Click for next page ( 74


The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 73
6 Hydrogen and Alternative Technologies for Reduction of U.S. Oil Use and CO2 Emissions Estimating future transportation fuel use is difficult period to bring hydrogen fuel cell vehicle technologies to because of the complexities and uncertainties inherent in cost competitiveness with gasoline vehicle technology; and the analysis. Petroleum may continue to be the dominant the costs for a future hydrogen infrastructure. The committee fuel, or production may become constrained and prices rise developed three scenarios in order to investigate the range much further. Hydrogen may replace petroleum as the main of possible outcomes. The hydrogen scenario analyses are fuel, or it may not become significant at all. As discussed in based on the results presented in Chapter 3. In addition, Chapter 3, fuel cell vehicles and hydrogen have the potential as discussed in Chapter 4, hydrogen is not the only way to to become competitive with conventional vehicles and fuels, reduce petroleum use. Two scenarios focused on alternatives but it is far from certain when that might be. Competitive- are analyzed, and a final scenario looks at combining all the ness depends in part on the cost of petroleum, which itself is approaches. highly uncertain, as witnessed by recent dramatic escalations in world oil prices. Nevertheless, as discussed in Chapter 2, • Case 1 (Hydrogen Success) assumes that development there appear to be compelling reasons why the nation may programs are successful, as shown in Table 6.1, and that have to reduce its use of petroleum, and hydrogen is among policies are implemented to ensure commercial deployment. the leading candidates proposed to achieve dramatic reduc- Hydrogen fuel cell vehicles are introduced starting with a tions. Moving to a hydrogen-based transportation sector few thousand vehicles in 2012, growing to a fleet of almost would be a revolutionary change that is unlikely to happen 2 million by 2020, 60 million in 2035, and 220 million in by itself. Mapping a route is essential to understanding how 2050 (Figure 6.1). This rapid-growth case corresponds to such a change might happen. Toward that end, this chapter a scenario recently developed by the U.S. Department of formulates and analyzes several scenarios to map plausible Energy (DOE) to 2025 (Gronich, 2007) and extended by the futures for the use and impacts of hydrogen fuel cell vehicles committee to 2050. By 2050, 80 percent of new vehicles sold (HFCVs) and other alternative vehicles and fuels. The sce- are assumed to be HFCVs (Figure 6.2). This is consistent narios and analyses necessarily depend on a host of assump- with other recent modeling studies (Greene et al., 2007). tions. None of the scenarios should be viewed as projections —Case 1a (Hydrogen Accelerated) assumes that hydro- of what the committee thinks is likely to happen. Rather, gen and fuel cell vehicles are introduced at twice the rate of they are intended to describe different paths along which Case 1, although technical and cost goals are met at the same events may unfold and the consequences, especially for oil rate as Case 1. By 2020, 4 million hydrogen fuel cell vehicles consumption and carbon dioxide emissions. are in the fleet. By 2050, 95 percent of new vehicles sold are assumed to be HFCVs This case is intended to investigate whether hydrogen could replace even more oil than in Case SCENARIOS and analysis 1 if the nation faces a crisis situation, perhaps from declin- ing worldwide petroleum production or rapidly worsening Scenarios global climate change. The main object of the scenario analysis is to estimate the —Case 1b (Hydrogen Partial Success) assumes that maximum practicable penetration rate of fuel cell vehicles, developing programs fall short of the costs and performance and then to estimate the resulting reductions of petroleum use of Case 1 (Hydrogen Success). Thus, the market introduction and emissions of carbon dioxide (CO2) in 2020 and beyond; rate is slower than for Case 1, similar to DOE’s “Scenario the investments that would be needed during a transition 1” (Gronich, 2007). By 2020, fewer than 1 million HFCVs 73

OCR for page 73
74 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A focus on hydrogen TABLE 6.1  Assumed Cost and Performance of Hydrogen Fuel Cell Vehicles and Gasoline Reference Vehicles a Case 1: Case 1b: Reference Hydrogen Hydrogen Gasoline ICEV Efficient Gasoline Success Partial Success (AEO 2008) ICEV (Case 2) FC drivetrain retail Costs fall with Costs fall with $54/kW $64/kW price (including learning and learning and fuel cell and manufacturing manufacturing hydrogen storage scale to $100/kW scale to $130/kW HFCV retail price >$100,000 >$100,000 — $1,000 increment compared (initially) → (initially)→ to gasoline $3,600 $6,100 reference vehicle (learned out) (learned out) FCV market 2012 (Case 1) 2015 — — Introduction 2010 (Case 1a) New car fuel 51 mpgge 45 mpgge 2005: 20.2 mpg 20.2 mpg economya (2015) → (2015) → 2015: 25.0 25.2 85 mpgge 73 mpgge 2050: 31.7 42.4 (2050) (2050) = 2 × efficient = 1.75 × efficient gasoline case gasoline case NOTE: Case 1a, Hydrogen Accelerated, is the same as Case 1, Hydrogen Success, for these values. Costs and fuel economy of HFCVs are based on a refer- ence midsize vehicle with an 80 kW fuel cell. While small relative to most current engines, this would give equivalent performance, in part because of weight reductions in the body. This vehicle is assumed to represent the range of vehicles from small to large. Given all the other uncertainties in this analysis and the limited resources available to the committee, this assumption was both reasonable and unavoidable. aOn-road fuel economy is assumed to be 80 percent of the EPA average. 400 1.2 Case 1 (H2 Success): Million light-duty vehicles 350 gasoline Case 1 (H2 1 Case 1a (H2 Accel): Success): gasoline 300 gasoline Case 1a (H2 Accel): 250 0.8 Case 1b (H2 Partial gasoline Fraction 200 Success): gasoline Case 1b (H2 Partial 0.6 Case 1 (H2 Success): Success): gasoline 150 HFCV 0.4 Case 1 (H2 100 Case 1a (H2 Accel): Success): HFCV HFCV 0.2 50 Case 1a (H2 Accel): Case 1b (H2 Partial HFCV 0 Success): HFCV 0 Case 1b (H2 Partial 2000 2010 2020 2030 2040 2050 TOTAL 2000 2010 2020 2030 2040 2050 Success): HFCV Year Year color Figure 6-1.eps FIGURE 6.1  Hydrogen cases: Number of gasoline and hydrogen FIGURE 6.2  Hydrogen cases: Fraction of new gasoline and hydro- color Figure 6-2.eps fuel cell vehicles in the fleet over time for three hydrogen cases. gen vehicles sold each year. are on the road, about the same rate of market penetration as • Case 3 (Biofuels) examines the large-scale use of bio- hybrid electric vehicles have experienced. fuels production from crop and cellulosic feedstocks. This • Case 2 (ICEV Efficiency) investigates the impact of level of production, equivalent to a maximum practical rate, improving conventional internal combustion engine ICEV is based on the results in Chapter 4. fuel economy with currently feasible and expected technol- • Case 4 (Portfolio), “all of the above,” analyzes the ogy. Fuel economy more than doubles by 2050, as shown in impact if HFCVs, more efficient conventional vehicles, and Table 6.1, in the committee’s judgment the maximum practi- biofuels are all pursued simultaneously. cal rate with evolutionary vehicle technology. This analysis is based on the results in Chapter 4.

OCR for page 73
HYDROGEN AND ALTERNATIVE TECHNOLOGIES FOR REDUCTION OF U.S. OIL USE AND CO2 EMISSIONS 75 TABLE 6.2  Hydrogen Supply Pathways Considered in This Analysis Hydrogen Production Hydrogen Delivery Method to Station Resource Technology (for Central Plants) Natural gas Steam methane reformation N/A (on-site)a Liquid hydrogen truck Steam methane reforming Compressed gas truck (central plant) Hydrogen gas pipeline Coal Coal gasification with carbon Liquid hydrogen truck capture and sequestration Compressed hydrogen gas truck (central plant) Hydrogen gas pipeline Biomass (agricultural, forest, Biomass gasification (central Liquid hydrogen truck and urban wastes) plant) Compressed hydrogen gas truck Hydrogen gas pipeline On-site reforming of ethanol N/A Electricity (from various electric Water electrolysis (on-site) N/A generation resources) NOTE: N/A = not applicable. aOn-site refers to hydrogen production at the refueling station. These scenarios are compared to a reference case, based (NRC) FreedomCar Fuel Partnership report (NRC, 2008). on the Energy Information Agency (EIA) 2008 Annual Available hydrogen supply pathways are listed in Table 6.2. Energy Outlook (EIA, 2008). The committee selected the Hydrogen production or storage technologies that would AEO high-oil-price scenario for its reference case as being require fundamental scientific breakthroughs (for example, more representative of conditions under which HFCVs hydrogen storage in carbon nanostructures or biological are promoted than the AEO reference case. This scenario production of hydrogen by algae) are not considered. includes improvements of gasoline ICEV technology to meet Cost and performance data for current and midterm CAFE (corporate average fuel economy) standards, although (2015-2030) hydrogen infrastructure technologies are dis- fuel economy continues to grow slowly after 2020, and some cussed in Chapter 3. Efficiency improvements in ICEVs and use of biofuels (blending up to 10 percent ethanol) but no biofuels are described in Chapter 4. introduction of hydrogen or advanced ICEV technology. Gasoline taxes continue as per AEO 2008. Modeling Tools for Scenario Analysis The time frame for analysis is 2008 to 2050. The commit- tee agreed that HFCVs were not likely to make a large impact The committee developed two EXCEL-based models for on U.S. oil use and greenhouse gas emissions by 2020, infrastructure and scenario analysis: because they are unlikely to enter the market before 2012- 2018, and then it will take time to build up a large enough 1. Hydrogen infrastructure model: designs and costs number of vehicles to impact oil use and carbon emissions hydrogen infrastructure to meet a specified market penetra- significantly. The committee recognizes that uncertainties tion for HFCVs. increase in such a long-term analysis, but it was necessary 2. Simplified transition model: estimates investment to for examining the time frame during which hydrogen could bring HFCV costs to competitive levels, investment costs for have a large impact. building hydrogen infrastructure, oil savings, and greenhouse gas emission reductions, over time. Technologies Considered The models were developed at the University of California Hydrogen and fuel cell technologies are based on tech- at Davis (UC Davis) and are described in detail in Appendix nologies currently in development, as discussed in Chapter C. Given the time and resources available to the committee, 3 and recently reviewed in the National Research Council the models were of necessity relatively simple, but they In this scenario, imported low-sulfur crude oil is projected to cost about $79 per barrel in 2010, rising to $90 in 2015, $102 in 2020, and $119 in Carbon dioxide is the main greenhouse gas of concern in this analysis, 2030 (all in 2006 dollars). Oil was over $130 per barrel in June 2008, but but other gases, especially methane and nitrous oxide, are emitted as part of that does not necessarily mean that the AEO numbers are wrong. Other the full fuel cycle. These are accounted for with global warming equivalency projections are both well above and well below this one. factors, taken from the literature and the GREET model.

OCR for page 73
76 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A focus on hydrogen TABLE 6.3  Assumptions in Reference Case 2005 2020 2035 2050 Total number of LDVs (millions) 220 280 331 369 Share of LDV fleet   Gasoline ICEVs 99.8% 96.6% 93.2% 91.4%   Gasoline hybrids 0.3% 3.4% 6.8% 8.6% New LDVs sold per year (millions) 16.2 18.2 20.8 22.4 Share of New LDVs per year   Gasoline ICEVs 98.6% 94.7% 91.8% 91.1%   Gasoline hybrids 1.4% 5.2% 8.2% 8.9% Average on-road fuel economy, new gasoline LDVs (mpg)   Gasoline ICEVs 20.2 29.3 30.6 31.7   Gasoline hybrids 32.1 41.0 42.9 44.5 Gasoline price ($/gallon) 2.32 3.19 3.54 3.96 Vehicle-miles traveled (billion per year) 2,556 3,251 4,243 5,364 Gasoline consumed (billion gallons per year) 124 132 140 157 (includes blends of ethanol up to 10%) Ethanol (billion gallons per year) consumed as:   Blend in gasoline to 10% 3.4 12.7 15.6 21   E 85 0.01 0.06 0.20 Greenhouse gas emissions (million tonnes CO2 equivalent per year) 1,345 1,442 1,527 1,710 were quite adequate for the purpose of scoping the potential on the DOE National Energy Modeling System (NEMS). For growth of HFCVs and their impact. input beyond 2030, a vehicle stock model was adapted from the Argonne National Laboratory VISION model to estimate numbers of LDVs, fuel economy, and vehicle energy use to Modeling Assumptions 2050 (Singh et al., 2003; Argonne, 2007). The vehicle stock • Only U.S. light-duty vehicles (LDVs) are considered. model keeps track of what vehicles are in the fleet (vehicles All scenarios assume the same total number of vehicles and are retired after a certain number of years) as increasing vehicle-miles traveled (VMT) as the reference case. numbers of hydrogen vehicles enter the market, allowing the • The AEO high-price case is used as the basis for energy calculation of energy use and greenhouse gas emissions for prices. each year. Table 6.3 and Figures 6.3 to 6.5 summarize the • The real discount rate is 15 percent (no inflation is reference case. After the committee had completed its initial included). analysis, Congress passed the Energy Independence and • The costs are given in 2005 constant dollars. Security Act of 2007 (EISA), which included a significant • Costs for hydrogen infrastructure technologies (hydro- increase in fuel economy standards for vehicles. The refer- gen production, storage, delivery, refueling stations) draw ence case reflects this new policy. heavily on DOE’s H2A database (DOE, 2007). Both current The reference case includes modest use of biofuels. In and 2015 technology numbers are used where available (for 2030, the reference case assumes that 15 billion gallons of production and refueling station technologies). ethanol are used per year, 12 billion from corn ethanol and 3 billion from cellulosic ethanol. This gradually increases to 21 billion gallons per year in 2050 (12 billion gallons from HYDROGEN SCENARIO ANALYSIS corn, 9 billion gallons from cellulosic feedstocks). This is equivalent to about 8-10 percent ethanol by volume in gaso- Reference Case line after about 2020. Cases 1 through 4 (hydrogen and other alternative fueled EISA includes ambitious goals for biofuels as well as fuel economy. vehicle) are compared against a reference case. The reference case is based on the high-price case of the AEO (EIA, 2008), The goal for 2020 is 36 billion gallons, of which 21 billion would be from advanced processes, such as cellulosic ethanol. The committee decided not which gives projections to 2030 for vehicle miles traveled, to include these goals in its reference case for two reasons. First, the likeli- vehicle fuel economy, and vehicle fleet composition based hood of meeting the biofuel goals appears to be much lower than for fuel

OCR for page 73
HYDROGEN AND ALTERNATIVE TECHNOLOGIES FOR REDUCTION OF U.S. OIL USE AND CO2 EMISSIONS 77 400 14 Million light-duty vehicles 350 Billion gallons per year 12 300 10 Gasoline ICEV 250 Gasoline HEV 8 Corn ethanol 200 HFCV 6 Cellulosic ethanol 150 Total 4 100 50 2 0 0 2000 2010 2020 2030 2040 2050 2000 2010 2020 2030 2040 2050 Year Year FIGURE 6.3  Reference case: Number of light-duty vehicles in color Figure 6-3.eps color Figure 6-5.eps FIGURE 6.5  Reference case: Assumed biofuel use. the fleet. 50 • For Cases 1 (Hydrogen Success) and 1a (Accelerated 45 40 Hydrogen), the fuel cell drivetrain (the fuel cell system, Fuel economy (mpgge) 35 hybrid battery, motor, and auxiliaries) costs the automaker 30 (original equipment manufacturer, or OEM) $50/kW. This 25 corresponds to a fuel cell system cost of $30/kW plus added 20 Gasoline ICE costs for a hybrid battery, electric motor, and other compo- 15 10 Gasoline HEV nents. Of the $30/kW fuel cell system cost, about half is due 5 to the fuel cell stack and half to the balance of plant. Hydro- 0 gen storage costs the OEM $10/kWh. A model from Kromer 2000 2010 2020 2030 2040 2050 and Heywood (2007) shows the total OEM manufacturing Year cost for drivetrain plus storage to be $71/kW, or a retail price FIGURE 6.4  Reference case: Assumed fuel economies for gasoline of about $100/kW, giving a drivetrain plus storage price of ICEVs and gasoline hybrid vehicles (HEVs). (The “dip” in hybrid color Figure 6-4.eps $7,920. fuel economy in 2004 occurred when hybrid sport utility vehicles • For Case 1b (Hydrogen Partial Success), the fuel cell and vans entered the market.) drivetrain costs the OEM $62/kW corresponding to a fuel cell system cost of $50/kW plus added costs for a hybrid battery, electric motor, and other components. Of the $50/kW fuel cell system cost, about 40 percent is due to the fuel cell stack and 60 percent to the balance of plant. Hydrogen Hydrogen Cases 1, 1a, 1b storage costs $15/kWh. The total OEM manufacturing cost is $93/kW or a retail price of about $130/kW, giving a drive- Table 6.1 lists cost and performance assumptions for train plus storage price of $10,400. hydrogen fuel cell vehicles and a gasoline reference vehicle for each hydrogen case. HFCV costs are based on an 80 The drivetrain and fuel storage for a reference gasoline kW fuel cell “engine” with 5 kg (165 kWh) of compressed internal combustion engine car are assumed to have an OEM hydrogen gas stored on board. (The drivetrain includes the cost of $35/kW plus $300 for the exhaust system. For an fuel cell and auxiliaries, a hybrid battery, electric motor, 80 kW engine, the OEM cost is $3,100 and the retail price wiring, and hydrogen storage.) This is consistent with the $4,300 ($54/kW). The price for each vehicle is broken down following assumptions: into a drivetrain and a “glider” (the rest of the vehicle). For all vehicles the glider price is the same, $18,750. The HFCV price is assumed to decrease according to a learning economy. Second, the net oil displacement and CO2 emission reduction are much less certain than for fuel economy improvements. These factors are curve model developed by Oak Ridge National Laboratory discussed in Chapter 4. researchers (Greene et al., 2007), based on automobile The 2006 reference gasoline vehicle is based on a midsized five pas- manufacturers’ estimates of fuel cell vehicle costs in mass senger car, with a curb weight of 1,570 kg. As efficiency improves over production (Figure 6.6). time, the weight is reduced to about 1,280 kg by 2030. The weight of the Cases 1 and 1a assume that the HFCV has twice the fuel corresponding HFCV is 1,320 kg, reflecting heavier components. This refer- ence vehicle is about average for the current new car fleet and is assumed economy of an efficient gasoline ICEV, described in Case 2 to represent the fleet. Similarly, the HFCV that replaces it is assumed to be (ICEV Efficiency) below. (The evolving efficient gasoline representative. HFCVs, like conventional vehicles, will range from small to ICEV in Case 2 has fuel economy of 25.2 miles per gallon large, but the fuel savings can still be determined from the average.

OCR for page 73
$40,000 $30,000 $20,000 2005 2010 2015 2020 2025 2030 2035 78 Year TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A focus on hydrogen Gasoline ICE Gasoline ICE HFCV $120,000 $120,000 $110,000 $110,000 HFCV $100,000 Vehicle first cost Vehicle first cost $100,000 $90,000 $90,000 $80,000 $80,000 $70,000 $70,000 $60,000 $60,000 $50,000 $50,000 $40,000 $40,000 $30,000 $30,000 $20,000 $20,000 2005 2010 2015 2020 2025 2030 2035 2005 2010 2015 2020 2025 2030 2035 Year Year FIGURE 6.6 Assumed retail prices for hydrogen and gasoline vehicles over time for Cases 1 and 1a (left) and Case 1b (right). Figure6-6.eps Gasoline ICE (mpg) in 2015 and 42.4 mpg in 2050). The hydrogen fuel cell 2007b). SSCHISM employs an idealized spatial model of vehicle has about 1.4 times the fuel economy of the gasoline $120,000 infrastructure layout in cities to design and cost alternative $110,000 hybrid in Case 2 (which is assumed to get 36.5 mpg in 2015 HFCV infrastructure pathways. Inputs include information about $100,000 Vehicle first cost and 60.3 mpg in 2050). In Case 1b (Hydrogen Partial Suc- $90,000 the level of demand (market fraction of hydrogen vehicles), cess), $80,000 vehicle has 1.75 times the fuel economy of the the fuel the city population and size, the number of stations, local $70,000 efficient gasoline ICEV. feedstock and energy prices, and constraints on viable types $60,000 Analysis of all three hydrogen cases is detailed in Appen- of supply. Outputs include the delivered hydrogen cost to $50,000 dix C.$40,000 (Hydrogen Partial Success) gave rise to only Case 1b the vehicle, hydrogen infrastructure costs, and energy use modest reductions in oil use and CO2 emissions by 2050. $30,000 and CO2 emissions for different supply pathways. Cost and $20,000 The cost of making a transition was roughly twice that of performance data about hydrogen production and delivery Case 1 (Hydrogen 2010 2015 2020 2025 2030 2035 2005 Success) and took several years longer technologies are discussed in Chapter 3. to complete. Year The committee makes several assumptions in designing Case 1a (Accelerated Hydrogen ) gave rise to a margin- the hydrogen infrastructure. ally faster transition, and the resulting reductions in oil use Figure6-6.eps and CO2 emissions were 25-33 percent greater than Case 1 1. Phased introduction. There is a phased introduction of (Hydrogen Success) by 2050. However, the estimated transi- hydrogen vehicles and stations in selected large cities, begin- tion cost for Case 1a was many times that for Case 1, because ning with cities such as Los Angeles and New York (with it assumed that more of the expensive early vehicles are interest and motivation to implement hydrogen) and moving pushed into the market in the early years of the transition. to other cities over time. This so-called lighthouse concept For these reasons the committee chose Case 1 (Hydrogen reduces infrastructure costs by concentrating development in Success) as the maximum practicable case as requested in its relatively few key areas termed “lighthouse cities.” A pos- statement of task. Cases 1a and 1b are not considered further sible schedule for phased introduction of hydrogen vehicles in this chapter. in various U.S. cities is shown in Figure 6.7. The list of 27 cities was chosen based on hydrogen scenario development work by DOE (Gronich, 2007; Melendez, 2006). Hydrogen Infrastructure Requirements and Costs 2. Station “coverage.” Initially, when hydrogen is intro- The UC Davis SSCHISM steady-state hydrogen supply duced in each lighthouse city, some minimum number of pathway model (Yang and Ogden, 2007b) is used to design hydrogen stations is needed to ensure adequate coverage and hydrogen infrastructure and estimate delivered hydrogen consumer convenience. This constraint is imposed to help costs for Case 1 (Hydrogen Success). Hydrogen equipment deal with the “chicken-and-egg” problem of assuring hydro- costs and performance are from the H2A model developed gen fuel availability to early non-fleet vehicle owners. This is by the Department of Energy (Paster, 2006). The H2A com- assumed to be 5 percent of existing gasoline stations in cities ponent-level data are combined into complete hydrogen sup- (Nicholas et al., 2004; Nicholas and Ogden, 2007). The per- ply pathways from hydrogen production through refueling centage of hydrogen stations and station capacity over time using the SSCHISM steady-state pathways model developed are shown in Figures 6.8 and 6.9. For the initial introduction at the University of California-Davis (Yang and Ogden, of hydrogen vehicles, it is assumed that 100 kg/d stations are

OCR for page 73
HYDROGEN AND ALTERNATIVE TECHNOLOGIES FOR REDUCTION OF U.S. OIL USE AND CO2 EMISSIONS 79 Figure6-7.eps FIGURE 6.7  DOE plan for introduction of light-duty hydrogen vehicles into 27 “lighthouse” cities (thousand vehicles per year introduced between 2012 and 2025). The overall build-up rate corresponds BITMAP total number of vehicles in 2025 is 10 million, and 2.5 million to Case 1. The vehicles are sold that year. SOURCE: Gronich (2007). available at 5 percent of gasoline stations for the first several added, and the fraction of gasoline stations offering hydro- years. (These very early stations might be supplied from the gen increases over time. To account for underutilization of existing industrial hydrogen system, using excess hydrogen hydrogen station equipment as demand grows, a relatively from refineries and other industrial or merchant sources.) low system capacity factor of 70 percent is assumed. This is followed by a brief period of building “medium- sized” 500 kg/d on-site steam methane reformers (SMRs) at The assumed capital costs of different hydrogen produc- 5 percent of gasoline stations. As demand grows, capacity tion systems are summarized in Table 6.4, based on H2A’s is added at each of these stations to make them 1,500 kg/d future (2015) technology assumptions (DOE, 2007). (See stations. Beyond about 2022, new 1,500 kg/d stations are also Chapter 3.) 0.6 1600 1400 0.5 Station capacity (kg/day) 1200 0.4 1000 Fraction 800 0.3 600 0.2 400 200 0.1 0 0.0 2000 2010 2020 2030 2040 2050 2000 2010 2020 2030 2040 2050 Year Year Figure6-9.eps FIGURE 6.8  Fraction ofFigure6-8.eps gasoline stations offering hydrogen, FIGURE 6.9  Capacity of new hydrogen stations by year, 2000-2050. 2000-2050.

OCR for page 73
80 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A focus on hydrogen TABLE 6.4  Assumed Capital Costs for Hydrogen 70,000 Production Systems 60,000 Pipeline Stations 50,000 Pipeline Delivery Million $ Plant Size FUTURE TECH H2A 2015 40,000 (tonne/d) Capital Cost (dollars/kg per day) Onsite SMR 30,000 20,000 Biomass Plants Central natural gas 50 621 Coal Plants 10,000 (SMR) 300 400 0 (production plant 400 375 2029 2023 2025 2027 2021 2019 2013 2015 2017 only) Central coal 250 1,275 Year (production plant 400 1,170 FIGURE 6.10  Early infrastructure capital costs for Case 1. color Figure6-10.eps only) 1,200 950 Central biomass 30 1,260 (production plant 155 860 only) 200 820 450,000 400,000 350,000 Pipeline Stations On-site SMR (station) 0.1 3,970 ($0.4 million per station) 300,000 Pipeline Delivery Million $ 0.5 1,811 ($0.9 million per station) 250,000 Onsite SMR 200,000 1.5 1,452 ($2.2 million per station) Biomass Plants 150,000 On-site electrolysis 0.1 4,325 ($0.4 million per station) 100,000 Coal Plants (station) 0.5 2,050 ($1.0 million per station) 50,000 0 1.5 1,673 ($2.5 million per station) 36 46 26 21 31 41 16 20 20 20 20 20 20 20 Year FIGURE 6.11  Capital costsFigure6-11.eps color for hydrogen infrastructure. The SSCHISM infrastructure model compares the dif- ferent possible supply options in Table 6.2 for 73 different 10 U.S. cities, finding the lowest-cost supply in each city at a $ per gallon gasoline equivalent energy specified market penetration. The best choice depends on the 8 H2: Case 1 (Hydrogen level of demand, the city size and demand density, and local 6 Success) energy and feedstock prices. For the first 5-10 years, on-site Gasoline AEO 2008 4 SMRs dominate the hydrogen supply. After that time, central high-price case 2 production plants begin to be built in large cities, with truck or pipeline delivery, although on-site SMRs are assumed to 0 persist in smaller cities. All coal hydrogen plants are assumed 2000 2010 2020 2030 2040 2050 to have carbon capture and sequestration (CCS). Biomass Year hydrogen plants are small in size (30-200 tonnes per day) to FIGURE 6.12  Estimated average cost of delivered hydrogen in the match the scale of regional biomass supply. This compares to United States and the assumed gasoline price. 250-1,200 tonnes per day for coal plants. The analysis uses color Figure6-12.eps a regional biomass supply curve (that specifies the amount of biomass available at a certain amount per tonne) (Perlack et al., 2005) to reflect biomass feedstock cost increases as contributions from biomass hydrogen as well. However, it is demand grows. important to note that the delivered costs of hydrogen from Figure 6.10 shows the capital costs for infrastructure up coal, biomass, and natural gas central plants are quite close to 2030. On-site SMRs dominate, with central production (within $0.50/kg). Thus, the choice of a feedstock may be and pipeline delivery beginning after about 2027, when the determined by other factors, such as state policies favoring first central production systems using biomass and coal are renewables. The long-term capital cost for infrastructure is built. These are accompanied by pipeline delivery systems roughly 25 percent on-site reformer stations, 25 percent cen- and stations. Biomass plants appear slightly earlier than coal tral production plants (most hydrogen comes from coal, with hydrogen plants, and more of them are built because they some from biomass), 25 percent delivery systems (pipelines are smaller in size. Later, central production dominates in predominate), and 25 percent refueling stations with truck or large cities, although on-site reformers persist in other areas pipeline delivery. The U.S. average cost of delivered hydro- (Figure 6.11). gen is shown in Figure 6.12. In terms of the amount of hydrogen produced, coal- The total infrastructure capital cost is about $2,000 per based hydrogen is the dominant source, with significant car served by the system. The total capital costs to build a

OCR for page 73
HYDROGEN AND ALTERNATIVE TECHNOLOGIES FOR REDUCTION OF U.S. OIL USE AND CO2 EMISSIONS 81 TABLE 6.5  Type of Hydrogen Supply over Time Case 1 (Hydrogen Success) 2020 2035 2050 No. of cars served 1.8 million (0.7%) 61 million (18%) 219 million (60%) (percentage of total fleet) Infrastructure capital cost $2.6 billion $139 billion $415 billion Total no. of stations 2,112 (all on-site SMR) 56,000 (40% on-site SMR) 180,000 (44% on-site SMR) No. of central plants 0 113 (20 coal, 93 biomass) 210 (79 coal, 131 biomass) Pipeline length (miles) 0 39,000 80,000 Hydrogen demand 1,410 38,000 (22% NG, 42% 120,000 (31% NG, (tonnes per day) (100% NG) biomass, 36% coal 25% biomass, 44% coal with CCS) with CCS) NOTE: NG = natural gas. “steady-state” hydrogen infrastructure to serve the demands 80 in 2020, 2035, and 2050 are estimated in Table 6.5. Note TOTAL Diff (Gas 60 minus H2) that more than $400 billion is required to build the hydro- gen infrastructure to supply the fuel for the HFCV fleet in 40 CMLTV Diff (Gas 2050. minus H2) Billion $ 20 CAP COST Diff Vehicles (Gas Investment Costs for Hydrogen Fuel Cell Vehicles to 0 minus H2) Reach Cost Competitiveness FUEL COST Diff -20 (Gas minus H2) Examining the annual cash flows reveals the total invest- ment required for hydrogen HFCVs to reach “breakeven” -40 with gasoline ICEVs. These are shown in Figure 6.13: 2010 2015 2020 2025 2030 Year • The “CAP COST Diff” (dollars per year) is the differ- FIGURE 6.13  Cash flows for Case 1. ence in vehicle price for a gasoline vehicle versus a hydrogen vehicle, summed over all the new HFCVs sold that year. This Figure6-13.eps starts out negative (HFCVs cost a lot more than gasoline vehicles), but small (only a few HFCVs are sold). In the assumes that for a new fuel such as hydrogen with a small longer term, the annual cost difference continues to grow, HFCV fuel tank (approximately 5 to 8 gallons of gasoline as HFCVs are assumed to always cost more than gasoline equivalent), consumers would value fuel on a cost-per-mile cars. traveled basis (dollars per mile) rather than a cost-per-gallon- • “FUEL COST Diff” (dollars per year) is the annual equivalent basis, as they do now for gasoline. difference in fuel costs for HFCVs (counting all HFCVs cur- • “TOTAL Diff” (dollars per year) is the cash flow, rently in the fleet) compared to what would have been paid which equals the economy-wide cost per year of pursuing a to fuel comparable gasoline-fueled vehicles. Hydrogen soon fuel cell market introduction plan. The cash flow is defined becomes less costly as a fuel on a cents-per-mile basis, so this difference becomes positive around 2017. This analysis on many complex factors. Gasoline prices include federal, state, and lo- cal taxes. One could argue that hydrogen should be competed against the Hydrogen fuel becomes cost competitive with gasoline (on a cents-per- untaxed gasoline price. However, other alternative fuels such as ethanol are mile basis) in about 2017, when hydrogen costs are still fairly high, about untaxed to encourage their adoption, and the committee decided to give $5.60/kg. This is because the hydrogen vehicle is assumed to have a fuel hydrogen the same advantage. It should be noted that much of the revenue economy 2.0 times greater than the gasoline vehicle, and the gasoline price raised by gasoline taxes goes to highway maintenance and other necessary in the AEO high-oil-price case is $2.80 per gallon. This analysis compares functions that continue no matter what type of vehicles travel on them. As the cost of hydrogen with the price of gasoline. The committee decided discussed in Chapter 7, these revenues will have to be replaced from other this would be the most straightforward comparison because it would be sources if alternative fuels remain untaxed. On the other hand, the price of difficult to estimate a price for hydrogen without a model for all its uses in gasoline does not include the cost of externalities that hydrogen is intended the economy, and it is hard to estimate the cost of gasoline, which depends to address: CO2 emissions and oil imports.

OCR for page 73
82 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A focus on hydrogen as the gasoline vehicle minus the annual cost of the hydro- TABLE 6.6  Transition Costs and Timing for Hydrogen gen vehicle. This starts out negative, but becomes positive Cases in 2023, when the cost of hydrogen vehicles becomes less Breakeven Year than that of a similar number of gasoline vehicles. While the (annual cash flow > 0) 2023 HFCVs first cost remains higher than for the gasoline car, Cumulative life-cycle cost difference $22 billion the net fuel cost savings make the annual cash flow positive (between HFCV and gasoline reference at the breakeven year. car) to breakeven year • “CMLTV Diff” (dollars) is the cumulative cost differ- Cumulative vehicles first-cost $40 billion ence, the sum of the cost difference over time (starting in difference (between HFCV and (~$3.3 billion/yr) 2012), providing a yearly tally of the total funds that would gasoline reference car) to breakeven have to be invested to make HFCVs competitive. At first, year there is a negative cash flow (early HFCVs cost more than Number of HFCVs at breakeven year 5.6 (1.9% of fleet) gasoline cars), but eventually as HFCV and hydrogen fuel (millions) costs come down, the negative cash flow “bottoms out” in Hydrogen cost at breakeven year $3.3/kg 2023 at a minimum of about $22 billion, when about 5.6 mil- Hydrogen demand; number of 4,200 tonnes/d; 3,600 lion fuel cell vehicles have been produced. This minimum is hydrogen stations at LCC breakeven stations the “buydown” investment that must be supplied to bring the year HFCV to cost competitiveness. Total cost to build infrastructure for $8.2 billion demand at LCC breakeven year Most of the negative cash flow is due to the high price of Year when hydrogen fuel cost per 2016 the first few million fuel cell vehicles. This is not surpris- kilometer = gasoline price per ing, since, initially, fuel cell vehicles cost a lot more than kilometer gasoline vehicles (see Figure 6.6). The subsidy that might be needed by automakers or buyers is the sum of the dif- Hydrogen cost ($/kg) 5.20 ference in costs between HFCVs and gasoline cars, each Gasoline price ($/gal) 2.70 year between vehicle introduction in 2012 and life-cycle cost (LCC) breakeven in 2023. This cumulative difference   Total cost to build infrastructure to $0.5 billion (1,000 small in vehicle first cost for HFCVs (as compared to a reference   meet demand in 2023 (LCC   on-site SMR stations) gasoline vehicle) is about $40 billion (averaged over the   breakeven year) 2012-2023 buydown period, this is about $7,000 per car, or an average of $3.3 billion per year for 12 years). Transition dates and costs are summarized in Table 6.6, relative to a reference gasoline vehicle. RESULTS: COMPARISON OF GREENHOUSE The buydown cost is quite sensitive to assumptions GAS EMISSIONS AND OIL DISPLACEMENT FOR for key factors. For example, if fuel cell vehicles could SCENARIOS be introduced at their “learned-out cost” (e.g., the cost of HFCVs once they have become technically mature and are Assumed Greenhouse Gas Emissions for Fuels manufactured at large scale), buydown cost requirements Until 2020, all hydrogen comes from on-site SMRs with for vehicles would be greatly reduced, and fuel cell vehicles a CO2 release of 100 g CO2 equivalent per megajoule of would become competitive almost immediately. In this case, fuel. After that time, low-carbon sources such as biomass the primary transition cost would be building a hydrogen hydrogen and hydrogen from coal with carbon capture infrastructure to the point at which hydrogen is competitive and storage are phased in. By 2050, roughly 31 percent of as a fuel (fuel cost per kilometer), on the order of a billion hydrogen is produced via on-site SMRs, the remainder via dollars. Note that this would happen much sooner than the low-carbon sources (44 percent coal with CCS, 25 percent vehicles reaching cash flow breakeven (see bottom row in biomass H2). Thus, the overall emissions for hydrogen sup- Table 6.6). Box 6.1 explores the sensitivity of the results to ply in 2050 are 37 g CO2/MJ fuel, based on the CO2 values assumptions on HFCV fuel economy and incremental costs, given in Table 6.7, which shows the assumptions regarding and the cost of hydrogen and gasoline. the well-to-wheels emissions associated with different fuel supply pathways. In all cases, the carbon emissions from the hydrogen supply are assumed to follow the curve shown in Figure 6.14, where CO2 emissions are shown as declining linearly between 2020 and 2050. The average CO2 emissions might fall faster than this, because most new capacity after

OCR for page 73
HYDROGEN AND ALTERNATIVE TECHNOLOGIES FOR REDUCTION OF U.S. OIL USE AND CO2 EMISSIONS 83 TABLE 6.7  Assumed Greenhouse Gas Emissions per Unit 120 of Fuel Consumed 100 g CO2 eq per MJ fuel Fuel Greenhouse Gas Content of Fuels (g CO2 equivalent/MJ fuel used 80 LHV basis) 60 Conventional gasoline   90 (AEO 2007 projections to 2030 show this staying 40 approximately constant) Hydrogen from on-site 100 (TIAX, 2007) 20 reformation Hydrogen from coal with CCS,   16 (92% CO2 capture rate, 0 pipeline delivery assuming U.S. electricity mix)  2000 2010 2020 2030 2040 2050    0 (100% CO2 capture rate, Year assuming zero-carbon electricity is used for CO2 FIGURE 6.14  Greenhouse gas emissions from hydrogen supply compression) (adapted over time. from Ruether et al., 2005). Figure6-14.eps Coal mining, transport, and plant and mine construction emissions are estimated to be 200,000 Million gallons per year about 3.8 g CO2 eq/MJ H2, but this is counterbalanced by 150,000 GHG savings of about 4.3 g Case 1 (H2 CO2 eq/MJ, due to exported Success) electricity from the plant) 100,000 Reference Hydrogen from biomass,   10 (TIAX, 2007) pipeline delivery 50,000 Hydrogen from electrolysis    0 using zero-carbon 0 electricity (wind, hydro, 2000 2010 2020 2030 2040 2050 solar) Year Ethanol from corn   70 (22% reduction relative to gasoline) FIGURE 6.15  Case 1 gasoline consumption relative to the refer- Ethanol from cellulose   13 (85% reduction relative to ence case. gasoline) Figure6-15.eps SOURCES: Ruether et al. (2005); TIAX (2007). 1800 Million tonnes CO2 eq/yr 1600 1400 1200 Case 1 (H2 2025 is likely to be very low carbon in these scenarios (70 1000 Success) percent biomass hydrogen or coal hydrogen with CCS). 800 Reference 600 Case 1 (Hydrogen Success) 400 200 Oil Displacement 0 2000 2010 2020 2030 2040 2050 Figure 6.15 estimates gasoline consumption for the Year Hydrogen Success case and the reference case. Oil displace- ment is about 0.8 percent in 2020, rising to 24 percent in 2035 FIGURE 6.16  Case 1 greenhouse gas emissions relative to the and 69 percent in 2050. Although it takes several decades for reference case. hydrogen’s impact to be seen, beyond 2025 it enables grow- Figure6-16.eps ing reductions in both greenhouse gas emissions per year and annual oil use. Hydrogen may be important to achieve long-term stabilization goals requiring deep cuts in carbon Greenhouse Gas Reductions or oil use. Figures 6.15 and 6.16 show a dip in the reference case after about 2020. This occurs even though the increase Figure 6.16 shows the committee’s estimate of the reduc- in fuel economy of new cars levels off because the entire on- tions in greenhouse gas emissions for Case 1 (Hydrogen road fleet fuel economy increases as new efficient vehicles Success) relative to a reference case, where no hydrogen replace older vehicles. technologies are introduced. Greenhouse gas emissions per

OCR for page 73
84 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A focus on hydrogen BOX 6.1 Sensitivity of Breakeven Analysis Results to Changes in Assumptions The results of an analysis of the costs and timing of a hydrogen transition depend on many assumptions and inputs and are sensitive to changes in four important input parameters: 1. The fuel economy of an HFCV compared to a reference gasoline vehicle; 2. The incremental cost of an H2FCV compared to a gasoline reference vehicle; 3. The cost of hydrogen; and 4. The cost of gasoline (dollars per gallon). The values of these parameters for Case 1 (Hydrogen Success) are shown in Table 6.1.1. Also shown are potential high and low values for each parameter. Each parameter is varied over this range. Three key metrics that describe a hydrogen transition are: 1. The “breakeven year”; 2. The “breakeven cost” (e.g., cumulative cash flow to get to the breakeven year); and 3. The total capital cost (the incremental cost for HFCVs + the infrastructure cost) to get to the breakeven year. TABLE 6.1.1  Range Over Which Parameter Values Can Vary for Case 1 Parameter Low Value Case 1 Value High Value Fuel economy of HFCV versus fuel 1.3 2 3 economy of efficient gasoline car Incremental cost of HFCV compared 1,713 3,600 6,800 to reference gasoline vehicle (FC system = $50/kW; (FC system = $50/kW; (FC system = $62/kW; H2 H2 storage = $2/kWh) H2 storage = $10/kWh) storage = $18/kWh) Incremental cost of H2 compared to −$2/kg 0 $2/kg Case 1 long-term cost Price of oil 0.5 × high-price case (oil at Oil price from AEO 2008 high- 1.3 × high price case (oil at $40-$60/bbl in 2012-2030) price case (oil at $80-$120/bbl $105-$160/bbl in 2030) in transition period 2012-2030) NOTE: bbl = barrel; FC = fuel cell.

OCR for page 73
HYDROGEN AND ALTERNATIVE TECHNOLOGIES FOR REDUCTION OF U.S. OIL USE AND CO2 EMISSIONS 85 2040 Breakeven year 2035 Fuel Economy Ratio (FCV/Gasoline ICEV) 2030 FCV Price Increment $ The value for each of these three metrics is shown as a func- 2025 tion of the variables in a series of three “spider plots” (Figures H2 Cost Increment (from Case 1) $/kg 6.1.1-6.1.3). In each plot, the x-axis is the ratio of the variable 2020 Oil Price Multiplier (X 0 0.5 1 1.5 2 to its Case 1 value (shown in Table 6.1.1.). (The normal Case 1 AEO hp) Variable/Base value input is represented by a value of “1” on the x-axis. So, an HFCV incremental cost of $3,600 corresponds to an x-value of 1, while FIGURE 6.1.1 Sensitivity of breakeven year to changes in HCFV fuel economy, HFCV price, H2 cost, and gasoline price. an HFCV incremental cost of $6,800 corresponds to an x-value of 6,800/3,600 = 1.88.) When the input parameter is varied from its color FigureBox6.1-1.eps low to its high values, the value of x changes from 0.5 to 2. The metric varies as the input changes. This shows the sensitivity of the results to changes in the input variables. 100 As expected, the breakeven year is delayed and buydown costs 90 Fuel Economy Ratio 80 (FCV/Gasoline ICEV) are higher if the HFCV price is higher, the HFCV is less efficient, 70 FCV Price Increment Billion $ or hydrogen costs more than in Case 1. Breakeven occurs faster 60 $ 50 H2 Cost Increment and the buydown cost is less for higher oil prices. 40 (from Case 1) $/kg For example, if hydrogen costs $1/kg more than expected, 30 Oil Price Multiplier there is relatively little impact on the breakeven year or the transi- 20 (X AEO hp) 10 tion cost. However, if the cost of hydrogen is $2/kg higher than 0 expected, this delays breakeven by 12 years (from 2023 to 2035) 0 0.5 1 1.5 2 and raises the breakeven cost by almost a factor of three (from Input value/Base value about $23 billion to $61 billion). If the oil price is 1.3 times the AEO’s projected high-price case (e.g., about $100 to $160 per color FigureBox6.1-2.eps FIGURE 6.1.2 Sensitivity of buydown cost (billion dollars) to changes in barrel of oil in the transition period between 2012-2030), the HFCV fuel economy, HFCV price, H2 cost, and gasoline price. breakeven year is accelerated slightly. However, if oil prices drop to 70 percent of the AEO projections (e.g., about $55 to $85 per barrel of oil during the transition period 2012-2030), breakeven is Fuel Economy Ratio delayed 5 years (from 2023 to 2028) and the buydown cost rises (FCV/Gasoline 600 from $23 billion to $31 billion. If oil prices drop to 50 percent of ICEV) 500 FCV Price the AEO high-price case ($40 to $60 per barrel during the transi- Increment $ tion), breakeven is delayed further to 2035, and the buydown cost 400 Billion $ almost triples to $61 billion. 300 H2 Cost 200 Increment (from Case 1) $/kg 100 0 Oil Price Multiplier (X 0 0.5 1 1.5 2 AEO hp) Variable/Base Value FIGURE 6.1.3 Sensitivity ofFigureBox6.1-3.eps color capital investment to breakeven year (incre- mental price of HFCVs + H2 infrastructure capital, billion dollars).

OCR for page 73
86 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A focus on hydrogen year are reduced by about 0.7 percent in 2020, 19 percent 70 in 2035, and 60 percent in 2050 compared to the reference Fuel economy (mpg) 60 case. 50 40 Gasoline ICEV Case 2 (ICEV Efficiency) 30 Gasoline HEV 20 Case 2 is an “evolutionary,” not revolutionary, scenario. 10 The committee assumes that currently available improve- 0 ments in gasoline internal combustion engine technology 2000 2010 2020 2030 2040 2050 are used to improve fuel economy (rather than power and Year acceleration). A range of more efficient advanced gasoline technolo- FIGURE 6.18  Case 2 assumed on-road fuel economy for new gies could be implemented in 2010-2035 as described in Figure6-18.eps gasoline ICEVs and gasoline hybrid ICEVs over time. Chapter 4. In this scenario, a new “high-fuel-economy” gasoline vehicle is introduced, as well as a hybrid gasoline vehicle, and these capture growing market share over time. By 2035 (2050), 42 percent (85 percent) of new LDVs and trains could offer an additional 15 percent reduction in fuel 30 percent (60 percent) of the fleet are gasoline hybrids, and consumption and CO2 emissions over advanced conventional the remaining non-hybrid cars have high fuel economy. This spark ignition power trains and have cost advantages over is shown in Figure 6.17. hybrid electric vehicles (see, for example, Adrian, 2004). In In 2010, the new gasoline vehicle is assumed to have an a high-fuel-cost environment, they could become a growing on-road fuel economy of 22.2 mpg, the hybrid 31.9 mpg. fraction of LDV sales with the some shifts in government (These values are selected to match the reference case up to positions on diesels and a positive public relations program. 2010.) The fuel economy of each vehicle is then assumed to Thus, to the extent that diesels can penetrate the market, this improve as follows and discussed in Chapter 4: scenario may understate potential fuel savings. The same vehicle stock model used in the reference case • 2.6 percent per year from 2010 to 2025, keeps track of the vehicle numbers and vintages of advanced • 1.7 percent per year from 2026 to 2035, and gasoline cars and gasoline hybrids on the road in any year. • 0.5 percent per year from 2035 to 2050. This allows calculation of oil consumption and greenhouse gas emissions for each year. The on-road new car fuel economy over time is plotted in Figure 6.18. Note that this is similar to the reference case Oil Displacement in Figure 6.4 up to about 2020. Beyond this, Case 2 is sig- nificantly more efficient; by 2050, gasoline ICEV and hybrid Gasoline consumption for the case above is estimated in cars are about 35 percent more efficient than in the reference Figure 6.19. Improving fuel economy is a very effective way case, which incorporates the new CAFE standards. to cut gasoline use. Gasoline consumption in 2020 is only The committee did not project increased market share for slightly reduced relative to the reference case, which includes diesel engines in this scenario because of the uncertainty over rapidly improving fuel economy, but in 2035 it is down by the costs of meeting future tailpipe emission specifications 35 billion gallons per year (25 percent), and in 2050, by 64 and consumer acceptance, considering the poor history of billion gallons per year (40 percent). diesels in U.S. automobiles. However, advanced diesel power Greenhouse Gas Reductions 400 350 Greenhouse gas emissions show a similar trend (Figure Million vehicles 300 6.20). Fuel economy improvements can yield increasing 250 Gasoline ICEV reductions in greenhouse gases. Greenhouse gas emissions 200 Gasoline HEV 150 are reduced by about 24 million tonnes of CO2 equivalent per TOTAL 100 year (1.7 percent) by 2020, 385 million tonnes (25 percent) 50 by 2035, and 700 million tonnes (41 percent) by 2050. 0 Based on projected gasoline prices and cost estimates 2000 2010 2020 2030 2040 2050 for improved fuel economy, it appears that gasoline hybrids Year and advanced gasoline vehicles would pay for themselves on a life-cycle cost basis, so no external subsidy should FIGURE 6.17  Case 2 assumed market penetration for gasoline be needed. A simple calculation shows that increasing fuel ICEVs and advanced gasoline HEVs. Figure6-17.eps economy from 30 to 45 mpg in a car that travels 15,000

OCR for page 73
HYDROGEN AND ALTERNATIVE TECHNOLOGIES FOR REDUCTION OF U.S. OIL USE AND CO2 EMISSIONS 87 180,000 miles a year saves 167 gallons of gasoline per year. If a car Million gallons per year 160,000 is kept 10 years and gasoline costs $2.5 per gallon, the pres- 140,000 120,000 Case 2 (ICEV ent value of fuel savings amounts to $2,565 (assuming a 10 100,000 Efficiency) percent discount rate). This would be enough to pay for the 80,000 Reference difference in first cost between a conventional gasoline car 60,000 and a hybrid, which is estimated to be $1,800-$2,500 per car 40,000 20,000 (Kromer and Heywood, 2007). 0 1990 2010 2030 2050 Case 3: Biofuels Year Case 3, Biofuels, considers an emphasis on biofuels in FIGURE 6.19  Gasoline consumption for Case 2 and for the refer- the transportation sector. Assumptions about the introduction ence case. Figure6-19.eps of biofuels are summarized in Table 6.8. Annual production levels for various biofuels are plotted in Figure 6.21. The 1800 detailed assumptions are discussed in Chapter 4. For refer- Million tonnes CO2 eq/yr 1600 ence, the maximum practicable case, in the committee’s 1400 judgment, is 700 million dry tons of sustainable biomass 1200 Case 2 (ICEV 1000 Efficiency) available in 2050, with all of the biomass used for cellulosic 800 Reference ethanol production at 90 gallons per dry ton. This would total 600 63 billion gallons of ethanol per year in 2050, the amount 400 assumed in Case 3. 200 The AEO 2007 reference case includes about 12 billion 0 gallons of corn ethanol by 2030 plus an additional 3 billion 2000 2010 2020 2030 2040 2050 gallons of cellulosic ethanol. The committee extended this to Year assume that in the reference case, cellulosic ethanol produc- tion reaches 9 billion gallons per year by 2050. The assumed FIGURE 6.20  Greenhouse gas emissions for Case 2 and for the reference case. Figure6-20.eps biofuels use in the reference case is shown in Figure 6.5. TABLE 6.8  Assumed Biofuel Use in Case 3 Corn Ethanol Cellulosic Ethanol Biobutanol Biodiesel F-T Diesel via Biomass Gasification Production Timing of introduction similar to that (billion gallons) for biomass H2, toward end of scenario time frame • 2002 • 2006 2.5 • 2008 6 0.25 • 2010 8 1 • 2015 10 6 0.6 (2012) • 2020 12 (max = 30% of 16 0.1 expanded corn crop) • 2025 28 0.6 1.5 (max = 30% of soy • 2030 36 2.6 crop, limited by land) • 2035 12 44 4.0 • 2050 63 CO2 reduction 22% relative to 85% relative to 80% relative to regular (energy equivalent gasoline gasoline diesel basis) Oil reduction 1 gallon gasoline 1 gallon gasoline 1 gallon biodiesel (energy equivalent equivalent ethanol equivalent ethanol replaces 0.95 gallon oil basis)a replaces 0.96 gallon replaces 0.93 gallon oil oil aGasoline = 119,000 Btu/gal, ethanol = 80,000 Btu/gal, and biodiesel is equivalent to regular diesel.

OCR for page 73
88 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A focus on hydrogen 70 45 Billion gallons per year 40 Billion gallons per year 60 35 50 Corn Ethanol 30 Cellulosic EtOH 40 Cellulosic EtOH 25 Biobutanol 30 Biobutanol 20 Biodiesel 20 Biodiesel 15 TOTAL 10 10 0 5 0 2000 2010 2020 2030 2040 2050 2000 2010 2020 2030 2040 2050 Year Year FIGURE 6.21  Annual production of biofuels assumed for Case 3. FIGURE 6.23  Case 3: Oil displacement relative to the reference color Figure6-21.eps case. color Figure6-23.eps The reference case corresponds to about 10 percent etha- 450 nol by volume in gasoline by 2050. In Case 3 (Biofuels), this Million tonnes CO2 eq/yr 400 Cellulosic EtOH is expanded to include an additional 54 billion gallons of 350 high biofuels per year (beyond the reference case). This is roughly 300 Biobutanol four times the biofuel use in the reference case or about 40 250 percent ethanol in gasoline plus some other advanced bio- 200 Biodiesel fuels. It is important to note that the greenhouse gas and 150 oil reductions shown for Case 3 are relative to a reference 100 TOTAL high case that already includes biofuel use up to 10 percent of 50 gasoline by volume. 0 In the Biofuels case, the assumed corn ethanol use is the 2000 2010 2020 2030 2040 2050 same as in the reference case. The difference is that more Year cellulosic and other advanced biofuels are produced. The additional biofuels production assumed in the Biofuels case FIGURE 6.24  Case 3: Greenhouse gas emission reductions relative (compared to the reference case) is shown in Figure 6.22. to the reference case. color Figure6-24.eps Gasoline displacement for Case 3 is shown in Figure 6.23. The total is about 12 billion gallons per year by 2020 and 39 billion gallons per year in 2050. Comparison of Scenarios Greenhouse gas emissions reductions are given in Figure 6.24 for the Biofuels case. The total reduction in greenhouse The estimated savings in gasoline use and greenhouse gas emissions is about 8 percent from the reference case by gas emissions for each case are plotted in Figures 6.25 and 2020, rising to 23 percent by 2050. The committee has not 6.26. In the near to mid term, improving the fuel economy estimated the costs of building biofuel production plants or of gasoline vehicles will be the most effective option for changes in the fuel distribution infrastructure that might be reducing oil use and greenhouse gas emissions. This is needed. already incorporated in the reference case up to 2020, but continued improvements thereafter could match the savings from hydrogen until about 2035. Biofuels could begin to make a difference sooner than hydrogen, which takes more time to implement, assuming cellulosic ethanol comes online 60 in 2010. After about 2032, however, Case 1 (Hydrogen Suc- Additional Corn cess) would lead to greater greenhouse gas reductions per Billion gallons Ethanol per year 40 Additional Cellulosic year than Case 3 (Biofuels). By 2040, the Hydrogen Success EtOH 20 scenario offers about twice the greenhouse gas reduction and Biobutanol 0 oil savings per year as the Biofuels scenario, and by 2050, 2000 2010 2020 2030 2040 2050 Biodiesel almost three times the reduction. This clearly illustrates the Year time frames for different technologies and the total contri- butions they might make by 2020 and beyond. Although FIGURE 6.22  Case 3: Added biofuel production relative to the efficiency and biofuels could contribute sooner, hydrogen reference case. would surpass the annual savings achievable with either after

OCR for page 73
HYDROGEN AND ALTERNATIVE TECHNOLOGIES FOR REDUCTION OF U.S. OIL USE AND CO2 EMISSIONS 89 180,000 combined case than for HFCVs alone. By 2020, efficiency Million gallons per year 160,000 Case 1 (H2 reduces greenhouse gas emissions by about 1.8 percent rela- 140,000 Success) tive to the base case. Beyond 2030, HFCVs lead to deeper 120,000 Case 2 (ICEV Eff) 100,000 cuts in emissions than would be possible with efficiency 80,000 Case 3 (Biofuels) alone. 60,000 40,000 Reference 20,000 Case 3 + Case 2: Biofuels + ICEV Efficiency 0 2000 2010 2020 2030 2040 2050 Combining higher gasoline vehicle efficiency with biofu- els yields much greater reductions in oil use and greenhouse Year gas emissions than are possible with biofuels alone. This FIGURE 6.25  Oil consumption for Cases 1-3 compared. is shown in Figures 6.29 and 6.30, which combine Cases 2 and 3. By 2020, biofuel use alone could reduce annual oil color Figure6-25.eps use by about 8 percent, with efficiency bringing the total 2000 to 10 percent. In the longer term, the effect of efficiency improvements dominates, with biofuels saving 23 percent of gasoline use or greenhouse gas emissions in 2050 and Million tonnes 1500 Case 1 (H2 Success) CO2 eq/yr Case 2 (ICEV Eff) efficiency an additional 41 percent. This strategy “stretches” 1000 limited biomass resources to fuel more vehicle miles traveled Case 3 (Biofuels) Reference per acre of land. 500 0 2000 2010 2020 2030 2040 2050 Year FIGURE 6.26  Greenhouse gas emissions for Cases 1-3 compared. 180,000 Million gallons per year color Figure6-26.eps 160,000 Case 1 + Case 140,000 2 (H2 Success 120,000 and ICEV 100,000 Efficiency) Reference 80,000 2035. This result highlights the long time constants inherent 60,000 in changing the energy system, as well as the need to develop 40,000 20,000 long-term, very low carbon options. 0 2000 2010 2020 2030 2040 2050 COMBINED APPROACHES TO REDUCING Year GREENHOUSE GAS EMISSIONS FIGURE 6.27  Oil use for Cases 1 and 2 combined. AND OIL USE Figure6-27.eps The previous section looked at the potential impact of implementing one technological measure at a time. In the 1800 future, reducing oil use and greenhouse gas emissions prob- Million tonnes CO2 eq/yr 1600 ably will be increasingly important, and it is likely that a 1400 Case 1 + Case 2 combination of approaches would be implemented. However 1200 (H2 Success and the reductions for each approach cannot simply be added 1000 ICEV Efficiency) because they affect each other. This section considers several 800 Reference combinations. 600 400 200 Case 1 + Case 2: HFCVs + ICEV Efficiency 0 2000 2010 2020 2030 2040 2050 This case combines higher gasoline vehicle efficiency Year with introduction of hydrogen fuel cell vehicles (Case 1 and Case 2). The results are shown in Figures 6.27 and 6.28. FIGURE 6.28  Greenhouse gas emissions with HFCVs for Cases Comparing Figure 6.27 and 6.25, gasoline consumption 1 and 2 combined. Figure6-28.eps in 2035 is about 18 billion gallons per year lower for the

OCR for page 73
90 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A focus on hydrogen Case 2 (ICEV 400 200,000 Million gallons per year Efficiency) Million vehicles 300 Gasoline ICEV 150,000 Case 3 (Biofuels) Gasoline HEV 200 100,000 Hydrogen FCV 100 TOTAL 50,000 Case 3 + Case 2 (Biofuels + ICEV 0 0 Efficiency) 2000 2010 2020 2030 2040 2050 2000 2010 2020 2030 2040 2050 Reference Year Year FIGURE 6.31  Assumed number of vehicles in the fleet for Case 4. FIGURE 6.29  Oil use for Cases 2 and 3 combined. color Figure6-31.eps color Figure6-29.eps 180,000 Million gallons per year 1800 160,000 Case 2 (ICEV 140,000 1600 Million tonnes CO2 /yr Efficiency) 120,000 1400 Case 4 (Portfolio) 100,000 1200 Case 3 (Biofuels) 80,000 Reference 1000 60,000 800 40,000 600 Case 3 + Case 2 20,000 400 (Biofuels + ICEV 0 200 Efficiency) 2000 2010 2020 2030 2040 2050 0 Reference Year 2000 2010 2020 2030 2040 2050 Year FIGURE 6.32  Oil use in million gallons per year for Case 4. Figure6-32.eps FIGURE 6.30  Greenhouse gas emission reductions for Cases 2 and 3 combined. color Figure6-30.eps Case 4 (Portfolio): Implement Efficient ICEVS plus Biofuels and Hydrogen FCVs TABLE 6.9  Gasoline Displacement for Cases 1-4 Compared to Reference Case Case 4 combines all three of the major options discussed Billion Gallons Gasoline Saved per Year (% Saved) above. Figure 6.31 shows the assumed numbers of vehicles in the fleet over time. Note that the number of hydrogen Case 2020 2035 2050 vehicles is the same as in Case 1 (Hydrogen Success) and Case 1 (Hydrogen 1.0 (0.8%) 34 (24%) 109 (69%) the number of gasoline ICEVs is the same as in Case 2 Success) (ICEV Efficiency). The number of advanced gasoline ICEVs Case 2 (ICEV 2.2 (1.7%) 35 (25%) 64 (41%) (hybrids) increases, but eventually loses market share to Efficiency) HFCVs. Figure 6.32 shows the reduction in petroleum consump- HFCVs + ICEV 3.0 (2.2%) 55 (39%) 125 (80%) tion over time for Case 4. Gasoline use is virtually eliminated Efficiency in the light duty vehicle fleet by 2050. Table 6.9 shows the Case 3 (Biofuels) 12 (9%) 28 (20%) 39 (25%) reduction in gasoline use for the four cases relative to the reference case. Case 3 + Case 2 14 (11%) 64 (45%) 103 (66%) Table 6.10 lists the emissions reductions relative to the Biofuels + ICEV reference case for the four cases. Emissions of greenhouse Efficiency gas over time are shown in Figure 6.33. The cumulative Case 4: 15 (11%) 83 (59%) 157 (100%) impact of reductions is shown in Figure 6.34. With a com- ICEV Efficiency bined approach including efficiency, biofuels, and hydrogen (Case 2) + Biofuels fuel cells, it is possible to reduce CO2 emissions by about 90 (Case 3) + percent and gasoline use by 99 percent by 2050. Hydrogen (Case 1)

OCR for page 73
HYDROGEN AND ALTERNATIVE TECHNOLOGIES FOR REDUCTION OF U.S. OIL USE AND CO2 EMISSIONS 91 TABLE 6.10  Greenhouse Gas Emission Reductions for CONCLUSIONS Cases 1-4 Compared to Reference Case CONCLUSION: In the judgment of the committee, the Million Tonnes CO2 Equivalent Avoided (% Avoided) maximum practicable number of HFCVs that could be on Case 2020 2035 2050 the road by 2020 is around 2 million. Subsequently, this Case 1   10 (0.7%) 295 (19%) 1,026 (60%) number could grow rapidly to as many as 60 million by (Hydrogen 2035 and more than 200 million by midcentury, but such Success) rapid and widespread deployment will require continued technical success, cost reductions from volume produc- Case 2 (ICEV   24 (1.7%) 385 (25%)    700 (41%) Efficiency) tion, and government policies to sustain the introduction   of HFCVs into the market during the transition period HFCVs 26 (1.8%) 475 (31%) 1,123 (66%) needed for technical progress. + ICEV Efficiency CONCLUSION: While it will take several decades for Case 3 118 (8%) 281 (18%)    386 (23%) HFCVs to have major impact, under the maximum prac- (Biofuels) ticable scenario fuel cell vehicles would lead to significant reductions in oil consumption and also significant reduc- Case 3 + 143 (10%) 666 (44%) 1,086 (64%) tions in CO2 emissions if national policies are enacted to Case 2 restrict CO2 emissions from central hydrogen production Biofuels + ICEV plants. Efficiency CONCLUSION: The unit costs of fuel cell vehicles and Case 4: 130 (9%) 747 (49%) 1,505 (88%) Hydrogen hydrogen in the Hydrogen Success scenario—the maxi- (Case 1) + mum practicable case—decline rapidly with increasing ICEV vehicle production, and by 2023 the cost premium for Efficiency HFCVs relative to conventional gasoline vehicles is pro- (Case 2) jected to be fully offset by the savings in fuel cost over the + Biofuels (Case 3) life of the vehicle relative to a reference case based on the EIA high-oil-price scenario. At that point, according to 1800 the committee’s analysis, HFCVs become economically 1600 competitive in the marketplace. Million tonnes CO2 eq/yr 1400 1200 Fully implementing the maximum practicable hydrogen case by 2050 would require construction of approximately 1000 Case 4 (Portfolio) 80,000 on-site distributed natural gas reforming units, 80 800 Reference coal gasification plants of 500 MW (electrical equivalent) 600 with CCS, 130 biomass gasification plants (each 100 MW 400 equivalent) with associated biomass growth and collection 200 farms, and roughly 80,000 miles of pipelines for hydrogen 0 supply and CCS. The committee estimates that more than 2000 2010 2020 2030 2040 2050 $400 billion would be required to fully build out hydrogen Year supply to fuel the HFCVs by 2050. FIGURE 6.33  Greenhouse gas emissions for Case 4 (combination The committee’s analysis indicates that at least two alter- Figure6-33.eps of HFCVs, efficiency, and biofuels). natives to HFCVs—advanced conventional vehicles and bio- fuels—have the potential to provide significant reductions in 30,000 Million tonnes CO2 eq 25,000 Case 2 (ICEV projected oil imports and CO2 emissions. However, the rate 20,000 Efficiency) of growth of benefits from each of these two measures slows 15,000 after two or three decades, toward the end of the committee’s Case 3 + Case 2 10,000 (Biofuels + ICEV analysis period, while the growth rate of projected benefits 5,000 Efficiency) from fuel cell vehicles is still increasing. The deepest cuts in 0 Case 4 (Portfolio) oil use and CO2 emissions after about 2040 would be from 2000 2010 2020 2030 2040 2050 hydrogen. Year Over the next 20 years, the greatest impact on U.S. oil and CO2 reductions would result from implementing existing FIGURE 6.34  Cumulative reduction of greenhouse gas emissions Figure6-34.eps for Case 2, Case 3 plus Case 2, and Case 4.

OCR for page 73
92 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A focus on hydrogen evolutionary vehicles focused on vehicle design to deliver Nicholas, M.A., S.L. Handy, and D. Sperling. 2004. Using Geographic efficiency improvements. Information Systems to Evaluate Siting and Networks of Hydrogen Stations. Transportation Research Record 1880:126-134. The potential of biofuels under the committee’s maximum NRC (National Research Council). 2008. Review of the Research Program practicable approach achieves a 23 percent reduction in CO2 of the FreedomCAR and Fuel Partnership: Second Report. Washington, and gasoline use by 2050, but has only a small impact prior D.C.: The National Academies Press. to 2035, compared to the reference case. Paster, M. 2006. Hydrogen Delivery Options and Issues. Presented at the USDOE Hydrogen Transition Analysis Workshop, Washington, D.C., August 9-10. Bibliography Perlack, R., et al. 2005. Biomass as Feedstock for a Bioenergy and Bioprod- ucts Industry: The Technical Feasibility of a Billion-Ton Annual Supply. Adrian, M. 2004. Advanced Electronics and Control Technologies for Fuel- DOE-USDA, Washington, D.C. efficient, Low-emission Diesel Powertrains. Doc. No. 2004-21-0083. Ruether, J., M. Ramezan, and E. Grol. 2005. Life-Cycle Analysis of Washington, D.C.: Greaney-Ricardo, Inc. Greenhouse Gas Emissions for Hydrogen Fuel Production in the United Argonne National Laboratory. 2007. The VISION Model. Available at http:// States from LNG and Coal. National Energy Technology Laboratory, www.transportation.anl.gov/software/VISION/index.html. Pittsburgh, Pa. DOE (Department of Energy). 2005. Life-Cycle Analysis of Greenhouse Sawyer, R. 2007. California’s Regulations to Control Greenhouse Gas Emis- Gas Emissions for Hydrogen Fuel Production in the United States from sions from Motor Vehicles. Presentation at the Hearing on Request for LNG and Coal. DOE/NETL-2006/1227, Table 5. National Energy Waiver of Preemption under Clean Air Act Section 209(b) by Robert Technology Laboratory, Pittsburgh, Pa. Sawyer, Chair, California Air Resources Board, Washington, D.C. DOE. 2007. H2A Hydrogen Analysis Model. Available at http://www. Singh, M., A. Vyas, and E. Steiner. 2003. VISION Model: Description of hydrogen.energy.gov/ h2a_analysis.htm. Model Used to Estimate the Impact of Highway Vehicle Technologies EIA (Energy Information Administration). 2008. Annual Energy Outlook and Fuels on Energy Use and Carbon Emissions to 2050. Center for with Projections to 2030. Report DOE/EIA-0383. Washington, D.C. Transportation Research, Argonne National Laboratory, Argonne, Ill. Greene, D., P. Leiby, and D. Bowman. 2007. Integrated Analysis of Market TIAX. 2007. Full Fuel Cycle Assessment: Well to Tank Energy Input, Transformation Scenarios with HyTrans. Oak Ridge National Labora- Emissions and Water Impacts. California Energy Commission Report, tory, Oak Ridge, Tenn. CEC-600-2007-002-D, February. Gronich, S. 2007. 2010-2025 Hydrogen Scenario Analysis. Presentation to Yang, C., and J. Ogden. 2007a. Determining the Lowest-cost Hydrogen the committee, February 20. Delivery Mode. International Journal of Hydrogen Energy 32(2):268- Kromer, M.A., and J.B. Heywood. 2007. Electric Powertrains: Opportuni- 286. ties and Challenges in the U.S. Light-Duty Vehicle Fleet. MIT Report Yang, C., and J. Ogden. 2007b. U.S. Urban Hydrogen Infrastructure Costs LFEE 2007-02 RP, May. Cambridge, Mass. Using the Steady State City Hydrogen Infrastructure System Model Melendez, M. 2006. Geographically Based Hydrogen Demand & Infrastruc- (SSCHISM). Presentation at the 2007 National Hydrogen Association ture Analysis. Presented at the USDOE Hydrogen Transition Analysis Meeting, San Antonio, Texas, March 18-22. (A beta copy of the model Workshop, Washington, D.C., August 9-10. is posted on Christopher Yang’s website at University of California Nicholas, M.A., and J.M. Ogden. 2007. Detailed Analysis of Urban Sta- at Davis Institute of Transportation Studies, http://www.its.ucdavis. tion Siting for California Hydrogen Highway Network. Transportation edu/people.) Research Record 1983:121-128.