The Pennsylvania State University. The Graduate School. College of Agricultural Sciences ECONOMICS OF BIODIESEL PRODUCTION FROM MICROALGAE

Transcription

1 The Pennsylvania State University The Graduate School College of Agricultural Sciences ECONOMICS OF BIODIESEL PRODUCTION FROM MICROALGAE A Thesis in Agricultural, Environmental and Regional Economics by Narishwar Ghimire 2008 Narishwar Ghimire Submitted in Partial Fulfillment of the Requirements for the Degree of Master of Science August 2008

2 ii The thesis of Narishwar Ghimire was reviewed and approved* by the following: David G. Abler Professor of Agricultural, Environmental & Regional Economics and Demography Thesis Adviser James Shortle Distinguished Professor of Agricultural and Environmental Economics Director, Environment and Natural Resources Institute James Dunn Professor of Agricultural Economics Stephen M. Smith Professor of Agricultural and Regional Economics Head of the Department of Agricultural Economics and Rural Sociology *Signatures are on file in the Graduate School.

3 iii ABSTRACT The study examines the economic potential of an algal biodiesel industry in the US using a simulation-based partial equilibrium modeling approach. Different scenarios are constructed under various assumptions, and the price, cost, and demand and supply structure of the algal biodiesel industry are analyzed. The results show that production of algal biodiesel would not be sufficient to meet the transport sector diesel demand in the US until 2029/30 without a very rapid rate of technical change (30 percent annually). Even with a 30 percent annual rate of technical change, meeting the entire transport sector diesel demand by algal biodiesel would be a formidable task without a continuation of government subsidies for biodiesel. Model results indicate that the current subsidy of about $42 per barrel of oil equivalent increases the production of algal biodiesel up to 80 percent relative to a scenario in which the subsidy is eliminated. Model results also indicate that the competitiveness of algal biodiesel with petro-diesel is hampered by the current high marginal cost of algal biodiesel. The average total cost of algal biodiesel was found by 2030 to range between $41 and $50 per barrel for price taker model and $31 to $70 per barrel for price maker model given an annual rate of Hicks- Neutral technical change of 30 percent.

4 iv TABLE OF CONTENTS LIST OF FIGURES LIST OF TABLES ACRONYMS UNITS ACKNOWLEDGEMENTS VII X XII XIII XIV CHAPTER 1. INTRODUCTION OBJECTIVES OF THE RESEARCH SIGNIFICANCE OF THE STUDY 5 CHAPTER 2. LITERATURE REVIEW IMPORTANCE OF ALGAL BIODIESEL IN THE US PRODUCTION COSTS OF ALGAL BIODIESEL MICROALGAE SPECIES AND OIL CONTENT PRODUCTION PROCESSES FOR MICROALGAE IMPORTANT ISSUES IN THE PRODUCTION OF MICROALGAE OVERVIEW OF WORLD PETROLEUM OIL DEMAND AND SUPPLY OVERVIEW OF DIESEL AND BIODIESEL DEMAND IN THE US 19 CHAPTER 3. METHODOLOGY PRICE TAKER MODEL Cost function 22

5 v Supply function Demand function for the price taker model PRICE MAKER MODEL Demand function Supply function 28 CHAPTER 4. RESULTS AND DISCUSSION COST STRUCTURE Scenario I Scenario II Scenario III DEMAND AND SUPPLY FOR PRICE TAKER MODEL Demand and supply of algal biodiesel in scenario I Demand and supply of algal biodiesel in scenario II Demand and supply of algal biodiesel in scenario III Impact of returns to scale parameter (γ) on the supply of algal biodiesel PRICE MAKER MODEL Demand and supply of algal biodiesel in scenario I Demand and supply of algal biodiesel in scenario II Demand and supply of algal biodiesel in scenario III IMPACT OF GOVERNMENT POLICY ON THE MARKET FOR ALGAL BIODIESEL Baseline subsidy scenario Eliminated subsidy scenario Doubled subsidy scenario 64

6 vi CHAPTER 5. CONCLUSIONS AND POLICY IMPLICATIONS CONCLUSIONS LIMITATIONS OF THE STUDY AND POLICY IMPLICATIONS 70 BIBLIOGRAPHY 71 APPENDIX A: ELASTICITY VALUES FOR DIESEL AND OTHER PETROLEUM PRODUCTS 78 APPENDIX B: EQUILIBRIUM PRICE AND QUANTITIES OF ALGAL BIODIESEL UNDER DIFFERENT SUBSIDY SCENARIOS 81

7 vii LIST OF FIGURES Fig Production of Biodiesel in US 8 Fig. 2.2a. Open raceway system of algae production 13 Fig. 2.2b. Open raceway system of algae production (photo) 14 Fig. 2.3 Algae production flow chart 15 Fig World projected oil demand 18 Fig World projected oil supply under different scenarios 19 Fig Projected diesel and biodiesel demand in transport in US 20 Fig Projected share of petro-diesel in transport fuel consumption in US 20 Fig Total cost curves 2015/17 scenario I (r =.20) 31 Fig MC/AC curves 2010/12 scenario I (r =.20) 32 Fig Total cost curves 2010/12 scenario I (r =.30) 32 Fig AC and MC curves 2010/12 scenario I (r =.30) 33 Fig Trend of AC and MC for price maker model scenario I (r =.30) 33 Fig Total cost curves 2015/17 scenario II (r =.30) 35 Fig AC and MC curves 2020/22 scenario I (r =.30) 36 Fig Trend of AC and MC scenario III 36 Fig Trend of AC under three scenarios 37 Fig Equilibrium price and quantities 2008/14 in scenario I (r =.30) 40 Fig Equilibrium price and quantity 2015/20 in scenario I (r =.30) 41 Fig Equilibrium price and quantity 2020/25 scenario I (r =.30) 41 Fig Equilibrium prices and quantities 2025/30 in scenario I (r =.30) 42

8 viii Fig Equilibrium prices and quantities 2008/14 in scenario II (r =.30) 44 Fig Equilibrium price and quantities 2015/20 scenario II (r =.30) 45 Fig Equilibrium price and quantities 2020/25 in scenario II (r =.30) 45 Fig Equilibrium price and quantities 2025/30 scenario II (r =.30) 46 Fig Equilibrium prices and quantities 2008/14 in scenario III (r =.30) 48 Fig Equilibrium prices and quantities 2015/20 in scenario III (r =.30) 48 Fig Equilibrium prices and quantities 2021/25 in scenario III (r =.30) 49 Fig Equilibrium prices and quantities 2025/30 in scenario III (r =.30) 49 Fig Trend of algal biodiesel supply under different values of γ (r =.25) 50 Fig Equilibrium prices and quantities of algal biodiesel in scenario I (r =.20) 52 Fig Equilibrium prices and quantities of algal biodiesel in scenario I (r = 0.30) 54 Fig Trend of price of algal biodiesel under scenario I 55 Fig Equilibrium price and quantity of algal biodiesel for 2029 in scenario II (r =.30) 57 Fig Equilibrium price and quantity of algal biodiesel for 2030 in scenario II (r =.30) 57 Fig Trend of price of algal biodiesel under scenario II 58 Fig Comparison of equilibrium prices and quantities under baseline and subsidy elimination scenarios for 2030 (r =.30, γ = 2.5) 61 Fig Comparison of equilibrium prices and quantities under baseline and subsidy elimination scenarios for 2028 (r =.30, γ = 2.5) 62 Fig Price trend under baseline and eliminated subsidy scenarios (γ = 2.5, r =.30) 63

9 ix Fig Quantity comparison between baseline and eliminated subsidy scenario (γ = 2.5) 63 Fig Trend of percent increase in quantity of algal biodiesel given a doubling of the subsidy (γ = 2.5) 64 Fig Comparison of quantities and prices under baseline and doubled subsidy scenarios for 2030 (γ = 2.5, r =.20) 65 Fig Equilibrium quantity and prices of algal biodiesel under three different scenarios for 2028 (γ = 2.5, r =.30) 66 Fig Equilibrium quantity and prices of algal biodiesel under three different scenarios for 2029 (γ = 2.5, r =.30) 66 Fig Equilibrium quantity and prices of algal biodiesel under three different scenarios for 2030 (γ = 2.5, r =.30) 67

10 x LIST OF TABLES Table 1.1. Land requirements to replace US transport sector petro-diesel demand 7 Table 2.1. Estimated costs for biodiesel production from microalgae 10 Table 2.2. Oil content of some selected species of microalgae for biodiesel 12 Table 4.1. TC, MC and AC scenario I 31 Table 4.2. TC, AC and MC scenario I for price maker model 34 Table 4.3. TC, MC and AC scenario II 35 Table 4.4. TC, MC and AC scenario III 37 Table 4.5. Equilibrium quantities of algal biodiesel in scenario I (φ =.80) 39 Table 4.6. Equilibrium quantities of algal biodiesel in scenario II 43 Table 4.7. Equilibrium quantities of algal biodiesel in scenario III 47 Table 4.8. Equilibrium price and quantities of algal biodiesel in scenario I 53 Table 4.9. Equilibrium price and quantities of algal biodiesel scenario II 56 Table Equilibrium price and quantity of algal biodiesel in scenario III 59 Table A.1 Price elasticities of petro-fuel demand 78 Table A.2 Income elasticities of petro-fuel demand 79 Table A.3 Elasticity of lagged petro-fuel demand 79 Table A.4 Price elasticity of supply, lagged supply and coefficient of time in supply of petro-fuel 80 Table B.1. Prices and quantities under eliminated subsidy scenario (γ = 2.5) 81 Table B.2. Prices and quantities under eliminated subsidy scenario (γ = 3.0) 82 Table B.3. Prices and quantities under eliminated subsidy scenario (γ = 3.5) 83

11 xi Table B.4. Prices and quantities under doubled subsidy scenario (γ = 2.5) 84 Table B.5. Prices and quantities under doubled subsidy scenario (γ = 3.0) 85 Table B.6. Prices and quantities under doubled subsidy scenario (γ = 3.5) 86

12 xii ACRONYMS $: US dollar AC: Average cost bbl: Barrel CO: Carbon monoxide CO 2 : Carbon dioxide CPI: Consumer price index EU: European Union FAO: Food and Agriculture Organization FAPRI: Food and Agricultural Policy Research Institute GHG: Greenhouse gas mb/d: Million barrels per day MC: Marginal cost NOx: Nitrous oxides NREL: National Renewable Energy Laboratory PABD: Price of algal biodiesel PBD: Price of biodiesel PPD: Price of petro-diesel TC: Total cost US: United States of America

13 xiii UNITS 1 BARREL = 42 GALLONS 1 MILLION BARREL = 10 6 BARRELS 1 BILLION GALLON = 10 9 GALLONS

14 xiv ACKNOWLEDGEMENTS I would like to gratefully acknowledge Prof. David Abler in the Department of Agricultural Economics and Rural Sociology, The Pennsylvania State University for his sincere cooperation and inspiration throughout my study here at Penn State University. Without his incessant encouragement, comments and invaluable suggestions it would not have been possible to bring the thesis in the present form. So I am greatly indebted to him and would be extending my sincere and deepest gratitude towards him. I would also like to express my gratitude to Prof. James Shortle and Prof. James Dunn in the Department of Agricultural Economics and Rural Sociology for their invaluable comments, corrections and moral support for the thesis and study. Moreover, I am grateful to Prof. Stephen M. Smith, Head of the Department of Agricultural Economics and Rural Sociology for moral support for the study.

15 xv DEDICATION I wish to dedicate this thesis to my respected late parents Mr. Nanda Ram Ghimire and Mrs. Deel Kumari Ghimire

16 Chapter 1 INTRODUCTION Widespread consumption of fossil fuels such as petroleum products has exacerbated problems of global warming and climate change due to increased emissions of carbon dioxide and carbon monoxide (CO) into the atmosphere. Even though fossil fuels currently contribute about more than 85 percent of the world s energy supply, their viability for the future has been seriously questioned from environmental and economic perspectives (Herzog, 1996). Economically, recent significant increases in energy prices have spurred interest in alternative energy sources. In the search for renewable energy sources, biodiesel has been presented as an alternative to petroleum products. Biodiesel consists of esters of fatty acid components of triglyceride derived from oils and fats in the presence of catalyst through a transesterification process (Brown et al., 1994; Duncan, 2003; ERC, 2006; Green-Trust, 2008). Although biodiesel produces more nitrous oxides (NOx) than petro-diesel, it performs nearly as well as petroleum products in motor engines and reduces GHG emissions (Sheehan et al., 1998). The production of biofuels from higher plants like soybeans, sugarcane, corn, wheat, beet and rapeseed has, however, been limited because of the large land mass required for their cultivation and their adverse impacts on food prices. For instance, it has been estimated that the US, Canada and EU would require percent of their cropland if they were to replace 10 percent of their transport fuel by biofuels, assuming constant technology and no

17 2 international trade in biofuels (OECD, 2006). As a result, an alternative source of biodiesel is needed. Biodiesel from microalgae 1 has been proposed as one of the most viable options to fill this gap. It has been hypothesized that biodiesel from algae could completely replace petro-diesel due to algae s high oil content, smaller land requirements for production and considerably shorter harvesting cycle compared to higher plants. In the US only 10 million acres of land (about 1% of the total land used for farming and grazing in US) would be sufficient to produce enough algae to fulfill the country s entire demand for diesel (Castoronline, 2006). Many species of microalgae are rich in lipid content (more than 50% of their mass). They can better capture the sunlight for photosynthesis than terrestrial plants and grow very rapidly, thereby making quicker harvests (almost daily) possible (Danielo, 2005; GreenFuel Technologies Press Release, 2007). There are claims that microscopic algae can synthesize 30 times more oil per hectare compared to terrestrial plants because of their unicellular nature and production in liquid media, and that they could completely replace the petroleum consumed in the US if cultivated on a surface area of km 2 or 4.5 million acres of land (about the size of Maryland) in high sun exposure locations (Danielo, 2005; Green Chip Stocks, 2008). Until now microalgae have been utilized for treatment of municipal and other waste water in the United States (Benemann et al., 2003). Marlborough-based Aquaflow Bionomic Corporation in New Zealand was the first company to produce biodiesel from algae on a commercial scale outside the laboratory (Aquaflow Bionomic Corporation, 2006). Algae can be grown every place where there is an abundance of sunlight and its production can be beneficial in reducing carbon dioxide and NOx emissions. In the 1 Algal strains, diatoms, and cyanobacteria are collectively called microalgae.

18 3 presence of sunlight algae consumes CO 2 and multiplies, and after harvest can be converted into biodiesel. More than 3,000 strains of microalgae have been studied and some of them have shown large potential for lipid production (Brown and Sprague, 1992). Microalgae are highly prolific producers of biomass and can be grown in continuous culture (Terry and Raymond, 1985). Algal technology not only helps to reduce GHG emissions but, if production is located near fossil fuel-fired power plants, can use recycled waste CO 2. Biodiesel reduces GHG emissions by 41 percent compared to 12 percent by ethanol and provides 93 percent more energy than the energy invested in its production whereas ethanol produces only 25 percent more energy than the actual energy invested in its production (Hill et al., 2006). Biodiesel produced from terrestrial plants and animal fat is insufficient to meet the growing worldwide demand for diesel. For instance, total vegetable oil and tallow supply in the US is equivalent to 3 billion gallons of biodiesel fuel (Brown et al., 1994). Other renewable energy sources like hydrogen fuel, wind power, solar power, hydropower, and geothermal power are at present not widely used in the US. For instance, hydroelectric contributes only seven percent of the total electricity use in the country, and solar power and geothermal energy each contribute less than one percent of the total energy needs of the US (Nationalatlas.gov, 2008). Nuclear energy is currently contributing 20 percent of the electricity generated in the US but nuclear power plant construction is expensive. The most viable alternative so far envisioned might be biodiesel from microalgae. But the question is whether biodiesel production from microalgae is economically viable as a substitute for petro-diesel. Benemann and Oswald (1996) reported production

19 4 costs of about $10,000/ton of algal biomass which in their own words was twice as expensive as alternative GHG mitigation techniques. Dimitrov (2007) estimated the average cost of algal biodiesel of about $20.31 per gallon, even under generous assumptions about the lipid content of microalgae and a conservative estimate of capital costs. Whether microalgae production technology can improve to the point where production costs are competitive with petro-diesel is an open question Objectives of the research The goal of this thesis is to analyze the economics of algal biodiesel production in the United States. The specific objectives of this study are to answer the following questions: o What rates of technical change in algal biodiesel production would be necessary for significant commercial production of algal biodiesel? o What will be the demand and supply structure of algal biodiesel in the US in coming years under different assumptions about technical change? o What will be the likely equilibrium price of algal biodiesel under different assumptions about technical change? Would it be competitive with petro-diesel? o What will be the impact of government biofuels subsidies on the market for algal biodiesel in the US?

20 Significance of the study As noted above, biodiesel from microalgae has been envisaged as a substitute for a substantial proportion of diesel fuel demand in the US if technologies can be developed to fully utilize its potential. There are dozens of studies to support this argument. Putt (2007) reported that growing microalgae on only 3 percent of Alabama land (1 million acre ponds) would produce 3 billion gallons per year of transportation fuel, sufficient to meet state demands. Production of algae on about 9.5 million acres of land would be sufficient to replace billion gallons of diesel fuel used in US annually (Briggs, 2004). The US National Renewable Energy Laboratory (NREL) started its work on production of biodiesel from microalgae during the energy crisis of the 1970s and discovered that growing native species of 30g algae /m2/day with at least 30 percent lipid content yielding 4,000 gallons of biodiesel per acre annually would be a capital-cost effective approach for biodiesel production in US (Putt, 2007). However, the work of NREL on algal biodiesel was terminated during the low energy prices of the 1990s. Increases in global energy prices and demand during this decade have revived interest in algal biodiesel. To date, studies of the potential competitiveness of algal biodiesel in the changing context of global energy markets have not been conducted. Some plant-level economic analysis has been done but it is inadequate to provide a detailed picture of future demand, supply and prices of algal biodiesel. This study is intended to be a contribution to the literature on formulating an appropriate strategy for addressing the potential of microalgae for the production of algal biodiesel.

21 6 Chapter 2 LITERATURE REVIEW 2.1 Importance of algal biodiesel in the US It has been shown that biodiesel has the potential to substitute for some proportion of petroleum fuel used in the transport sector (BMELV, 2006). Chisti (2007) and Briggs (2004) go so far to argue that microalgae can completely replace petro-diesel if the technology is developed properly. Assuming the current total cropland of US to be 450 million acres (Briggs, 2004) and bearing in mind that vegetable oil on average has an 80 percent biodiesel recovery, canola (rapeseed) would require 100 percent of the US cropland to replace the entire petro-diesel demand in the transportation sector ( million barrel/year) while microalgae would require only 2.5 percent of total US cropland (assuming a biodiesel yield of 4,000 gallon/acre from microalgae). If we increase the oil content of microalgae through genetic engineering or other scientific techniques the land requirement for biodiesel production would be significantly reduced. Hence, microalgae seem to a potential source of biodiesel in the US (Table 1.1). However, high production costs are hindering the expansion of algal biodiesel. Technology plays an essential role in reducing production costs, and is required in order to make the algal biodiesel industry a profitable investment opportunity. This thesis examines the role of Hicks-neutral technical change in reducing production costs and thereby expanding biodiesel production.

22 7 Table 1.1. Land requirements to replace US transport sector petro-diesel demand Crops Oil Biodiesel yield Land required Percent of US yield* (gal/acre) (million acre) cropland (gal/acre) Corn (maize) , Cotton , Soybean , Linseed (flax) , Pumpkin seed Mustard seed Canola Coconut Oil palm Chinese tallow Algae (moderate yield)** 5, , Algae (high yield) 50, , *Source: Journey to forever, 2008 **Source: PESWiki, 2008; Approximate value from Chisti, 2007 An important feature of many biodiesels is their environmental friendliness, thereby enhancing sustainability and reducing GHG emissions (Demirbas, 2007). Thus, a study of pricing and the cost structure as well as the demand and supply of algal biodiesel is very important as it helps policy makers to design sound policy in this sector. Biodiesel production is increasing every year due to high petro-diesel prices and government biofuel mandates. Use of biodiesel helps to reduce the harmful carbon monoxide from diesel engines as well as unburned hydrocarbons and particulate matter (NBB, 2008). Use of biodiesel reduces CO 2 emissions by 78 percent and polycyclic aromatic hydrocarbons (PAH) by about 75 to 78 percent compared to petro-diesel.

23 8 Despite this fact a close scrutiny of biodiesel production from oil seeds and fats indicates that their production will eventually decline due to competition for resources such as land with food crops and increases prices of biofeedstocks (Raneses et al., 1999; Coltrain, 2002; Collins, 2006; FAPRI, 2007). Although the US government has targeted to increase biodiesel production from 500 million gallons in 2009 to one billion gallons in 2012, the rise in biofeedstock prices will make the achievement of this goal increasingly difficult unless steps are taken to produce biodiesel from other sources such as algae which do not compete with food crops for land (Fig. 2.1). Fig Production of Biodiesel in US Total biodiesel Soya oil Canola Fats and others 700 Production (millions of gallons/yr) Years Source: FAPRI, 2007 The average return over operating costs for some biodiesel plants in the US has already declined because of high prices of biofeedstocks and in many cases is currently negative. One indicator of this is idled production capacity: total US production of

24 9 biodiesel was 500 million gallons in 2007 as against 1.85 billion gallons of production capacity (Carriquiry and Babcock, 2008) Production costs of algal biodiesel The cost of producing biodiesel from microalgae consists of two major components: capital costs and operating costs. Capital costs include the cost of the algae production infrastructure such as the establishment of a pond, if a so-called raceway system is used for growing algae, or installation of photobioreactors if a tubular photobioreactor system is used; the equipment for distributing water and nutrients (CO 2, nitrogen, phosphorous, and other nutrients) to the algae; harvesting and extracting equipment; power/energy distribution and supply systems; and engineering costs. The raceway and tubular photobioreactor systems are described in detail below. Operating costs include labor costs, CO 2, energy and nutrient supply costs, and maintenance costs (Benemann and Oswald, 1996). Brown and Sprague (1992) reported that CO 2 is by far the biggest input and thus a major component of cost of production of microalgae because atmospheric CO 2 (0.033%) is not enough to produce a large algal biomass. The cost of biodiesel from microalgae was estimated in the 1980s to be about $5.05/gal of which $1.48 was for carbon dioxide (Brown and Sprague, 1992). After algal oil is produced, the cost of producing biodiesel includes the cost of converting algal oil into biodiesel minus the value of glycerol that is produced as a byproduct during the transesterification process (Duffield et al., 1998). Benemann and Oswald (1996) reported average production costs for algal oil of $59 and $39 per barrel under productivity assumptions of 30g/m2/day and 60g/m2/day plant capacity and 380

25 10 and 760 barrel/ha/yr biodiesel production respectively (Table 2.1). However, these costs do not include the cost required for converting algal oil into the biodiesel. Table 2.1. Estimated costs for biodiesel production from microalgae Cost ($/bbl) References Productivity assumption/remarks 69.3 Neenan et al. (1986) 50g/m2/d Brown and Sprague (1992) $62.16 for CO 2 56 Huntley and Redalje (2007) 30-60g/m2/d 39 & 59 Benemann and Oswald (1996) 60 & 30g/m2/d respectively Briggs (2004) Simulation based 127 & 65 Sheehan et al. (1998) Conservative & optimistic approach 62 Weissman and Goebel (1987) 1000 acre scale 853 Dimitrov (2007) Conservative estimate for capital costs These cost estimates were made for raceway system of algae production based on 1980s research. Huntley and Redalje (2007) estimated an average cost for algal biodiesel production of $56/barrel of oil equivalent (in 2003 dollars) for either a raceway system or a tubular photobioreactor system. These cost estimates, however, might not represent actual algal biodiesel costs as they are the costs for pilot plants and laboratory facilities. On the other hand, Dimitrov (2007) argued that the production cost of algal biodiesel would be so high that it would be economically viable only if the price of petroleum was about $853 per barrel ($20.31 per gallon), and this was with what he termed a conservative estimate for capital costs (Table 2.1). The surprisingly large difference in cost estimates between Benemann and Oswald (1996) and Dimitrov (2007) might be due the differences in the system of production of microalgae, different cost estimation procedures, and different time periods. Benemann and Oswald (1996) used the raceway system of production for microalgae while Dimitrov (2007) used a tubular photobioreactor, which is usually a more expensive production system.

26 11 Briggs (2004) estimated a total cost of $354.2 billion ($308 billion in pond construction where the algae would be grown plus $46.2 billion per year in operating costs) to produce 138 billion gallons of biodiesel, which would completely replace petrodiesel in the US transportation sector. This implies an average cost for algal biodiesel of about $2.60 per gallon. Sheehan et al. (1998) estimated the cost of algal biodiesel to be two times higher than petro-diesel even with aggressive assumptions about the biological productivity of the algae, given energy prices prevailing in the late 1990s. Assuming that algae production is located near a coal-fired power plant and utilizes flue gas as a source of CO 2, they estimated average production costs of $127 and $65 per barrel of oil equivalent for their conservative and optimistic cases, respectively. Weissman and Goebel (1987) concluded that production costs would be $62 per barrel of algal lipids at a 1,000 acre scale. They also found that a 50 percent increase in biomass productivity induced a 20 percent decrease in the production cost of algal oil. Neenan et al. (1986) reported production costs of about $1.65 per gallon of biodiesel with a productivity assumption of 50g/m2/day and a microalgae biomass feedstock production cost of $211/metric ton. Production costs depend largely on the availability of an inexpensive source of CO 2, water resources availability, high lipid content in the algae, and an efficient method for harvesting the algal biomass (Weissman and Goebel, 1987). The production cost for algal oil also depends on the type and scale of production system employed. A multi-acre open pond system for growing algae (such as 10 acre pond) would be less expensive compared to a one acre pond because of the lower fixed cost involved in the construction of multi-acre pond (Putt, 2007).

27 12 Government policy is a key factor in deciding the fate of algal biodiesel. Wassell and Dittmer (2005) studied the economic efficiency of subsidies for biodiesel in US and concluded subsidies to be effective in increasing biodiesel used in non-road equipment Microalgae species and oil content Microalgae vary in their oil content from less than 10 percent to as high as 80 percent of their dry weight (Table 2.2). Table 2.2. Oil content of some selected species of microalgae for biodiesel Species of microalgae Lipid content (% of dry wt) References Amphora sp Benemann and Oswald (1996) Botryococcus braunii Chisti (2007) Chlorella vulgaris Becker (1994) Chlorella sorokiniana 20 Illman et al. (2000) Spirogyra sp Becker (1994) Dunaliella salina 14.4 Zhu and Lee (1997) Dunaliella primolecta 23 Chisti (2007) Euglena gracilis Becker (1994) Isochrysis sp Chisti (2007) Tetraselmis suecica Otero and Fabregas (1997) Prymnesium parvum Becker (1994) Neochloris oleoabundans Chisti (2007) Isochrysis galbana Fidalgo et al. (1998) Scenedesmus sp FAO (2008) Porphyridium sp FAO (2008) Schyzochytrium sp Chisti (2007) Future efforts to bring down the average cost of algal biodiesel production depend in part on the ability to develop species of microalgae with high oil content. It has been reported that the lipid content in microalgae can be increased substantially by limiting the nitrogen supply in the algal culture (Benemann and Oswald, 1996). Nitrogen sufficiency increases productivity but decreases the oil content (Tornabene et al., 1983; Lewin 1985).

28 13 High productivity and high lipid content were inversely related in many species of microalgae under the application of genetic engineering techniques utilizing manipulation of biosynthetic pathways (Sheehan et al., 1998; Huntley and Redalje, 2007) Production processes for microalgae Microalgae can be grown in water given sufficient sunlight availability and temperatures ranging between 20 and 30 0 C (Sheehan et al. 1998, Chisti, 2007). Microalgae production also requires a sufficient amount of carbon dioxide and appropriate amounts of phosphorus, nitrogen, iron and silicon (diatom) for proper growth. There are basically two systems of large scale microalgae production: raceway ponds and tubular photobioreactors. In the raceway system algae are grown in open shallow ponds containing water and nutrients. Paddlewheels are designed in order to mix and circulate the flow thereby preventing any sedimentation. Fig. 2.2a. Open raceway system of algae production Source: Sheehan et al. (1998)

29 14 Fig. 2.2b. Open raceway system of algae production (photo) Source: Enhanced Biofuels & Technologies (2007) A tubular photobioreactor consists of a tubular array of straight transparent tubes where sunlight is captured. The system is designed in such a way that the microalgal broth is constantly circulated between the solar collector and a reservoir (Chisti, 2007). Although a larger algal biomass production is possible in this system, it has a higher capital cost than the open pond raceway system of production. Harvesting can be done through filtration, flocculation or settling methods. A detailed diagram of the production process for crude algal oil and algae biomass using a photobioreactor is shown in Fig The wet algae paste produced from centrifugation is used for oil extraction and in turn the production of crude algal oil.

30 15 Fig. 2.3 Algae production flow chart Source: AlgaeWay (2008) 2.5 Important issues in the production of microalgae Microalgae thrive well in a shallow water body in a sunny location with ample CO 2 and optimum amounts of nutrients. One of the advantages of algal biodiesel is that microalgae can be grown even in brackish water, saltwater or wastewater, and high levels of CO 2 can be effectively utilized for the production of biodiesel. It has been shown that 160 billion kg CO 2 is needed for producing 50 billion kg of algal biomass annually (Brown and Sprague, 1992). Similarly one million tons of CO 2 can be captured by growing 183 million tons of algae (Chisti, 2007). The need for high solar radiation, warm

31 16 weather, and flat land with sufficient water makes the southwestern part of United States a potentially suitable location for the production of algal biomass (Brown and Sprague, 1992). Even though the Southwest is rich in saline aquifers, the sustainability of this region in providing a sustained volume of saline water for mass culture of microalgae is yet to be determined because of the high evaporation loss of water if algae are produced in an open pond system (Neenan et al., 1986). It would be better if a microalgae production facility was located near a coal- or oil-fired power plant as the flue gas from these plants contains a high level of CO 2 (as high as 13%), which can be utilized for enhanced algal biomass production (Abler and Ghimire, 2008; Li et al., 2008). Although the CO 2 used by the microalgae during its growth is ultimately released to the atmosphere when the biodiesel is combusted, the processes can double the quantity of energy produced for a given level of CO 2 (Brown and Sprague, 1992). Another important facet of microalgae production near power plants is that the hot water produced in the plant can be utilized in order to maintain the required level of water temperature for algae growth, which can be important during nighttime, even in warm locations. Many power plants install expensive cooling systems in order to treat and convert hot wastewater produced in the plant before finally disposing it to nearby water bodies. If the hot water can be utilized for maintaining the optimum temperature for algae growth, algae production costs can be reduced while power plants can reduce water cooling costs. Heat from the flue gas can also be used for drying the algae, which is necessary before oil is extracted (Abler and Ghimire, 2008). Microalgae can not only be used for the production of various types of biofuels including biodiesel but its by-products produced during the extraction of algal oil can

32 17 also be used for the production of pharmaceuticals, fertilizer, animal feed, etc. Nothing need go as waste in extracting oil from microalgae. However, commercial production efforts have been largely unsuccessful mainly due to the difficulty of mass scale culture and maintenance of the microalgae in an open pond production system (Neenan et al., 1986). For more successful mass culture of microalgae, the biological productivity of microalgae species should be improved either through increased photosynthetic efficiency (currently 12-16%), increased lipid content (currently 50-60% in many cases), or both (Neenan et al., 1986). Genetic engineering shows promise in achieving these objectives (Sheehan et al., 1998). Considerable attention is also being paid in engineering research and development to the design and development of an efficient system of harvesting algal biomass, processing the biomass, and extracting algal oil Overview of world petroleum oil demand and supply Some have argued that the world supply of petroleum will eventually begin to decline as early as 2010 but that demand for oil in the transport sector will continue to increase due to increases in population and per capita income (Marinelink.com, 2002). This is commonly known as the peak oil argument.

33 18 Fig World projected oil demand Oil demand Transport sector demand Demand (mb/d) Year Source: Energy Information Administration (EIA) (2008) The growth in transport sector fuel demand is expected to be met largely by growth in petroleum-based fuels in developing countries. The demand for petroleum in the world is expected to increase from 84 million barrels per day (mb/d) in 2008 to almost 113 mb/d by 2030 (Fig. 2.4). The demand for liquid fuel in the transport sector is forecasted to reach as high as 63 mb/d which is about 56 percent of the total petrol-oil demand in the world by Currently about 51 percent of the total oil demand in the world is from the transport sector (EIA, 2008). The supply of oil in the world is expected by EIA to increase to as much as 118 mb/d from 88 mb/d in 2008 in the reference case. World oil production is projected to be different under different oil prices. Under high oil prices the production of petroleum is projected to be low and vice-versa for the low oil price case (Fig. 2.5).

34 19 Fig World projected oil supply under different scenarios 160 High oil price Low oil price Reference case 140 Petro-liquid production (mbd) Year Source: EIA (2008) Under high oil prices the production of biofuels like algal biodiesel would be more competitive and the demand for petroleum might decline as consumers substitute biofuels for petro-fuel. Under low oil prices biofuels would be less competitive Overview of diesel and biodiesel demand in the US Currently the US consumes about 2.96 million barrels per day of petro-diesel in the transport sector alone. This figure is projected by the Energy Information Administration (EIA) to increase to 4.19 millions of barrel per day (mb/d) by 2030 which will be about 22.8 percent of total transport fuel consumption in the country (Fig. 2.6).

35 20 Fig Projected diesel and biodiesel demand in transport in US Consumption (mb/day) Petro-diesel demand Biodiesel demand Years Source: EIA (2008) Biodiesel production from vegetable oil and fats contributes only 0.67 percent of the total petro-diesel consumption in EIA s reference case. The consumption of biodiesel turns out to be very small at about 0.03 millions of barrels per day. Fig Projected share of petro-diesel in transport fuel consumption in US Share of petro-diesel in transport Share o transport fuel in total liquid fuel Share of fuel (%) Year

36 21 Source: EIA (2008) In 2008 the transport sector consumed about 67 percent (13.9 mb/d) of the total liquid fuel consumption in the United States, and that percentage is projected by EIA to increase to 74 percent (18.4 mb/d) by 2030 (Fig. 2.7).

37 22 Chapter 3 METHODOLOGY The study applies simulation modeling techniques in order to determine the future demand, supply and price structure of algal biodiesel under different scenarios. A partial equilibrium analysis of supply and demand of algal biodiesel was carried out under two basic modeling approaches: a price taker model and a price maker model Price taker model In this model the supply of algal biodiesel was derived from a Cobb-Douglas type cost function. The demand of algal biodiesel was assumed to be perfectly elastic at its prevailing price. The price of algal biodiesel was determined based on its energy content relative to petro-diesel Cost function Constructing a cost function for biodiesel production from microalgae requires information on all costs including the production of microalgae and processing for oil extraction. The cost function was specified as follows: C = f(v, q algalbiodiesel, q lag, T) Where C is the cost of algal biodiesel production, v is an index of prices of inputs into production

38 23 q algalbiodiesel is the quantity of algal biodiesel produced, q lag is the lagged quantity of algal biodiesel production, and T is technology. The specific form of cost function is specified as follows: -rt γ ϕ C(v, q) = Ae vq t q t 1. The input price index includes prices of capital inputs, labor, nutrients, land, energy, and other inputs. A, α, β, φ and γ are parameters, r is the rate of technical change.. γ is the returns to scale parameter. q t and q t 1 represent current and the lagged output (quantities of algal biodiesel). T is technology. Technical change can be achieved by either an increase in output from a given bundle of inputs, a decrease in the inputs required to achieve a given level of output, or both. In other word it implies a downward shift in isoquants and an upward shift in the production possibilities frontier Supply function The quantity supplied of the algal biodiesel is derived from the cost function. Differentiating the cost function with respect to q we obtain, C q t rt γ 1 MC P Aγ e = = = vq q t t ϕ t 1 γ 1 1 rt 1 q p ϕ = e v q t γ A t t 1 1 ( ) ( ) 1 rt γ 1 γ A ϕ ( γ 1) ( γ 1) ( γ 1) ( γ 1) q t = p t e v q t 1

39 24 where, 1 1 r (0) 1 γ ϕ A = p 0 e v 0 q 0 q γ 0 1 Equating marginal cost with price we get rt γ 1 ϕ p t = MC = Aγ e vq t q t 1 Taking log in both sides we get ln p t = ln A + ln γ rt + ln v + ( γ 1) ln q t ϕ ln q t 1 Rearranging the equation, we get ln p t = ln A + ln γ + ln v ϕ ln q t 1 rt + ( γ 1) ln q t or ( γ 1) ln q t = c + ln p t + rt + ϕ ln q t 1 where c = ln A + ln γ + ln v. This means that c 1 r ϕ ln q t = + ln p T ln q ( 1) ( 1) t + + γ γ ( γ 1) ( γ 1) t 1, 1 r ln q t = k + ln p T ln q ( 1) t φ γ + ( γ 1) + t 1 where k = c ( γ 1)

40 25 and φ ϕ ( γ 1) = Demand function for the price taker model In the price taker model demand is assumed to be perfectly elastic at the prevailing real market price of algal biodiesel. The price (P t ) of algal biodiesel was derived from the price of petro-diesel by multiplying by the factor (.914), which is an estimate of the energy content of algal biodiesel relative to petro-diesel (NBB, 2005; Szulczyk, 2007): P t =.914*Price of petro-diesel (RPPD) The real (inflation-adjusted) price of petro-diesel was calculated as follows: pd RPPD =f t *P 0 where ft = CPI f t CPI t CPI f 0 CPI0 = f CPI t CPI t CPI. 0 f CPI 0 and where CPI f = Consumers price index for fuel and power taken from the Energy Information Administration (EIA) database, CPI = Consumer price index for all commodities taken from EIA, where the subscript t indicates the time period and the subscript 0 indicates the base period (2008).

41 Price maker model In this model demand for algal biodiesel is downward sloping. The demand function was derived from total demand for diesel fuel in transport sector after subtracting the supplies of petro-diesel and other (non-algal) biodiesel. The supply function was derived from the cost function as in the price taker model Demand function The demand function for algal biodiesel was specified as follows: D algal biodiesel = D diesel S petrodiesel -S other biodiesel where D diesel = f(pd, D t-1, I) S petrodiesel = g(ppd, SPD t-1, T) S biodiesel = h(pbd, SBD t-1, T) D diesel is demand for all types of diesel in transport sector D t-1 is the lagged demand for all types of diesel in transport sector PD is the price of diesel I is per capita income, S is the supply of petro-diesel PPD is the price of petro-diesel PBD is the price of biodiesel T is the technology SPD t-1 is the lagged supply of petro-diesel SBD t-1 is the lagged supply of biodiesel

42 27 All demands and supplies are measured in millions of barrels of oil equivalent per day. The structure of diesel demand can be written as: DD β0 β1 β2 β = e PPD I D 3 t 1 where DD is demand for all types of diesel PD is price of diesel I is per capita gross domestic product, β 0 is the intercept, and β, β, β are elasticities of diesel demand with respect to own price, income and lagged demand respectively. The supply function for petro-diesel can be specified as: SPD = γ T e 0 γ PPD 1 γ SPD 2 γ t e 3 1 where SPD is the supply of petro-diesel, PPD is the price of petro-diesel SPD t 1 is the lagged supply of petro-diesel, and T is the technology or time trend. The coefficients of price and lagged supply are interpreted as elasticity of petro-diesel supply with respect to price and lagged supply. By the same token, supply function for the other biodiesel can be specified as: α T SBD e 0 α PBD 1 α S 2 α = t e 3 1

43 28 where PBD =.914*PPD So, SBD = α T e 0 α α PPD 1 α S 2 α t e 3 1 where SBD is the supply of other biodiesel, PBD is the price of biodiesel PPD is the price of petro-diesel SBD t 1 is the lagged supply of other biodiesel, and T is the technology or time trend The coefficients of price and lagged supply are interpreted as elasticities of biodiesel supply with respect to price and lagged supply. Hence, the demand for algal biodiesel will be as follows: D a lg ae β e 0 β PPD 1 β I 2 β = D 3 t 1 α T e 0 α α PBD 1 α S 2 α t e 3 1 γ T e 0 γ PPD 1 γ SPD 2 γ t e Supply function The supply function for algal biodiesel was derived from the cost function in exactly the same manner as in the price taker model.

44 29 Chapter 4 RESULTS AND DISCUSSION This chapter contains the simulation results under different parameter values and different scenarios regarding rates of technical change in algal biodiesel production. All the simulation analyses begin with a conservative base-period value of algal biodiesel production of 2,000 gallons per year. This is based on reports that Solazyme Inc., a company producing biodiesel from microalgae, has been producing thousands of gallons of algal oil for biodiesel (Green Car Congress, 2008). Also, in all the simulation analyses the inputs into algal biodiesel production are assumed to be perfectly elastic in supply, and the input price index (v) is normalized to one. Unless otherwise specified, the coefficient on lagged supply (φ ) is set equal to 0.8. The price of algal biodiesel in the simulation analyses is calculated from the price of petro-diesel based on relative energy content. The literature indicates that the energy content of biodiesel is of that of petro-diesel (NBB, 2005; Szulczyk, 2007), so that the price of algal biodiesel in the model is 91.4% of the price of petro-diesel. 4.1 Cost structure The cost structure of algal biodiesel includes an analysis of total cost, average cost and marginal cost. The simulation results for different parameters and technical change scenarios are discussed here.

45 Scenario I In this scenario the returns to scale parameter is γ = 2.5 and r = 0.20 or 0.30, where r stands for rate of technical change. This value of γ implies that the short-run own-price elasticity of algal biodiesel supply is 1/(γ 1) 0.7 and the long-run own-price elasticity is [1/(γ 1)]/(1 φ ) 3.3. The values of r imply an annual rate of Hicks-Neutral technical change of 20 or 30 percent. In this scenario it turns out that with a 20 percent annual rate of technical change production increases very slowly. Total cost is lower at a 20 percent rate of technical change compared to the 30 percent case due to lower algal biodiesel production. However, average costs and marginal costs tend to be very similar for both technical change scenarios. Here 30 percent technical change implies either a 30 percent increase in the production of algal biodiesel at given levels of inputs, a 30 percent decrease in the inputs required to achieve given level of biodiesel production, or some combination of the two. A 20 percent technical change can be interpreted similarly. The average cost for 2008 turns out to be about $43.20/bbl which increases to $50.30/bbl in 2030 (Table 4.1). The total cost curves are given in Fig Marginal cost curves are steeper than average cost curves (Fig. 4.2), which is a property of the Cobb-Douglas cost function under decreasing returns to scale. Marginal cost is equal to the market price which was derived from EIA data. Total cost curves for 20 and 30 percent rates of technical change are given in Fig. 4.1 and Fig. 4.3 respectively.

46 31 Fig Total cost curves 2015/17 scenario I (r =.20) TC15 TC16 TC17 TC (m$) Quantity (mb/yr) Table 4.1. TC, MC and AC scenario I 20% Technical change 30% Technical change year TC(m$)* MC($/bbl) AC($/bbl) Q(mb/yr) TC(m$) MC($/bbl) AC($/bbl) Q(mb/yr) E E E E E E E E E E E E E E *Total cost is expressed in millions of dollars