaway C-4907&Profloc 17ORR  45  Volume of hydrolysate (ml) 10 10 10 10 10 10 10 10 10  3.5.5 Solvent extraction for detoxification of bio-oil hydrolysate The solvents used in the project were alamine 336, aliquat 336, tri-n-octylamine, tributyl phosphate, primene-JMT, oleic acid and oleyl alcohol. Alamine 336 and aliquat 336 are the trade name of chemicals, which are tri-octyl/decyl amine and had a structure of R N 3 and 3 NCH respectively [74, 75]. Primene JMT is also a trade name of a primary R , aliphatic amine which had a structure of R -2 3 C-NH [76]. Each of the five amines was used in conjunction with two different co-solvents. The co-solvents were I -octanol and kerosene.  In order to study the effect of the co-solvent, a set of four different co-solvent concentrations was tested on the amines. Solvent extractions were carried out in 50 mL falcon tubes at room temperature. Equal volume of the solvents and hydrolysates were used in the extraction experiment. They were mixed for approximately two hours in an environmental shaker at room temperature (---20°C). After two hours, the solvents and the hydrolysates were separated by centrifugation at 2285g for 10 minutes (DamonIIEC division CU-5000). The hydrolysates were then transferred by pipette to serum vials for fermentation. Table 9 shows the experimental design for the different sets of chemicals used for solvent extraction. The solvent extraction experiments were initially conducted using a model solution containing 10 gIL glucose and 60 gIL acetic acid. Performances of the solvents were evaluated based on the amount of acetic acid removed and the amount of sugar retained after the extractions. Then, experiments were conducted to test if the solvents were inhibitory to fermentation. This was tested by extracting 15 mL water with 15 mL of the proposed solvents. The water was then added to 35 mL of YP medium and supplemented with 10 gIL of glucose for fermentation.  46  Table 9 Experimental design for solvent extraction using model hydrolysate Solvents  (1  —  Co-solvents octanol or kerosene)  Alamine 336  Ovol%  Alamine 336  25 vol%  Alamine 336  5Ovol%  Alamine 336  75 vol%  Aliquat 336  Ovol%  Aliquat 336  25vo1%  Aliquat336  5Ovol%  Aliquat 336  75vo1%  Primene-JMT  0 vol%  Primene-JMT  25 vol%  Primene-JMT  50 vol%  Primene-JMT  75 vol%  Tri-n-octylamine  0 vol%  Tri-n-octylamine  25 vol%  Tri-n-octylamine  50 vol%  Tri-n-octylamine  75 vol%  Tributyl phosphate  0 vol%  Tributyl phosphate  25 vol%  Tributyl phosphate  50 vol%  Tributyl phosphate  75 vol%  Oleic acid  0 vol%  Oleyl alcohol  0 vol%  3.5.6 Hydrogenation for detoxification of bio-oil hydrolysate The reactors used in hydrogenation treatments are cylindrical reactors from PARR and Autoclave Engineers. The two reactors are similar and can be used interchangeably when the other reactor is not available. Both reactors are capable of handling a pressure of at  47  least 2000 kPa and are equipped with a rupture disk as a safety mechanism. The temperature and the stirring speed of the two reactors are controlled by controllers with digital output. Figure 22 24 shows the two reactors and the controllers used in this —  project.  Figure 20 Reactor from Autoclave Engineers used for hydrogenation of artificial hydrolysates  Figure 21 Temperature control (bottom unit) and stirrer control (top unit) for Autoclave Engineers reactor  48  Figure 22 PARR reactor used for hydrogenation of artificial hydrolysate  The catalyst used in the reaction was molybdenum disuiphide 2 (MoS ) , which was directly prepared in the reactor. The preparation of the catalyst involved the addition of 0.1 mL of carbon disuiphide and 2 mL of ammonium heptamolybdate (AHM) solution. Reactions of hydrogen gas with carbon disuiphide and AHM solution would generate the desired catalyst. 2 CS  +  2 H  —*  4 CH  +  S 2 H  S+AHM H + 2 + 3 0 NH H —*MoS  C + + 4 + 3 O 2 AHM +M N H S —C H oS H The concentration of the catalyst MoS 2 prepared in the reactor was approximately 600 ppmw. The experiments were conducted with 100 mL of artificial solutions that simulate the bio-oil hydrolysate and the aqueous phase. The temperature was set at 250°C. The stirrer was set at approximately 400 rpm. During the startup, the reactor was purged with nitrogen for 5 minutes and then pressurized with hydrogen to 150 psi for the low pressure condition and 400 psi for the high pressure condition. The compositions of the artificial solutions are shown in Table 10.  49  Table 10 Concentration of levoglucosan, glucose and acetic acid in the artificial solutions Compounds  Solution A (gIL)  Solution B (gIL)  Levoglucosan  7  0  Glucose  0  10  Acetic acid  50  50  3.5.7 Adaptive evolution Adaptive evolution is not a detoxification method, but rather a technique that is aimed at alleviating the effect of the inhibitory compounds through the adaptation of the yeast to inhibitory environments. A known percentage of bio-oil hydrolysates were diluted in YP medium (Table 11). Yeast inoculums were cultured and prepared in YPG medium. Then, the yeasts were inoculated to the YP medium that contained bio-oil hydrolysate, which allowed the yeast to adapt at a non-lethal level. The yeasts were cultured in shake flasks and transferred to new fermentation medium under the same stress conditions every day. It had been reported that the mutants will replace their progenitor genotype every 50 generations [72]. A growth curve was prepared for the yeasts and by solving the equation of the curve, each generation took approximately 7 hours (Figure 23). In this work, the yeasts were adapted for a total of 36 days. In this period of time, the yeasts were adapted for approximately 121 generations. The parameters investigated were concentration of hydrolysate and the growth conditions of the yeasts (aerobic and micro-aerophilic). Table 11 shows the experimental design for this experiment.  50  Table 11 Growth conditions tested for adaptive evolution of yeast Sample  Culturing conditions  Hydrolysate volume in medium (vol %)  1  aerobic  35%  2  aerobic  35%  3  aerobic  10%  4  aerobic  10%  5  Micro-aerophilic  35%  6  Micro-aerophilic  35%  7  Micro-aerophilic  10%  8  Micro-aerophilic  10%  6 5 4  0.92251n(x) + 2.5884 =0.97 2 R  y  3  —  C? C?  2 1 0 0  5  10  15  20  25  30  Timer(hours) Figure 23 Typical yeast growth curve for S. cerevisiae strain T2  51  3.6 Analytical methods The samples taken from hydrolysates and fermentation broths were analyzed with gas chromatography (Varian Inc., CP-3 800) and high pressure liquid chromatography (Dionex Corp., ED-50) for ethanol and levoglucosan analysis, and sugar analysis, respectively. The concentration of yeast was determined by generating a standard curve formulating the linear relation with optical density and concentration. The optical density was measured with a spectrophotometer (Mandel Scientific Inc., Pharmaspec UV-1700) at 600 nm.  3.6.1 Determining yeast concentration In order to determine the yeast concentration during fermentation, 1 mL of sample was taken from the fermentation medium at the predetermined sampling time. Each sample was then spun down by centrifugation at l0000g for 10 minutes (Sanyo, MSE micro centaur). The supernatants were used for ethanol analysis, while the yeast pellets were resuspended with I mL of deionized water. This process was repeated twice to wash the yeast and ensure the complete removal of any fermentation medium. The re-suspended yeast pellets were then stored in a freezer and used for optical density (OD) analysis. In order to analyze the yeast samples, they were thawed, diluted 20 40 times and were transferred to a cuvette for absorbance analysis in a spectrophotometer. The wavelength —  used for yeast analysis was 600 nm. The OD was compared with the standard curve prepared for the yeast to determine the yeast concentration. The standard curve was prepared by diluting a known concentration of the yeast. The yeast sample was diluted to an OD reading between 0.2—0.9 and the OD reading was recorded at each dilution (Figure 24).  52  0.9  0.6 0.5 0.4  .  _y=0.753x. = 0.99 2 R  0.1 0.0  —..  0.0  0.2  .  0.4  .  0.6  0.8  1.0  1.2  OD600nm Figure 24 Optical density calibration curve for S. cerevisiae strain T2  3.6.2 Analysis of ethanol and levoglucosan by gas chromatography A gas chromatography (GC) system (CP-3800, Varian Inc.) was used in the project to analyze the concentration of levoglucosan and ethanol. Samples used for ethanol analysis were stored in a freezer. Prior to analysis, the samples were thawed and diluted ten times. The solvent used for dilution was either deionized water for ethanol analysis or acetone for levoglucosan analysis. Then, an internal standard was added to determine the accuracy of the signal output by the GC. The internal standard used was 1 -butanol for ethanol analysis or 1 -octanol for levoglucosan analysis. The diluted samples and internal standards were added to the sample vials manufactured specifically for the auto-sampler (CP-8400, Varian Inc.) and mixed for 2 minutes by placing them on a vortex. A software called Star was used to communicate between the GC and the computer. There were five methods programmed to operate the OC. The methods were called standby, ethanol analysis, sugar analysis, condition-ethanol and condition-sugar. When the GC was not in use, the standby method was selected to keep the column at 40°C and maintain the flames for the flame ionization detectors (FIDs). Prior to ethanol analysis  53  and levoglucosan analysis, the condition-ethanol or condition-sugar method was selected to prepare the GC column for analysis. The condition methods would run for about an hour to purge any impurities out of the system. When one of the analysis methods was selected, the auto-sampler would inject, and analyze the samples automatically. A summary of the methods is tabulated in Table 12. Table 12 Summary of gas chromatography methods used for ethanol and levoglucosan analysis  Specification Volume of injection Temperature of column-oven Injector temperature Detector temperature Auto-sampler cleaning solvent  Internal standard Analysis time/duration  GC methods Condition Condition ethanol sugar  Ethanol analysis  Levoglucosan analysis  lpL  1tL  50 °C  250 °C  45  200 °C  270 °C  200  200 °C  285 °C  Deionized water I -butanol (10 g/L)  Methylene chloride 1 -octanol (10 gIL)  minutes  72 minutes  Opi.  Standby  Opt  OpT.  DC  250 °C  40 °C  DC  270 °C  200/270 °C  200 °C  285 °C  200/285 °C  N/A  N/A  N/A  N/A  N/A  N/A  1 hour  1 hour  N/A  3.6.3 Analysis of sugars by high pressure liquid chromatography The high pressure liquid chromatography (HPLC) system (ED-50, Dionex Corp.) is connected to a computer and controlled by a program called Chromeleon Chromatography Management System. This program can control the temperature, pressure, injection volume, as well as the output of the results. Table 13 summarizes the operating conditions for carbohydrate analysis with the HPLC.  54  Table 13 Operating conditions for high pressure liquid chromatography analysis of carbohydrate Carbohydrate method 30 °C 200 2000 psi 10-13 25 !IL 250 jiL Fucose (5g/L) 60 minutes  Specification Temperature Pressure pH Injector volume Flush volume Internal standard Analysis time  -  The procedures for sample preparation for HPLC analysis are similar to those for the GC. Samples in storage were thawed and diluted 50 times in nanopure water. Due to the sensitivity and vulnerability of the HPLC column, nanopure water must be used throughout the entire system. The specification of the nanopure water is shown in Table 14.  Table 14 Specifications of nanopure water used in high pressure liquid chromatography Specification Resistivity at 25 °C Total organic carbon (TOC) Bacteria pHat25°C Particles  18.2 MQ cm2/cm 1 ppb 1 CFU/ml 7 <0.2 tm <  <  In order to analyze the sugar content of the samples with the HPLC system, a number of preparation procedures are required. Before every set of analysis samples, it was necessary to prime the GP-50 gradient pump and the TTL pump in the HPLC system to make sure that there were no leaks throughout the piping network. Then, a ready check was performed using the Chromeleon program. When all the preparations were complete, data acquisition was activated and ran for 10  —  15 minutes until signals from the HPLC  stabilized. After the signal from the HPLC system was stabilized, the analysis could then begin manually.  55  Chapter 4 Levoglucosan extraction from bio-oil 4.1 Bio-oil titration to determine phase behavior The first goal in levoglucosan extraction was to determine the critical weight ratio of water and bio-oil that would induce phase separation. This was determined through several titration experiments. In the experiments, 20.14 mL of bio-oil was used as the solvent and water was slowly added to bio-oil until phase separation occurred. Phase separation was observed at about 2.38 mL of water. In mass ratio, this was approximately 9.86 wt% of water added to bio-oil. Comparing this with the results from Peacocke et al. [33], who indicated that phase separation occurred at 20 23 wt%, shows a significant difference [33]. Other researchers have reported that phase separation occurs at 27 wt% for birch, 23 25 wt% for pine and 31 wt% for poplar [6, 30]. However, these results —  —  were reported as the weight percentage of total water within the system [6, 33]. The water content of the bio-oil for this project (produced by VTT) was reported to be in the range of 17.4 21.1 wt% [6, 30, 77]. Thus, the total water content that induced phase —  separation was found to be in the range of 27.3 —31.0 wt%.  4.2 Suspended solids in aqueous phase Upon separation, the aqueous phase was poured into a falcon tube for centrifugation at 4650g (Damon/IEC division, CU-5000). This was to remove any suspended solids that might be present in the aqueous phase. Initially, the aqueous phase was clear after centrifugation and solid particles were removed. However, suspended solids were observed at the bottom of the container when aqueous phase samples were stored in a refrigerator at 4°C or at room temperature for more than one day. Since bio-oils were known to be chemically unstable, it was hypothesized that suspended solids were forming in the aqueous phase.  56  4.3 Effect of temperature on levoglucosan extraction An experiment was conducted to determine if temperature has any effect on the extraction of levoglucosan. Levoglucosan was extracted with 100 wt% of water at two different temperatures (25°C and 80°C). It was found that altering the temperature had no significant effect on the amount of levoglucosan being extracted from bio-oil. The average concentration for levoglucosan extracted at 25°C was found to be 51.4±5.2 g/L and 50.3±1.5 g/L at 80°C. At high temperature, bio-oil was observed to be completely dissolved into water in equal mass ratio (1:1). This resulted in a single phase solution. However, phase separation was observed once the solution cooled down and the amounts of levoglucosan extracted were similar to cold water extraction.  4.4 Effect of the order of so’vent addition on levoglucosan extraction Initial experiments were conducted to determine whether the addition of water to bio-oil and the addition of bio-oil to water would generate the same results. The results obtained from this experiment are shown in Figure 25. It was found that at water to bio-oil extraction ratios of 50 wt% and 100 wt%, the differences were 1.96% and 1.89%, respectively. The standard deviation of the levoglucosan method was evaluated with five runs of standard aqueous phase extracted from the same batch of bio-oil and was found to be 4.9%. Comparatively, the differences in levoglucosan concentrations between adding bio-oil to water and adding water to bio-oil were small. Thus, it was safe to conclude that the amount of levoglucosan extracted was independent of whether water was added to bio-oil or vice versa. From this result, the subsequent levoglucosan extraction experiments were conducted by the addition of bio-oil to water.  57  60 50 40 30 20 10 0 50% 100% Weight percentage of water to bio-oil examined in levoglucosan extraction(wt%) Figure 25 Comparison of levoglucosan extracted between the addition of water to bio-oil (light colored bars) and bio-oil to water (dark colored bars)  4.5 Effect of mixing on levoglucosan extraction In addition to the order of adding bio-oil and water, two different mixing methods were evaluated. They were referred as the handheld mixer and the vortex mixer method (Figure 16). The extractions ratios examined were 50 wt% and 100 wt% of water in bio oil. These two methods were to examine if the tar like organic phase formed during the initial contact of bio-oil and water was interfering with the levoglucosan extraction. Thus, the handheld mixer was used to break and mash the congealed organic phase, while the vortex mixer provided only gentle mixing. It was found that the difference between the two method of mixing were small (Figure 26). At 50 wt% and 100 wt% the difference between the two mixing methods were 1.5% and 4.6%, respectively. This indicated that the organic phase did not act as a barrier to the extraction of levoglucosan.  58  70 60 ;50 C  20  io 50% 100% Weight percentage of water to bio-oil examined in levoglucosan extraction [wt%] Figure 26 Comparison of levoglucosan extraction between handheld mixer (light colored bars) and vortex mixer (dark colored bars)  4.6 Optimal water-to-bio-oil ratio for levoglucosan extraction It was seen in section 4.3  4.5 that variation of the parameters (temperature, mixing, adding water to bio-oil or vice versa) had minimal effect on levoglucosan extraction. —  However, a common trend was observed for both of the above experiments, which exhibited a decrease in the concentration of levoglucosan extracted when the extraction ratio decreased from 100 wt% to 50 wt%. The average concentration of levoglucosan at 100 wt% extraction ratio was 4% and 7% higher than extraction at 50 wt% in the first experiment (addition of water to bio-oil or vice versa) and the second experiment (different mixing method), respectively. This suggested that the amount of water used in levoglucosan extraction could have more influence than the other parameters. Thus, a set of experiments with extraction ratios from 100 wt% to 2000 wt% was conducted. The results were shown in Figure 27. A decreasing trend was observed for the ratios examined. The data were fitted with a second degree polynomial with coefficient of determination of 0.97.  59  6.0% 5.0% 0 4.0/o  3.0% 2.0% 1fl0/ 1.V/0 —•  0.0% 0%  500%  1000%  1500%  2000%  Weight percentage of water used in levoglucosan extraction(wt%) Figure 27 Effect of water addition on levoglucosan extraction (100— 2000 wt% water-to-bio-oil)  Further extraction experiments with less than 100 wt% of water are needed in order to determine the optimal extraction ratio of water to bio-oil. From the titration results, 9.86 wt% of water in bio-oil was needed to induce phase separation. Thus, the lowest ratio chosen for examination was 10 wt%. Figure 28 shows the results for experiments with low extraction ratios (10 wt% 50 wt%). -  It can be seen that there was an increasing trend in the extraction of levoglucosan with an increase in the mass ratio of water used in the extractions (Figure 28). A logarithmic trend was able to fit the set of results with a coefficient of determination value of 0.86. Combining Figure 28 with previous results for higher water to bio-oil extraction ratios (Figure 27) it was found that the optimal ratio for levoglucosan extraction with water was 100 wt%.  60  5%  0.86  0%  0%  I  I  10%  20%  30%  40%  50%  Weight percentage of water added to bio-oil examined in levoglucosan extraction (wt%) Figure 28 Effect of water addition on levoglucosan extraction (10— 50 wt% water to-bio-oil)  The maximum levoglucosan concentration obtained was approximately 4.98 wt% (g levoglucosan! g bio-oil). In Bennett’s work, the maximum value of levoglucosan extracted was reported to be 5.1 wt% (g levoglucosan! g bio-oil) at an extraction ratio of 60 wt% (water to bio-oil) [6]. Although the maximum value of levoglucosan extracted coincided with Bennett’s work, the optimal extraction ratio was significantly different. In Bennett’s work, at 100 wt% extraction ratio of water to bio-oil, the weight of levoglucosan extracted was 2.2 wt% (g levoglucosan! g bio-oil) [6]. A possible explanation to the differences was that the aging of the bio-oil caused a change in the composition of compounds and properties of bio-oil.  4.7 Acetic acid in aqueous phase Acetic acid was one of the inhibitory compounds that got extracted along with levoglucosan. The amount of acetic acid found in the aqueous phase varies depending on  61  the extraction conditions. Figure 29 shows the amount of acetic acid extracted for the tested extraction ratios of water to bio-oil.  8% C?  7%  C?  6%  -.  C? C?  5% C?  _  C?  4% 3%  C?  2% 1% 0% 0%  500%  1000%  1500%  2000%  Weight percentage of water added to bio-oil examined in levoglucosan extraction (wt%) Figure 29 Weight percentage of acetic acid extracted from bio-oil (10-2000 wt% water-to-bio-oil) An increase in the concentration of acetic acid was observed as the extraction ratio increased from 10 wt% to 50 wt%. The maximum amount of acetic acid extracted from bio-oil was 6.56 wt% (at extraction ratio of 100 wt% water to bio-oil). Similar to the levoglucosan samples, the consistencies of the GC results were tested with five aliquots from the same sample of bio-oil extract. The standard deviation was found to be 8.2%.  4.8 Other compounds in aqueous phase A number of compounds were extracted to the aqueous phase. An average of 63 peaks (excluding internal standard) were detected in the aqueous phase for the proposed extraction ratios. In a similar experiment, Siplar et al. [78] induced phase separation by adding water to bio-oil at a mass ratio of 1:10 and identified 45 compounds in pine wood  62  derived bio-oil with gas chromatography and mass spectrometry [78]. The name of the compounds can be found in Table 18 in the Appendix. In this project, the identities and concentrations for most of the compounds were unknown. However, they were quantified indirectly according to the mass transferred from the bio-oil fraction to the aqueous fraction. Figure 30 shows the results of mass transferred from bio-oil to the aqueous fraction after extracting bio-oil with water in various extraction ratios.  70% ?  60%  ‘  50% 40% . 20% 10% 0%  0%  500%  I  I  I  1000%  1500%  2000%  Weight percentage of water added to bio-oil examined in levoglucosan extraction (wt%) Figure 30 Mass transferred from bio-oil to aqueous phase (1 0-2000 wt% water-to bio-oil)  From the results obtained, it was found that the mass transferred from the bio-oil to the aqueous phase decreases by 7 30% as the water-to-bio-oil extraction ratios deviated —  from the 1:1 mass ratio of water to bio-oil. This finding reinforced previous finding that at 100 wt% water to bio-oil or 1:1 mass ratio, extraction was most efficient. However, the efficiency was not limited to levoglucosan but also includes some of the compounds that are extractable by water.  63  Chapter 5 Methods for improving the fermentability of bio-oil hydrolysate 5.1 Effect of hydrolysate concentration on fermentation by S. cerevisiae Due to the presence of inhibitors in bio-oil, it was necessary to dilute the bio-oil hydrolysate with sterile water to lower the concentration of the inhibitors so that the yeasts were able to survive in the medium. An experiment was conducted to detenriine the highest concentration of the bio-oil hydrolysate that the yeasts were able to survive. The bio-oil hydrolysates were diluted based on the volume in milliliters and were referred as the strength of the hydrolysate. For example, 25% strength hydrolysate would mean 25 vol% of hydrolysate and 75 vol% of sterile water. The strength of the hydrolysate tested ranged from 2  —  45%. Figure 31 shows the concentration of ethanol produced in the  fermentation medium (52 mL), and Table 15 shows the ethanol yield based on available glucose for each set of the experiments.  Figure 31 Ethanol produced in various strengths of hydrolysate  64  Table 15 Effect of hydrolysate concentration on ethanol yield Fermentation medium 2% hydrolysate 5% hydrolysate 10% hydrolysate 20% hydrolysate 25% hydrolysate 30% hydrolysate 35% hydrolysate 40% hydrolysate 45% hydrolysate  Yield (gethanol/ g glucose) 0.20±0.08 0.49±0.0 1 0.49±0.04 0.42±0.09 0.18±0.01 0.18±0.01 0.17±0.0 1 0.0 0.0  It was found that the ethanol yield was highest for 5% hydrolysate. At 2% hydrolysate, the ethanol yield was only 0.20±0.08 (g ethanol/g glucose). This was probably due to the scarce amount of the sugar available to the yeast. In terms of toxicity, hydrolysates at 5% and 10% strength were able to achieve a relatively high yield of 0.49±0.01 and 0.49±0.04 (g ethanol/g glucose). These two values were close to the theoretical ethanol yield of 0.57 (g ethanol/g levoglucosan), which had an equivalent concentration of 0.51 (g ethanol! g glucose) reported by Bennett [6]. However, it was found that the hydrolysate became more inhibitory at 20% strength. The yield of ethanol was decreased by 14% when the strength of hydrolysate increased from 10 to 20%.  As the strength of hydrolysates  increased from 20% to 25%, the ethanol yield was decreased by 58.4%. At 40% v!v hydrolysate, the yeasts were not able to produce any ethanol because the concentration of bio-oil hydrolysate was too high in the fermentation medium. This set of experiments provided insights for evaluating the detoxification methods proposed.  5.2 Detoxification of artificial aqueous phase and hydrolysate by hydrogenation The first detoxification method experimented was hydrogenation of artificial solutions. As mentioned in the materials and methods (section 3.5.6), the solutions contained 7 g!L of levoglucosan and 50 gIL of acetic acid in solution A and 10 gIL of glucose and 50 gIL of acetic acid in solution B. The use of levoglucosan and glucose could provide some  65  insight as to whether this detoxification method was best applied before, or after the hydrolysis of the aqueous phase. The results for the two sets of experiment are tabulated in Table 16.  Table 16 Concentration of levoglucosan, glucose and acetic acid in artificial solution A and B after hydrogenation Rxn time (hr) 0 1 2 3 4  Artificial solution A LG conc. AA conc. (gIL) (g/L) 7.0 50.0 4.4 43.6 0.8 25.8 0 17.3 0 23.0  Artificial solution B Glucose Acetic acid conc. (gIL) conc. (gIL) 10.0 50.0 3.4 30.4 0.9 26.3 0 19.4 0 20.2  It was found that the degradation of levoglucosan was very rapid. This was especially true at higher temperature. This observation was made because the temperature was steadily increasing from 0 hour to 1 hour. After about 25 minutes of reaction time, the temperature reached 250°C, which was the proposed hydrogenation temperature. Thus, the slower degradation rate of levoglucosan during the first hour was very likely due to the relatively milder overall reaction temperature. This was also true for acetic acid degradation. However, the degradation of acetic acid was much slower than levoglucosan. It took about 2 hours to reduce the amount of acetic acid by 50% while levoglucosan degraded by 88.6% in 2 hours. It was found that the degradation of levoglucosan, glucose and acetic acid followed first order reaction kinetics (Figure 32). The first order reaction constant was found to be 1.08 hr, 1.20 hr and 0.34 hr for levoglucosan, glucose and acetic acid, respectively (Figure 32).  66  0.5 0 -0.5 c-)  -2 • Levoglucosan SGlucose Acetic Acid  -2.5 -3  Time (hr)  Figure 32 First order reaction kinetics for hydrogenation of levoglucosan, glucose and acetic acid  In order to determine the feasibility of this technique the reaction time of sugar was also considered. This was to determine if the bio-oil hydrolysate could generate better results. Table 16 shows the results of hydrogenation of the artificial hydrolysate. It can be seen that dextrose had been reacted completely in 3 hours while a substantial amount of acetic acid remained unreacted. This was similar to the results reported by van Gorp et al [79]. In van Gorp’s work [79], glucose was hydrogenated at 100°C and 120°C with a hydrogen pressure of 40 bar [79]. All of the glucose was reacted in approximately 80 minutes at 100°C and 140 minutes at 120°C [79]. Comparing the rate of degradation of dextrose and levoglucosan, it was found that dextrose degraded 30% more than levoglucosan during the first hour of hydrogenation.  There was a possibility that the difference in concentration between the sugars and acetic acid could generate biased results. Nevertheless, the initial experiments suggested that hydrogenation would not be very beneficial if the degradation was not selective. In addition to the non-selective nature, the degradation rate of acetic acid was relatively  67  slow compared to the sugars. All in all, hydrogenation might not be beneficial for the project under the proposed operating conditions and catalyst.  5.3 Detoxification using activated carbon, air stripping and overliming Commonly used detoxification techniques were applied to treat bio-oil hydrolysate. The techniques used were air stripping, adsorption on activated charcoal and overliming. These techniques were targeted toward compounds such as phenolics, furans and volatile organic compounds in bio-oil hydrolysates, which were inhibitory to fermentation. As mentioned in the materials and methods section, instead of identifying and quantifying the compounds adsorbed, the evaluation of the detoxification techniques were based on the ethanol yield from the fermentation of the treated bio-oil hydrolysate. All of the treated bio-oil hydrolysates were diluted with sterilized water to 40% and 50% strength before initiating fermentation. Air stripping at the proposed temperature, time and pH (25°C, 50°C and 80°C for 30 minutes and 1 hour, pH 10) failed to detoxify the bio-oil hydrolysates to a sufficient level for ethanol production. A possible explanation was that the volatile components in the aqueous phase were not the major inhibitors in the fermentation process. Thus, removing the volatile components via air stripping did not improve the fermentability of the hydrolysates or none of the inhibitory compounds was removed from the hydrolysate.  For adsorption with activated charcoal, the same results were observed. In the experiment, 1 wt% and 10 wt% of activated charcoal were added to the hydrolysate at two different temperatures (25°C and 40°C). Hydrolysates treated with activated charcoal under these conditions showed no improvement in the fermentability. In addition, no ethanol was produced at 40% and 50% strength of bio-oil hydrolysate. However, the treated hydrolysates did become lighter in color which suggested some degree of chemical removal by the activated charcoal.  68  The overliming technique showed more promising results. At 50% strength of bio-oil hydrolysates, the yield of ethanol was 0.19±0.01 (g ethanol/g glucose). This was a significant improvement in the fermentability of the hydrolysates. The improvement in the fermentability and the yield of ethanol was most significant on the 40% strength bio oil hydrolysate. As one might recall from previous results, the hydrolysate was found to be too toxic for the yeast to produce any ethanol at 40% strength. However, the ethanol yield at 40% strength was 0.45±0.05 (g ethanol/g glucose) when treated with the overliming techniques. This ethanol yield was not far from the ethanol yield from the controlled fermentations with YPG mediums (0.47 g ethanol/g glucose) and the theoretical ethanol yield reported (0.51 g ethanol/g glucose) [6].  5.4 Detoxification by coagulation and flocculation Treatments using coagulant and flocculant removed 3.4—6.5 wt% of materials from bio oil hydrolysate by settling. Table 17 shows the weight of the materials removed, as well as the weight percentage of the settled material. However, the removal of these materials did not improve the fermentability of bio-oil hydrolysate at 40% and 50% strength. Thus, a higher dosage of coagulant and flocculant (100 tL/10 ml) was attempted but this resulted in the hydrolysate forming a thick paste which behaved like single phase. This was undesirable as no material could be removed by settling or filtration.  69  Table 17 Materials removed by treatment with coagulation and flocculation Settled weight (g)  Settled weight (wt%)  0.575 1  3.60%  0.5983  3.82%  0.7018  4.91%  0.5488  3.38%  0.6846  4.76%  0.8501  6.30%  Both (10pL/10ml hydrolysate)  0.6 136  4.03%  Both (2OpL/lOml hydrolysate)  0.6000  3.96%  Both (5OjiL/lOml hydrolysate)  0.8663  6.52%  Chemicals added Callaway C-4907 (lOpL/lOml hydrolysate) Callaway C-4907 (2OiiL/IOml hydrolysate) Callaway C-4907 (5OpL/lOml hydrolysate) Profloc 17ORR (lOjiL/lOml hydrolysate) Profloc I7ORR (2011L/lOrnl hydrolysate) Profloc I7ORR (5OiiL/lOml hydrolysate)  5.5 Detoxification by extraction with organic solvents In the solvent extraction experiments conducted, the targeted inhibitor in biooil hydrolysate was acetic acid. Initial experiments with solvent extractions involved artificial hydrolysates, which consisted of only sugar and acetic acid. Further experiments were then conducted to determine if the selected solvents were inhibitory to fermentation. This was achieved by comparing the amount of ethanol produced with fermentation medium extracted with the proposed solvents and the control medium as described in materials and methods (Chapter 3, section 3.5.5). Based on these results, the solvents were screened and selected to apply to bio-oil hydrolysates. In addition, the results of the best three solvents were compared.  70  5.5.1 Solvent extraction for acetic acid removal from artificial bio-oil hyd rolysate  The artificial hydrolysates used in the experiments contain approximately 10 gIL of glucose and 60 gIL of acetic acid. Figure 33 shows the percentage removal of acetic acid with 1 -octanol and kerosene as the co-solvent. The percentage removals were calculated based on the concentration of acetic acid after the extractions. From Figure 33, it can be seen that the removal of acetic acid increases as the volume of 1 -octanol in the system increases for alamine 336, primene-JMT and tri-n-octylamine. The two highest amounts of acetic acid removal were by the solvents alamine 336 and tri-n-octylamine. At 25 vol% alamine 336 in 1-octanol and 25 vol% tri-n-octylamine in 1-octanol removal of up to 86±5.2% and 90.4±6.8% of acetic acid were achieved.  Kerosene, another co-solvent examined in the extraction experiments, was compared against 1-octanol. The comparison is shown in Figure 33. It was found that the amount of acetic acid removed by the solvents with kerosene was lower than with 1 -octanol. There was a decreasing trend in the amount of acetic acid being removed with an increasing volume of kerosene in the solvent system. This indicated that as a co-solvent, kerosene was impairing the extraction of acetic acid by other solvents. However, contrary to this general trend, an increase in acetic acid removal was observed with increasing volume of kerosene for primene-JMT.  71  100%  80%  60%  40%  20%  0% TBP TBP TBP TBP ALl ALl ALl ALl ALA ALA ALA ALA JMT JMT JMT JMT TOA TOA TOA TOA 100% 75% 50% 25% 100% 75% 50% 25% 100% 75% 50% 25% 100% 75% 50% 25% 100% 75% 50% 25%  Figure 33 Acetic acid removals, co-solvents: 1-octanol (light colored bars) and kerosene (dark colored bars), solvents: TBP ALA  =  alamine 336, JMT  =  =  tributyl phosphate, ALl  primene JMT, TOA  =  =  aliquot 336,  tri-n-octylamine  The amount of glucose retained was also an important aspect to the evaluation of the solvent extraction methods. The results of this experiment are shown in Figure 34. Some of the tested solvents were able to selectively remove acetic acid while completely retaining glucose. This was true for aliquot 336 in 1-octanol, alamine 336 in 1-octanol, tributyl phosphate in 1 -octanol and tri-n-octylamine in 1 -octanol. The rest of the solvents also showed very good selectivity towards acetic acid removal. The retention of glucose ranged from 76.3% to 100%. The two best solvents for acetic acid removal, 25 vol% alamine 336 with 1-octanol and 25 vol% tri-n-octylamine with 1-octanol retained 97.3±1.9% and 100±1 .4% glucose, respectively. The selective removal of acetic acid further supported the feasibility of detoxifying the bio-oil hydrolysates with these two solvents.  72  100%  ——  ;  —-——  ——  .4  80%  • 60%  40%  20%  0% TBP TBP 100% 75%  TBP 50%  TBP ALl ALl 25% 100% 75%  AL! 50%  ALT ALA ALA ALA ALA JMT JMT JMT JMT TOA TOA TOA TOA 25% 00% 75% 50% 25% 100% 75% 50% 25% 00% 75% 50% 25%  Figure 34 Percentage of glucose retained after extraction, co-solvents: 1-octanol (light colored bars) and kerosene (dark colored bars), solvents: TBP phosphate, ALl  =  aliquot 336, ALA  =  alamine 336, JMT  =  =  tributyl  primene JMT, TOA  =  tri-n-octylamine  5.5.2 Evaluation of the toxicity of the solvents towards fermentation  During the extraction process, trace amounts of the solvents could be dissolved in the bio-oil hydrolysate. This could introduce new inhibitors to the hydrolysate. Thus, it was necessary to evaluate the toxicity of the solvents. This was done by comparing the yeast growth and ethanol yield of extracted artificial hydrolysates with control samples. The control samples were YP medium (1% yeast extract, 2% peptone) with a supplemented sugar concentration of 10 gIL. The proposed solvents were used to extract distilled water as described in the materials and methods section (Chapter 3, section 3.5.5). Thus, the extractions were only meant to contaminate the model hydrolysate with the proposed solvents from the detoxification experiment. In this experiment, acetic acid was not  73  added to the artificial hydrolysate to reduce the number of factors and make the results easily comparable.  It was found that three of the solvents, aliquot 336, primene-JMT and tributyl phosphate severely inhibited yeast activity, with no growth observed in the artificial hydrolysates (Figure 35). Furthermore, no ethanol was detected in the hydrolysates after 48 hours of fermentation which agreed with the result that there were no yeast activity. The other two solvents tri-n-octylamine in diluted 1-octanol and alamine 336 diluted in 1-octanol exhibited less severe inhibition of yeast growth (Figure 35). Nevertheless, these two solvents did exhibited 22 and 12  —  —  30% reduction in yeast growth for alamine 336 in 1-octanol  20% reduction for tri-n-octylamine in 1-octanol. From this results, the only  suitable solvents for the project were aLamine 336 in 1-octanol and tri-n-octylamine in 1octanol.  ‘100% 80%  160% 40%  20% c.e  >.  0% Control TBP 75%  TBP 50%  TBP 25%  ALl 75%  ALl 50%  ALl 25%  ALA 75%  ALA 50%  ALA 25%  JMT 75%  JMT 50%  JMT 25%  TOA 75%  TOA 50%  Figure 35 Comparison of yeast concentration after 48 hours fermentation, co solvents: 1-octanol (light colored bars) and kerosene (dark colored bars), solvents: TBP  =  tributyl phosphate, ALl  =  aliquot 336, ALA  JMT, TOA  =  =  alamine 336, JMT  tri-n-octylamine  74  =  primene  TOA 25%  The inhibition of ethanol production was also evaluated by comparing with the control (Figure 36). Surprisingly, for both of the solvents the inhibition of biomass production was more severe than the ethanol production. This could be due to the difference in mechanism between biomass production and ethanol production for the yeast. Once the ethanol production was compared, it was clear that the best solvent for detoxifying the bio-oil hydrolysates for the project was tri-n-octylamine in 1-octanol.  100% 80% 60%  I400 .E0  20% 0% control  ALA 25% ALA 50% ALA 75% TOA 25% TOA 50% TOA 75%  Figure 36 Comparison of ethanol concentration after 48 hr fermentation (alamine 336 and tri-n-octylamine in 1-octanol)  5.5.3 Fermentation of bio-oil hydrolysate treated with solvent extraction Oleic acid and oleyl alcohol were used in this project to compare with the best solvent found from the extraction results. Their abilities to selectively extract acetic acid while retaining glucose were compared (Figure 37). As well, the practicability of these solvents was compared based on ethanol production.  75  100% .  fl80v0  c  60%  .—  40% E 20%  0% 25% TOA in 1-octanol  oleyl alcohol  oleic acid  Figure 37 Comparison of acetic acid removed (dark colored bars) and glucose retained (light colored bars) between 25% TOA with 1-octanol, oleyl alcohol and oleic acid  A comparison on the amount of glucose retained after the extractions indicated that tn-n octylamine was able to achieve 11 % higher glucose retention than oleyl alcohol and 10% higher than oleic acid. For the removal of acetic acid, tri-n-octylamine was able to remove 6% more than oleyl alcohol and 18% more than oleic acid. Similar to other solvent extraction experiments, the ethanol yields from the fermentation of the extracted artificial hydrolysates were compared with the control fermentation medium. The comparison between the solvents and the control were based on the ethanol concentration after 48 hours of fermentation.  76  100% C C  nroI ou/o  control  25% TOA in 1octa no!  oleyl alcohol  oleic acid  Figure 38 Comparison of ethanol concentration after 48 hr fermentation (tn-n octylamine, oleyl alcohol and oleic acid)  It was found that oleic acid showed a 26% inhibition towards ethanol production, and oleyl alcohol showed 14% inhibition (Figure 38). The key objective of the project was to increase the productivity of ethanol. Thus, while oleic acid showed adequate extraction results, the inhibition towards ethanol production exhibited by the solvent would limit its usefulness. The best three solvents were selected to extract actual bio-oil hydrolysate. The bio-oil hydrolysates were then diluted to 40% strength for fermentation. The ethanol yields from these detoxified bio-oil hydrolysates were compared. It was found that the three solvents had similar ethanol yields. The ethanol yield was found to be 0.24±0.01 (g ethanol/g glucose), 0.24±0.02 (g ethanol/g glucose) and 0.23±0.02 (g ethanol/g glucose) for 25 vol% tri-n-octylamine with 1-octanol, 50 vol% alamine 336 with 1-octanol and oleyl alcohol, respectively. This was not foreseeable for 25 vol% tri-n-octylamine and oleyl alcohol as their results for the retention of glucose differed by 15% and ethanol production in YPG medium differed by 14%. The reason for the slight difference in ethanol yield was unknown. 77  Although an increase in ethanol yield from 0 to 0.24 (g ethanol/g glucose) was quite significant, the improvement was only half of the overliming technique. The overliming technique generated an ethanol yield of 0.45±0.050 (g ethanol/g glucose) under the same fermentation conditions. However, the cost of the two detoxification methods should also be considered in order to determine the practicability of the two methods.  5.6 Adaptive evolution of yeast Adaptive evolution is a more passive approach than the other detoxification methods that increase the fermentability of bio-oil hydrolysates. Adaptive evolution occurs when bacteria and yeast experience a growth limiting or stressful environment. Under these conditions, they have the potential to survive and to adapt to these stresses. An example of this phenomenon is the antibiotic resistance of bacteria.  The two parameters studied in this experiment were to adapt the yeast in aerobic or micro-aerophilic conditions and to adapt the yeast in either high or low concentration of bio-oil hydrolysates. The high and low concentrations were chosen based on the results of fermenting bio-oil hydrolysate obtained from section 5.1. The high concentration medium contained 35 vol% bio-oil hydrolysate, which was the highest concentration of bio-oil hydrolysate that had shown ethanol production. The low concentration medium contained 10 vol% bio-oil hydrolysate, which was the highest concentration of bio-oil hydrolysate that exhibited an ethanol yield close to the theoretical yield.  It was observed that under high concentration of bio-oil hydrolysates, each successive transfer would make the yeast darker in color. In addition to the color change, when the yeasts were spun down, the pellets became less compact with each successive transfer. On the seventh day, all of the yeasts had perished under high concentration of bio-oil hydrolysate and micro-aerophilic conditions. This was determined when there was no observable yeast pellet when the fermentation medium was centrifuged at l0000g for 10 minutes (Sanyo, MSE micro centaur). Two days later, the same result was found for the  78  yeasts that were under high concentration of bio-oil hydrolysate and aerobic conditions. This revealed that the yeasts were able to survive longer under aerobic conditions. Nevertheless, it was a surprising result because previous experiments confirmed that the yeasts were able to ferment and produce an ethanol yield of 0.17±0.01 (g ethanol/g glucose) at 35% volume of bio-oil hydrolysates. The failure of the adaptations was probably due to the prolonged exposure to the high toxic level in the bio-oil hydrolysate.  All cultures adapted using the lower (10% v/v) concentration of bio-oil hydrolysate under micro-aerophilic conditions (samples ML 1 and 2) and aerobic (samples AL 1 and 2) conditions survived. They were transferred to agar plates and were used in the fermentation of bio-oil hydrolysates of 30% strength. The results were compared with an unadapted strain of T2 yeast and are shown in Figure 39.  140% 120% 100% 80% ‘—  40% 20% 0% control 30% hydrolysate  aerobic  microaerophilic  Figure 39 Comparison of ethanol produced after 49 hours of fermentation (yeast adapted in aerobic and low hydrolysate concentration, yeast adapted in micro aerophilic and low hydrolysate concentration)  Yeast adapted under aerobic conditions was able to produce 39.4% more ethanol than a normal strain and 15.4% more ethanol than yeasts adapted under micro-aerophilic  79  conditions. It was hypothesized that aerobic culturing of the yeast produced more biomass under the same stress conditions, which could result in higher mutation rate. In addition, yeast under micro-aerophilic condition would have slower growth rate, which would mean that these yeast adapted for less than 121 generations. A comparison of the yield of ethanol was also prepared in Figure 40.  0.40  0.30 0.20  ‘:5  0.10  0.00 control 30% hydrolysate  aerobic  microaerophilic  Figure 40 Effect of hydrolysate concentration and culture conditions on ethanol yield after 49 hours fermentation  In previous results with 30% v/v hydrolysate, the ethanol yield was 0.18±0.01 (g ethanol/g glucose). In this set of experiments, the control showed an increase in the ethanol yield when compared with previous results. Perhaps the increase in the ethanol yield could be a result of the additional hour of fermentation and random errors during the fennentation process. The improvement with the adapted yeasts was much more significant. Yeasts from sample AL had 9% higher ethanol yield than the control, while the yeast adapted at ML conditions had 6% higher ethanol yield than the control. The adapted yeast strain had the potential to increase the ethanol yield and it was likely that further adaptation could increase the ethanol yield further. In addition, this technique could be used in combination of other detoxification methods without incurring extra costs once the yeasts were adapted.  80  Chapter 6 Conclusions One of the objectives in this project was to find the optimal conditions to extract levoglucosan from bio-oil. The dominating parameter in levoglucosan extraction was the amount of water used for the extraction. It was found that the maximum amount of levoglucosan was extracted when 100 wt% of water-to-bio-oil was used for the extraction. Under the optimal conditions, 4.98 wt% of levoglucosan was extracted from the bio-oil. Other parameters such as sufficient mixing and extracting with hot water at 80°C would also be beneficial to the extraction process. Extraction with hot water had the benefit of lowering the viscosity of the organic phase of bio-oil. This could reduce the maintenance cost of the equipment.  The other objective was to reduce the toxicity of the bio-oil hydrolysate, thereby increasing the yield of ethanol. It was found that at 10% and 20% strength of bio-oil hydrolysate, the ethanol yield was comparable to the control (ethanol yield in YPG medium). However, the ethanol yields decreased as the concentration of the hydrolysates were increased. At 40% strength, the medium was too inhibitory for any ethanol production. The detoxification methods used in an attempt to remove the inhibitors included: air stripping, adsorption with activated carbon, overliming, hydrogenation, solvent extraction and coagulation and flocculation. Among these detoxification methods, overliming and solvent extraction were found be the most efficient. Bio-oil hydrolysates treated with overliming and fermented at 40% strength produced an ethanol yield of 0.45±0.05 (g ethanol/g glucose). Detoxification by solvent extractions (40% hydrolysate) produced ethanol yields of 0.24±0.01 (g ethanol/g glucose), 0.24±0.02 (g ethanol/g glucose) and 0.23±0.02 (g ethanol/g glucose) for the solvents 25 vol% tri-n-octylamine in 1-octanol, 50 vol% alamine 336 in 1-octanol and oleyl alcohol, respectively. A technique called adaptive evolution of yeast was applied to increase the resistance of yeast to inhibitors in bio-oil hydrolysates. Using high concentration of hydrolysate (35% strength), the yeast failed to adapt under both aerobic and micro-aerophilic cultivation. At low concentration of hydrolysates (10% strength), the yeast were able to adapt to the  81  hydrolysate. When subsequently used to ferment 30% v/v hydrolysates the adapted yeasts achieve an ethanol yield of 0.32±0.09 (g ethanol/g glucose) and 0.29±0.02 (g ethanol/g glucose) for yeast adapted under aerobic conditions and micro-aerophilic conditions, respectively. This was an increase of 9% and 6% compared with the controls (unadapted yeast) for yeast adapted under aerobic conditions and micro-aerophilic conditions, respectively.  82  Chapter 7 Recommendations for future work The optimal extraction ratio (water to bio-oil) of levoglucosan has been found. However, it would be beneficial to test if other solvents can be used to extract more levoglucosan from bio-oil. In addition, using dilute sulfuric acid as a solvent for extraction has the potential to combine extraction and hydrolysis in one step.  This project has showed that overliming, solvent extraction and adaptation of yeasts were able to increase the ethanol yield from the fermentation of bio-oil hydrolysate. However, a more detail analysis with additional parameters could be added to the investigation. For example, in solvent extraction, a more detailed study of the ratio of the solvents and co solvent system could be conducted to find an optimal ratio for extraction. In addition, investigation of other organic solvents such as 2-ethylhexyl alcohol, hexanol and decanol to remove fermentation inhibitors would contribute positively to this project. For adaptive evolution, it would be beneficial to see if the yeast adapt to other concentrations of hydrolysate and/or adapt better after more mutation cycles.  Modifications to the examined experimental parameters and further research can potentially improve the detoxification methods that failed to provide promising results, such as air stripping, hydrogenation, and adsorption on activated charcoal,. For example, a screening of a number of different catalysts can be performed for hydrogenation of bio oil hydrolysates to find catalysts that are able to degrade acetic acid much faster than the sugars. Finally, the monetary costs and benefits of these methods have yet to be studied and quantified. Therefore, an economic analysis on the project is highly recommended.  83  References 1. Energy Information Administration, Department of Energy (DOE), 2008, http://www.eia.doe.gov/fuelrenewahle,html (Accessed June 2008). 2. Natural Resource Canada, 2008, http://www.fuelfocus.nrcan.gc.ca, (Accessed August 2008). 3. Renewable Fuels Association (RFA), 2005 2008, http://www.ethanoIrfa.org/, (Accessed June 2008). 4. Bridgwater, A., (2004), “Biomass fast pyrolysis”, Review paper, Bio-energy research group, Aston University. 5. Zhang, L.L.H, (2004), “Preparing levoglucosan derived from waste cellulose pyrolysis”, Energy sources, vol. 26, . 1059 — 1053 pp. 6. Bennett, N., (2006), “Extraction, hydrolysis, and fermentation of levoglucosan derived from pyrolysis oil”, Master of Applied Science thesis, Department of Chemical and Biological Engineering, University of British Columbia. 7. Yu, Z., Zhang, H., (2004), “Ethanol fermentation of acid-hydrolyzed cellulosic pyrolysate with Saccharomyces cerevisiae”, Bioresource technology vol. 93, pp. 199—204. 8. “AgriSuccess Journal”, Farm Credit Canada, September/October 2007. 9. Galbe, M., Zacchi, G., (2002), “A review of the production of ethanol from softwood”, Applied microbiology and biotechnology, vol. 59, pp. 618—628. Canadian 10. The Renewable Fuels Association (CRFA), 2008, http://www.grcenfuels.org/cdnstats.php, (Accessed December 2008). 11. Lagercrantz, J., (2006), “Ethanol production from sugar cane in Brazil”, Report for GrOna Bilister, report 1.06. 12. Roberts, J.D., Caserio, M.C., (1977), “Basic principles of organic chemistry”, W.A Benjamin Inc. 13. Millidge, A.F., (1969), “Ethylene and its industrial derivatives”, S.A. Miller (Ed.), Ernest Benn Ltd., London, pp. 709. 14. Fougret, C.M., Holderich, W.F., (2001), “Ethylene hydration over metal phosphates impregnated with phosphoric acid”, Applied catalysis A, vol. 207, pp. 295—301. 15. Kargi., S., Shuler., M., (2006), “Bioprocess engineering”, Dorling Kindersley Pvt. Ltd. 16. Lodgsdon, J.E., (1994), “Ethanol.” In J.1. Kroschwitz (Ed.) Encyclopedia of chemical technology, 4th ed. vol. 9, pp. 817. New York: John Wiley & Sons. 17. Pimentel, D. and Patzek T., (2005), “Ethanol production using corn, switchgrass, and wood; biodiesel production using soybean and sunflower”, Natural resources research, vol. 14, no. 1, pp. 65 76. 18. Pimentel, D. (2003), “Ethanol fuels: energy balance, economics and environmental impacts are negative”, Natural resources research, vol. 12, no. 2, pp. 127— 134. 19. Hodge, C., (2002), “Ethanol use in US gasoline should be banned, not expanded” Oil and gas journal, vol. 100, no. 37, pp. 20—30. —  —  84  20. Sun, Y, Cheng, J., (2002), ‘Hydrolysis of lignocellulosic materials for ethanol production: a review” Bioresource technology, vol. 83, pp. 1 11. 21. Chum, H.L., Johnsoon, D.K., Black, S., (1988), “Organosolv pretreatment for enzymatic hydrolysis of poplars: 1. enzyme hydrolysis of cellulosic residues” Biotechnology and bioengineering, vol. 31, pp. 643—649. 22. Thring, R.W., Chorent, E., Overend, R., (1990), “Recovery of a solvolytic lignin: effects of spent liquor/acid volume ration, acid concentration and temperature” Biomass, vol. 23, pp. 289—305. 23. Foody, B. E., Foody, K. J. (1991), “Development of an integrated system for producing ethanol from biomass.” Energy biomass wastes, vol 14, pp. 1225-1243. 24. McMillan, J.D., (1994), “Pretreatment of lignocellulosic biomass- enzymatic conversion of biomass for fuels production”, American Chemical Society, Washington, DC, pp. 292—324. 25. Grous, W.R., Converse, A.O., Grethlein, H.E., (1986), “Effect of steam explosion pretreatment on pore size and enzymatic hydrolysis of poplar”, Enzyme microbial technology, vol. 8, pp. 274—280. 26. Mononmani, H.K., Sreekantiah, K.R., (1987), “Saccharification of sugar-cane bagasse with enzymes from Aspergillus ustus and Trichoderma viride”, Enzyme microbial technology, vol. 9, pp. 484—488. 27. Yang, R., Lucia, L., Jameel, H., Ragauskas, A.J., (2003), “Oxygen delignification chemistry and its impact on pulp fibers”, Journals of wood chemistry and technology, vol. 23, no. 1, pp. 13-29. 28. Hakalaa, T.K., Lundella, T., Galkina, S., Maijalaa, P., Kalkkinenb, N., Hatakkaa, A., (2005), “Manganese peroxidases, laccases and oxalic acid from the selective white rot fungus Physisporinus rivulosus grown on spruce wood chips”, Enzyme microbial technology, vol. 36, pp. 461 —468. 29. Bridgwater, A.V., (1999), “Principles and practice of biomass fast pyrolysis process for liquids”, Journal of analytical and applied pyrolysis, vol. 51, pp. 3-22. 30. Oasmaa, A., Peacocke, C., (2001), “A guide to physical property characterization of biomass derived fast pyrolysis liquid”, Technical research center of Finland. 31. Diebold, J., (2000), “A review of the chemical and physical mechanisms of the storage stability of fast pyrolysis bio-oils”, National renewable energy laboratory. 32. Bridgwater, A., (2005), “Fast pyrolysis based biorefineries”, Bio-energy research group, Aston University, Birmingham, UK 33. Peacocke, G., Russell, P., Jenkins, J., Bridegwater., A., (1994), “Physical properties of flash pyrolysis liquids.”, Biomass and bioenergy, vol. 7, no. 16, pp. 169—177. 34. Oasmaa, A. and Sipila, K. (1996), “Pyrolysis oil properties: use of pyrolysis oil as fuel in medium speed diesel engines”, Bio-oil production and utilization, CPL scientific information services, Ltd, Newbury, UK, pp. 175-185. 35. Czernik, S., (1994), “Storage of biomass pyrolysis oils.” National Renewable Energy Laboratory, Golden, CO, pp. 67-76. 36. Diebold, J.P., Czernik, S. (1997), “Additives to lower and stabilize the viscosity of pyrolysis oils during storage,” Energy and fuels, vol. 11, pp. 108 1-1091. —  85  37. Nguyen, Q. A., Saddler, J. N. (1991), “An integrated model for the technical and economic evaluation of an enzymatic biomass conversion process.” Bioresource technology, vol 35, pp 275-282. 38. Helle, S.S., Bennett, N.M., Lau, K., Matsui, J.H., Duff, S.J.B., (2007), “A kinetic model for production of glucose by hydrolysis of levoglucosan and cellobiosan from pyrolysis oil”, Carbohydrate research, vol. 342, pp. 2365-2370. 39. Sandvig, E., Walling, G., Brown, R., Pletka, R., Johnson, W., (2003), “Integrated pyrolysis combined cycle biomass power system” Alliant energy, center for sustainable environmental technologies Iowa State University. DynaMotive Energy Systems Corp, Black & Veatch Corporation. 40. Mussatto, S.I., Roberto, I.C., (2004), “Alternatives for detoxification of dilutedacid lignocellulosic hydrolyzates for use in fermentative processes: a review”, Bioresource technology, vol. 93, pp. 1—10. 41. Lambert, R. 0., Moore-Bulls, M. R., Barrier, J. W. (1990), “An evaluation of two acid hydrolysis processes for the conversion of cellulosic feedstocks to ethanol and other chemicals.”, Applied biochemistry and biotechnology, vol. 24/25, pp. 773-783. 42. Lee, Y.Y., Praschant, I., Torget, R.W., (2002), “Dilute acid hydrolysis of lignocellulosic biomass”, Department of chemical engineering, Auburn University and National Renewable Energy Laboratory, US. 43. Sun, Y., Cheng, J., (2002), “Hydrolysis of lignocellulosic materials for ethanol production: A review”, Bioresource technology, vol. 83, pp. 1—11. 44. Rodrigues, F., Ludovico, P., Leao, C., (2006), “Sugar metabolism in yeasts: an overview of aerobic and anaerobic glucose catabolism”, The yeast handbook, vol. 6, pp. 101 123, life and health sciences research institute, health sciences school, University of Minho. 45. Lin Y., Tanaka 5., (2006), “Ethanol fermentation from biomass resources: current state and prospects”, Applied microbiology and biotechnology, vol. 69,pp. 627— 642 46. Ingram, L.O., Conway, T., Clark, D.P., Sewell, G.W., Preston, J.F., (1987) “Genetic engineering of ethanol production in Escherichia coil” Applied environmental microbiology, vol. 53, pp. 2420—2425 47. Gong, C.S., Maun, C.M., Tsao, G.T., (198 1),”Direct fermentation of cellulose to ethanol by a cellulolytic filamentous fungus Monilia sp”, Biotechnology letter vol. 3, pp. 82 — 77 48. Gervais, P., Sarrette, M., (1990), “Influence of age of mycelia and water activity on aroma production by Trichoderma viride”, Journal of fermentation and bioengineering, vol. 69, . 50 — 46 pp. 49. Saddler, J.N., Chan, M.K.H., (1982), “Optimization of Clostridium thermocelium growth on cellulose and pretreated wood substrates.”, Applied microbiology and biotechnology, vol. 16, pp. 104 — 99 50. McMurry, J., (2004), “Organic chemistry”, Thomson learning Ltd. 51. Bakker, B.M., (2001), “Stoichiometry and compartmentation of NADH metabolism in Saccharomyces cerevisiae”, FEMS microbiology reviews, vol. 25, pp. 15-37. —  86  52. Helle, S.S., Duff, S.J.B., (2004) “Supplementing spent sulfite liquor with a lignocellulosic hydrolysate to increase pentose / hexose co-fermentation efficiency and ethanol yield”, final report, Department of chemical and biological engineering, University of British Columbia. 53. Deak, T., (2006), “Environmental factors influencing yeasts”, The yeast handbook, vol. 8, pp 155 175, Department of microbiology and biotechnology, Corvinus University. 54. Larsson, S., Palmqvist, E., Hahn-Hagerdal, B., Tengborg, C., Stenberg, K., Zacchi, G., Nilvebrant, N.O., (1999), “The generation of fermentation inhibitors during dilute acid hydrolysis of softwood”, Enzyme microbial technology, vol. 24, pp. 151 159 55. Nigam, J.N., (2001), “Ethanol production from wheat straw hemicellulose hydrolysate by Pichia stipitis”, Journal of biotechnology, vol. 87, pp. 17-27. 56. Helle, S.S., Lin, T., Duff, S.J.B., (2008), “Optimization of spent sulfite liquor fermentation” Enzyme and microbial technology, vol. 42, pp. 259-264. 57. Mussatto, S.I., (2002), “Influence of treatment of straw hemicellulosic hydrolyzate in Rice Production of Xylitol by Candida guilliermondii Faculty of chemical engineering, Lorena, Brazil. 58. Parajo, J.C., Dominguez, H., Dominguez, J.M., (1995), “Study of charcoal adsorption for improving the production of xylitol from wood hydrolysates.” Bioprocess engineering, vol. 16, pp. 39—43. 59. Sene, L., Converti, A., Zilli, M., Felipe, M.G.A., Silva, S.S., (2001), “Metabolic study of the adaptation of the yeast Candida guilliermondii to sugar cane bagasse hydrolysate.” Applied microbiology and biotechnology, vol. 57, pp. 738—743. 60. Martinez, A., Rodriguez, M.E., Wells, M.L., York, S.W., Preston, J.F., Ingram, L.O., (2001), “Detoxification of dilute acid hydrolysates of lignocellulose with lime.” Biotechnology progress, vol. 17, pp. 287—293. 61. Yu, Z., Zhang, H., (2003), “Pretreatment of cellulose pyrolysate for ethanol production by Saccharomyces cerevisiae, Rich Ia sp. YZ- 1 and Zymomonas mobilis”, Biomass and Bioenergy, vol. 24, pp. 257-262. 62. Horvath, I.S., Alriksson, A.S.B., Jonsson, L.J., Nilvebrant, N.O, (2005), “Critical conditions for improved fermentability during overliming of acid hydrolysates from spruce”, Applied biochemistry and biotechnology, pp. 121 124. 63. Senol, A, (2004), “Effect of diluent on amine extraction of acetic acid: modeling considerations”, Industrial engineering chemistry research, vol. 43, pp. 64966506. 64. Yang, S. T., White, S. A., Hsu, S. T., (1991), “Extraction of Carboxylic Acids with Tertiary and Quaternary Amines: Effect of pH.”, Industrial engineering chemistry research, vol. 30, pp. 1335-1342. 65. Zausten, R.R.M., Maugeri-Filho, F., Vaz-Rossell, C.E., Straathof, A.J.J, (2008), “Liquid-liquid extraction of fermentation inhibiting compounds in lignocellulose hydrolysate”, Biotechnology and bioengineering, vol. 102, no. 5, pp. 1354 1360. 66. Fisk, C., Crocker, M., Store), J., Crofcheck, C., Andrews, R., Lewis, S., (2006). “Novel approach to catalytic upgrading of hio-oil”. American society of agricultural and biological engineers, paper no. 066035. pp. 1 10. —  —  —  —  —  87  67. Furmisky, E, (20()0), “Catalytic hydrodeoxygenation: review”, Applied catalysis vol. 199, pp. 147—190. 68. Zhang, S., Yan, Y., Ren, Z., Li, T., (2003), “Study of hydrodeoxygenation of hio oil from fast pyroiysis of biornass”, Energy sources, vol. 25, pp. 57 65. 69. Ryashentseva, M.A., Minachev, K.M., Belanova, E.P., (1974), “Hydrogenation of acetic acid over rhenium sulfides”, Russian chemical bulletin, vol. 23, no. 8, pp. 1833-183 5. 70. McBryde, C., Gardner, J., Lopes, M., Jiranek, V., (2006), “Generation of novel wine yeast strains by adaptive evolution”, American society for enology and viticulture, vol. 57. no. 4, pp. 423 430. 71. Ferea, I., Botstein, D., Brown, P., Rosenzweig, F., (1999). ‘Systematic changes in gene expression patterns following adaptive evolution in yeast”, Proceedings of national academy of sciences, USA, vol. 96, pp. 972 1—9726. 72. Lombardo, M., Aponyi, 1., Rosenberg, S., (2004), “General stress response regulator RpoS in adaptive mutation and amplification in Escherichia co/i”, Genetics society of america vol. 166, pp. 669 -680. 73. Marechal., P.A., de Maranon., I.M., Molin., P., Gervais., P., (1995), “Yeast responses to water potential variations”, International journal of food microbiology, vol. 28, pp. 277—287. 74. Alamine® 336 Fact Sheet, Cognis Corporation Mining Chemicals, http://www.products.cognis.com/cogn:is/prodleafR2.nsf/($Productsbyl)ocIDPL Header)/REFEC0F524763F20 I 22C I 256CAA002DD839/$file/Alaminer336E. p, (Accessed on 2009) 75. Aliquat® 336 Fact Sheet, Cognis Corporation Mining Chemicals, http://www.products.cognis.com/cognis/prod1eafR2.nsf/($Productsby[)ocl [)PL Header)/REFABE5FB30096BF55485256D2EOO5CC9AE/$file/Aliguat r 336 E. pf, (Accessed on 2009) 76. Alguacil, F.J., Caravaca, C., (1996), “Synergistic extraction of gold (I) cyanide with the primary amine Primene JMT and the phosphine oxide Cyanex 921”, Hydrometallurgy vol. 42, pp. 197-208. 77. Scholzer, B., Meier, D., (2001), “Characterization of the water-insoluble fraction from pyrolysis oil (pyrolytic lignin). Part I. PY-GC/MS, FTIR, and functional groups”, Journal of analytical and applied pyrolysis, vol. 60, pp. 4 1-54. 78. Sipilar, K., Kuoppala, E., Farernas, L., Oasmaa, A., (1998), “Charaterization of biomass based flash pyrolysis oil”, Biomass and bioenergy, vol. 14, no. 2, pp. 103-113. 79. Van Gorp, K., Boerman, E., Cavenaghi, C.V., Berben, P.H., (1999), “Catalytic hydrogenation of fine chemicals: sorbitol production.” Catalysis today, vol. 52, pp. 349—361. —  —  88  Appendix Table 18 List of identified compounds in the aqueous phase for pine wood derived bio-oil [46] Compounds 1  2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23  Compounds  acetic acid 1-hydroxy-2-propanone 1-hydroxy-2-butanone 2-cyclopenten-1 -one 2-furaldehyde 2-methyl-2,3-dihydrofuran-3-one 1 -acetyloxypropan-2-one 2-methyl-2,3-dihydrofuran-3-one acetylfuran 2-methyl-2-cyclopenten-1 -one isomer of a-angelica lactone 4-(hydroxy-methyl)tetrahyd ropyran-3-one 5-methyl-2-furaldehyde phenol 2-hydroxy-3-methyl-2-cyclopenten-1 -one 2-methylphenol 3- or 4-methylphenol guaiacol 3-hydroxy-2-methyl-(4H)-pyran-4-one 2,4- or 2,5-dimethyiphenol ethylphenol catechol 4-methylquaiacol  89  24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45  5-(hydroxymethyl)-2-fural-dehyde 3-methylcatechol 3-methoxycatechol 4-methylcatechol 4-vinylguaiacol dimethylpyranone dimethylcatechol eugenol 4-ethylcatechol van illin isoeugenol 4-propylguaiacol acetoquaiacone coniferyl alcohol (cis) vanillic acid con-iferyl alcohol (trans) a-oxypropioquaiacone dihydroconiferyl alcohol acetosyringone coniferaldehyde syringyl acetone sinapaldehyde  

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.24.1-0058763/manifest

Comment

Related Items