3
Hydrogen Technology

This chapter reviews the hydrogen production and fuel cell vehicle technologies that will be needed to establish a budget roadmap to achieve the “maximum practicable” goal spelled out in Chapters 6 (“U.S. Carbon Dioxide (CO2) and Oil Reductions from Hydrogen and Alternatives”) and 7 (“A Budget Roadmap,” which covers the R&D, vehicle, and infrastructure costs entailed in moving to a self-sustaining hydrogen-based system). It details how the hydrogen production and delivery infrastructure might grow to meet the demand for hydrogen fuel and lays out a picture of how the future hydrogen infrastructure could evolve to support the growing number of hydrogen fuel cell vehicles (HFCVs), referred to as the maximum practicable rate (MPR) of penetration, developed and discussed in Chapter 6. In the MPR case, HFCVs are introduced starting with a few thousand in 2012, growing to a fleet of 2 million in 2020, 60 million in 2035, and 220 million in 2050.

In assessing these developments, the committee analyzed hydrogen production, delivery, and dispensing technologies and evaluated fuel cell and hydrogen storage. In doing so, its members met representatives of major hydrogen fuel and auto companies that have announced hydrogen programs, participated in studies of the National Research Council (NRC) Committee on the Review of the FreedomCAR and Fuel Partnership. (To allow for greater interaction, the two committees shared two members.) It also met with U.S. Department of Energy (DOE) officials to review its technical programs, conducted a detailed review of the literature with the help of a consultant, and reviewed publicly available sources that have a history of technology evaluations and/or product introductions.

Resource calculations in Chapters 6 and 7 include costs of both the HFCVs and the hydrogen infrastructure (for production, delivery, and dispensing), and this chapter includes estimated cost data for key technologies. The cost of the hydrogen fuel cell and onboard hydrogen storage system on the HFCV will be critical to its competitiveness with alternative vehicle technologies described in Chapter 4. The cost and technical status of hydrogen producing and delivering technologies from various primary feedstocks determined which technologies were included in the roadmap. This chapter focuses on the least expensive, most fully developed technologies with potential for low well-to-wheels (WTW) carbon dioxide (CO2) release that are the most practical for different stages of the move to hydrogen. Using the criteria of cost and stage of technological advancement in this screening process effectively ruled out some technologies that are in early development phases or have high current costs but may have good potential for improvement. This fact does not imply that these technologies are unimportant or will not play roles in the future. The committee’s task included estimating the costs of building the infrastructure to meet growing hydrogen demand during the transition to a self-sustaining hydrogen transportation system (i.e., one without subsidies). The technologies chosen represent just one scenario of what could happen.

This chapter addresses the readiness of the hydrogen production and delivery and HFCV technologies to begin to meet the MPR case. Much of the current development of HFCVs is aimed at the year 2015 for the most crucial early-stage technologies to be at the point at which decisions to move to mass commercialization could be made. This focus is consistent with the MPR case.

HYDROGEN PRODUCTION AND DELIVERY

Hydrogen can be produced from various resources, either in small facilities at the point of sale or in larger facilities farther away and requiring delivery. Getting to a self-sustaining market will require an evolution of the supply infrastructure to ensure the lowest possible costs and fuel availability at all times. The long-term vision of the hydrogen production and supply infrastructure is to produce large amounts of hydrogen from domestic resources with low WTW CO2 emissions and deliver this to customers at a cost that is competitive with gasoline on a cost-per-mile traveled basis. The report



The National Academies | 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 31
3 hydrogen Technology This chapter reviews the hydrogen production and fuel and technical status of hydrogen producing and delivering cell vehicle technologies that will be needed to establish a technologies from various primary feedstocks determined budget roadmap to achieve the “maximum practicable” goal which technologies were included in the roadmap. This spelled out in Chapters 6 (“U.S. Carbon Dioxide (CO2) and chapter focuses on the least expensive, most fully developed Oil Reductions from Hydrogen and Alternatives”) and 7 technologies with potential for low well-to-wheels (WTW) (“A Budget Roadmap,” which covers the R&D, vehicle, and carbon dioxide (CO2) release that are the most practical for infrastructure costs entailed in moving to a self-sustaining different stages of the move to hydrogen. Using the criteria of hydrogen-based system). It details how the hydrogen pro- cost and stage of technological advancement in this screen- duction and delivery infrastructure might grow to meet the ing process effectively ruled out some technologies that are demand for hydrogen fuel and lays out a picture of how the in early development phases or have high current costs but future hydrogen infrastructure could evolve to support the may have good potential for improvement. This fact does not growing number of hydrogen fuel cell vehicles (HFCVs), imply that these technologies are unimportant or will not play referred to as the maximum practicable rate (MPR) of pen- roles in the future. The committee’s task included estimat- etration, developed and discussed in Chapter 6. In the MPR ing the costs of building the infrastructure to meet growing case, HFCVs are introduced starting with a few thousand in hydrogen demand during the transition to a self-sustaining 2012, growing to a fleet of 2 million in 2020, 60 million in hydrogen transportation system (i.e., one without subsidies). 2035, and 220 million in 2050. The technologies chosen represent just one scenario of what In assessing these developments, the committee analyzed could happen. hydrogen production, delivery, and dispensing technologies This chapter addresses the readiness of the hydrogen and evaluated fuel cell and hydrogen storage. In doing so, production and delivery and HFCV technologies to begin its members met representatives of major hydrogen fuel and to meet the MPR case. Much of the current development of auto companies that have announced hydrogen programs, HFCVs is aimed at the year 2015 for the most crucial early- participated in studies of the National Research Council stage technologies to be at the point at which decisions to (NRC) Committee on the Review of the FreedomCAR and move to mass commercialization could be made. This focus Fuel Partnership. (To allow for greater interaction, the two is consistent with the MPR case. committees shared two members.) It also met with U.S. Department of Energy (DOE) officials to review its techni- hYdroGeN ProdUcTioN aNd deliVerY cal programs, conducted a detailed review of the literature with the help of a consultant, and reviewed publicly available Hydrogen can be produced from various resources, either sources that have a history of technology evaluations and/or in small facilities at the point of sale or in larger facilities far- product introductions. ther away and requiring delivery. Getting to a self-sustaining Resource calculations in Chapters 6 and 7 include costs market will require an evolution of the supply infrastructure of both the HFCVs and the hydrogen infrastructure (for pro- to ensure the lowest possible costs and fuel availability at all duction, delivery, and dispensing), and this chapter includes times. The long-term vision of the hydrogen production and estimated cost data for key technologies. The cost of the supply infrastructure is to produce large amounts of hydro- hydrogen fuel cell and onboard hydrogen storage system on gen from domestic resources with low WTW CO2 emissions the HFCV will be critical to its competitiveness with alter- and deliver this to customers at a cost that is competitive native vehicle technologies described in Chapter 4. The cost with gasoline on a cost-per-mile traveled basis. The report 

OCR for page 31
 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A FOCUS ON HyDROGEN region need to offer hydrogen to provide adequate coverage. The Hydrogen Economy (NRC, 2004) indicated that this As more HFCVs are sold, the infrastructure expands, eventu- long-term potential could be possible once a large number ally to mid- and late transition supply options discussed in of HFCVs are in the market and a mature hydrogen industry the next two sections. This is then repeated in more cities. is in place. This section discusses the transitional period and This concept is more fully explained in Chapter 6, as is a list lays out a picture of what this might look like. The technolo- of possible cities. gies discussed are those that are the most well developed and The demand for hydrogen at these stations will be very nearest to technical and economic readiness. These are the low for the first several years of operation because there will technologies that are included in the resource estimates in be an excess of stations for the few HFCVs in the market. All Chapter 6, along three alternative time frames: of the hydrogen stations will be underutilized for a period of years. During the first several years of operation there will 1. Early hydrogen delivery for first vehicle owners— be fewer than 10 HFCV fills (at perhaps 5 kg each) per sta- supplied from existing industry excess capacity or small tion per day on average. At very low demand the technology skid-mounted production appliances at existing gasoline choices to provide hydrogen include truck delivery of hydro- stations gen from current industrial hydrogen gas suppliers or from 2. Midtransition vehicle owners—supplied mainly by excess hydrogen production at some refineries and chemical on-site production at full-size refueling stations plants directly to the filling station where it is stored, much as 3. Late transition to self-sustaining hydrogen transporta- in today’s gasoline station. Where hydrogen is not available, tion system—supplied mainly in large central production small skid-mounted natural gas reformers or small water facilities and delivered by pipeline to refueling stations electrolysis systems could be used to generate hydrogen at the refueling station. As demand grows, these small units early hydrogen infrastructure might be moved to new areas and replaced by larger facilities. The initial hydrogen cost for all of these options is high, but Before HFCVs can be marketed generally, there must be as the hydrogen sales volume increases the cost will decline. enough fueling facilities to convince buyers that fuel will be The hydrogen sales volume might increase quickly during available when and where they need it. Based on the anal- this early stage if hydrogen-powered internal combustion ogy with diesel fuel, between 10 and 25 percent of stations engine vehicles (ICEVs) as well as HFCVs were also fill- providing an alternate fuel gives customers confidence in fuel ing at these stations. Box 3.1 contains a discussion of the availability. Meeting this standard for early hydrogen vehicle hydrogen ICEV. users will be very challenging because hydrogen fueling To illustrate this point, for a 500 kg/d natural gas reformer equipment is much more expensive than traditional liquid (about one-third the size of a full commercial-scale unit fuels. Reducing the number of locations while still providing needed for a full-size refueling station) producing hydrogen good coverage will be important to hold down capital costs at a station and dispensing at 5,000 pounds per square inch and to increase the hydrogen volume sold at each location. (psi), the hydrogen cost will be $3.50/kg when operating at Recent estimates (Nicholas et al., 2004) are that as few as 5 full capacity (70 percent). Full capacity will provide about percent of stations offering hydrogen might satisfy customer 70 refuelings per day. For this station in the very early years concerns if their locations are closely coordinated to provide when there are only 10 refuelings per day, the hydrogen cost adequate coverage of urban and suburban areas. This very close coordination in locating stations is a new concept increases to $7.70/kg. compared with today’s free market and individual company decisions to determine locations. This concept involves midtransition infrastructure for expanding hydrogen coordination between auto companies, energy providers, availability to Full-size stations and local governments that is not practical today because of With increasing HFCV sales, hydrogen demand will antitrust concerns. eventually catch up to the initial capacity of the first stations. Convincing hydrogen suppliers to build hydrogen sta- In addition, coverage must expand to cover distant suburban tions before the introduction of many HFCVs is sometimes areas as well as some rural areas and highways between referred to as the “chicken-and-egg problem.” Hydrogen urban areas. The amount of hydrogen needed around any suppliers are reluctant to invest large sums before they know urban center would still be small, which favors hydrogen that many HFCVs will be sold. Similarly auto companies will technologies that do not need a large hydrogen distribution not be able to sell many HFCVs without an adequate number system. This includes distributed reforming of natural gas of hydrogen fueling stations. A way around this quandary or renewable liquid fuels such as ethanol and electrolysis of is to stage HFCV introduction in phases by region. This water using electricity. The distributed approach uses large approach is referred to as the “lighthouse concept” (Gronich, appliance-type devices located at the refueling site to convert 2007). For example, if the HFCV is first introduced in the the raw material to hydrogen. These devices would be factory Los Angeles area, then only 5 percent of the stations in that

OCR for page 31
 HyDROGEN TECHNOLOGy difficult. DOE analysis estimates that the footprint for an existing gasoline station will have to be increased by about BoX 3.1 7,200 square feet for either a full-size natural gas reformer or The hydrogen-powered iceV a water electrolysis system (Gronich, 2007). Even a smaller unit (e.g., 100 kg/d), would require about 2,200 square feet Hydrogen-powered ICEVs (HICEVs) have been demonstrated additional area. In any case, these are significant increases by several auto companies including Ford and BMW. BMW even that will limit the number of existing sites that could pos- offers a model to the public for lease. The hydrogen ICEV has sibly be used for dispensing hydrogen. This opens up the many of the same advantages as an HFCV in that the fuel can be possibility that many of the hydrogen refueling sites will be made from various sources and the primary combustion product at nontraditional locations such as shopping malls and big- is water. However since the HICEV still uses an oil-based lubricant, box retailer parking areas or even auto dealerships. Box 3.2 a small amount of this is consumed and very small CO2 releases discusses the latter option. result. In addition since combustion still takes place at a high Distributed natural gas reforming is the lowest-cost temperature, some NOx emissions also result. method of delivering hydrogen to an HFCV during this An HICEV can be about 25 percent more efficient than a gaso- period of low but growing demand for hydrogen. Most large line ICEV, but this is offset by increases in engine cost resulting urban areas have an existing natural gas infrastructure allow- from the light nature of hydrogen. It is difficult to get enough air ing its use in such places. For locations in which natural and hydrogen into the cylinders at normal pressure. Either much gas is not available, the outer reaches of population centers, bigger cylinders are needed or boosting the air pressure through or areas between cites along highways. other methods are turbocharging or supercharging is needed to achieve the power a needed. Water electrolysis is a proven, higher-cost, method gasoline engine provides. Either choice increases the cost. of hydrogen production. Table 3.1 summarizes hydrogen Perhaps the most difficult technical problem to resolve is storing enough hydrogen on board to travel a reasonable distance. Since the efficiency of the HICEV is only about 60 percent that of an HFCV, much more hydrogen is needed to travel the same distance. This probably will limit the number of HICEVs that enter the market. BoX 3.2 Because of the ease in converting an existing gasoline ICEV auto dealers selling hFcVs and hydrogen vehicle design to an HICEV it is likely that if a hydrogen infra- structure were built, some number of HICEVs would also be built. Auto dealers selling and servicing HFCVs will need hydrogen Since the cost of this conversion is much lower than the cost of an on-site for initial fills and likely for some types of service. They HFCV the number could be influenced by future carbon policy or may be willing to either sell hydrogen to the general public or to other government actions aimed at increasing hydrogen demand provide land to a fuel supplier at their dealer site. It would be a in order to reduce hydrogen cost especially at the beginning of selling point, especially in the early stages of the transition, be- the transition. cause dealers could assure customers that hydrogen fuel would be available at least at their facility and neighboring dealerships. Personnel at the dealership would have to be familiar with fueling procedures anyway, so the need for specially trained personnel would be reduced. Many dealerships also would have sufficient land area, unlike most current gasoline stations. manufactured and delivered to a site ready for installation. To see if this concept is feasible, the committee examined Research and development has focused on small-scale natu- General Motors dealerships in the Los Angeles Basin. There are ral gas reforming and water electrolysis. 131 GM dealers in the 2,027 square mile basin, or one for every This distributed approach likely will include adding 15.5 square miles. On average, there is one GM dealership within hydrogen generating and storage to some existing gasoline 2.2 miles of every person, which is not a great driving distance filling stations as well as building retail stations specifically to refuel. designed for hydrogen. The size of the stations can also be Although this is only one example, and indeed a very simpli- increased to become commercial scale (1,500 kg/d), reducing fied way of viewing the concept, it appears that dealerships could the cost of hydrogen further. contribute to hydrogen availability early in the transition. Dealers Full-size hydrogen stations, however, require much larger and customers would have to adapt to a different mode of busi- footprints than today’s gasoline equivalent. Typical modern ness, but this should be feasible. Many other types of locations gasoline stations are approximately 6,500 square feet, includ- would also be needed because the number of auto dealerships is ing a mini-mart. Area requirement is a very significant issue small compared to the number of gasoline stations. because suitable large sites may be difficult to find in urban areas and acquiring the needed permits is also likely to be

OCR for page 31
 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A FOCUS ON HyDROGEN TABLE 3.1 Dispensed Hydrogen Costs for Distributed The cost of producing hydrogen at a central plant and Generationa delivering it to a station is highly dependent not only on the feedstock cost and conversion technology, but also on Future Current the size of a commercial plant and the method and distance Technology Technology of delivering the hydrogen. In a fully developed hydrogen Process ($/kg) ($/kg) economy, delivery and dispensing of hydrogen could cost as Natural gas reforming 2.60 3.30 much as its production and consume significant energy. Electrolysis 5.60 7.20 The first central hydrogen production plants could be NOTE: Production plant is 1,500 kg/d; natural gas costs $6.10/MBtu; elec- needed as early as 2025. With the long lead times needed to tricity costs $0.08/kWh (see Chapter 6 for other assumptions). plan, permit, and construct large plants, the decisions on what aCosts are at the dispensing nozzle and are derived from the H2A model; they represent the final dispensed hydrogen cost using either technology type of feedstock to use would be made about 5 years prior demonstrated today or with anticipated future improvements and a minimum to this. These plants will have to use commercially proven of 500 installed units. technology. Natural gas reforming and coal gasification tech- nologies are available today. The cost of hydrogen from both feedstocks is similar when feedstock costs are $6 per million British thermal units (Btu) for natural gas and $27 per ton for costs for the main distributed options. Reforming a bioliquid coal. Twice as much CO2 is produced in a coal-fed plant as such as ethanol also is possible, although that is in an earlier one using natural gas. Both types could include CO2 capture stage of development (Paster, 2007a). The alternative tech- technology (at 80 to 90 percent efficiency), which produces nologies to natural gas reforming could be useful because a high-purity CO2 stream that can then be sequestered in they could lead to lower CO2 emissions than natural gas underground geological formations. Although CO2 capture reforming. Electricity for electrolysis could be generated by is a proven technology, CO2 sequestration (i.e., permanent low-CO2 methods, such as wind power, solar technologies, disposal underground) has not yet been adequately demon- or nuclear methods. Reforming ethanol or another bioliquid strated for commercial readiness. could yield much lower CO2 releases, depending on how the Biomass gasification for hydrogen production is under bioliquid is produced. This may become an important dis- development but not commercially ready yet. It is likely that tinction depending on future carbon policy and regulations. with continued emphasis on development it could be ready For example, California currently requires that at least 33 for commercial decisions in about 2025. Hydrogen from bio- percent of the energy used to produce hydrogen dispensed at mass gasification is more expensive than from natural gas or state-funded stations be provided from renewable sources. coal, but the net CO2 releases are very low if land use issues can be kept small (see Chapter 4, “Alternative Technologies late Transition to self-sustaining centralized hydrogen for Light-Duty Vehicles”). If CO2 sequestration is not com- Production mercially available when central plants are needed, biomass could be an important source of hydrogen that would not At some point in the hydrogen transition, the drive to contribute to carbon emissions. increase production and to lower both the cost of hydrogen Table 3.2 reviews the estimated costs of central hydro- and the CO2 emissions associated with hydrogen production gen production. The costs shown in this table are derived would shift the emphasis away from distributed production toward very large centralized production plants. This is likely to start between 2025 and 2030 in the MPR case. These cen- tral plants could use a variety of primary feedstocks includ- TABLE 3.2 Centralized Plant Gate Hydrogen Production ing natural gas, coal, and biomass. In the longer term there Costs could be additional technology options such as hydrogen from high-temperature nuclear and concentrated solar, pho- Future Current tobiochemical, and photoelectrochemical methods and from Technology Technology Process ($/kg) ($/kg) centralized electrolysis using solar or wind energy. The concept of centralized hydrogen production is very Natural gas reforming (379 tons/d) 1.50 1.60 Coal gasification (306 tons/d) 1.50 1.90 different from today’s centralized refining system for making Biomass gasification (155 tons/d) 1.80 2.50 gasoline, which is characterized by a few very large refining centers. They serve nearby population centers but also feed NOTE: Costs are at the plant gate and represent a learned out cost derived from the H2A model for either technology understood today or with an- a large pipeline network for delivering to areas hundreds ticipated future improvements. CO2 capture costs are included for coal or thousands of miles away. Gasoline is cheaper to ship by gasification but not the other technologies. CO2 sequestration costs are not pipeline than hydrogen (on an energy-equivalent basis). As included for any technology. Future delivery and dispensing costs of about a result, the optimal distribution system for hydrogen would $1 to $2/kg (pipeline) or $3.50/kg (liquid) must be added to the production lead to smaller-sized plants located closer to population cost for the final delivery cost (Paster, 2007b). Feedstock costs: natural gas $6.10/MBtu; coal $27/ton; biomass $38/ton. centers, with few if any long-distance pipelines.

OCR for page 31
 HyDROGEN TECHNOLOGy distributed Natural Gas reforming from the H2A model that is the basis of the resource needs analysis in Chapter 6. These costs are plant gate Catalytically reforming natural gas is now the most costs. Delivery and dispensing costs have to be added on common method of producing hydrogen at a large scale for a case-by-case basis for a final dispensed cost that could refineries and chemical plants. The challenge for near- and be directly comparable to those in Table 3.1 for distributed midterm use in a distributed system is to package this indus- hydrogen generation. trial process into small units or “appliances,” which can be As with distributed hydrogen production, it is unlikely installed inexpensively at existing retail stations. Making that centralized hydrogen production will be done using hydrogen directly at the filling station in a distributed sys- only one feed type or method. The best method in any par- tem eliminates the need for a large and expensive hydrogen ticular case will be determined based on regional and local transportation infrastructure. This concept will be very issues, including resource availability and hydrogen delivery important in the early and mid years of the transition. This options, all within any future existing carbon policy. concept includes all of the processes that are found in large- scale commercial plants, including the catalytic reformer, hydrogen safety catalytic shift reactors, and hydrogen cleanup processes to achieve the appropriate hydrogen purity. Several companies Safety issues, both real and perceived, along with the are developing this concept and have demonstrated small creation of appropriate codes and standards are significant appliances that are practical and have an appearance such barriers to the introduction of HFCVs and the hydrogen that they would not look out of place in neighborhood filling refueling infrastructure and are a significant risk to achiev- stations (James et al., 2007). ing the maximum practicable penetration rate for hydrogen The entire process has been successfully demonstrated, vehicles. Safety issues along with the need for codes and and most future developmental work now is directed at standards were addressed in The Hydrogen Economy (NRC, improving the efficiency and reducing the cost of equipment. 2004) and in the FreedomCAR and fuel partnership review Engineering cost analysis indicates that the overall cost of (NRC, 2008). The latter source noted that more urgency producing, storing, compressing to 5,000 psi, and delivering is needed to address these issues. It states that “work on hydrogen to a car at a future full-scale 1,500 kg/d station safety, codes, and standards is an essential federal role. The is now $3.30/kg, with the potential to reduce this further individual companies and states cannot do it on their own. to $2.60/kg.1 If the vehicle stores hydrogen at 10,000 psi, The manufacturers want and need uniform national (and as most manufacturers are proposing, the cost will be even hopefully international) standards so they can market on a higher, but the vehicle will be able to travel farther. worldwide basis.” Natural gas reformers appear to offer the lowest-cost method of producing hydrogen for the near and mid term hYdroGeN FeedsTocks aNd TechNoloGies and as such receive continued industrial emphasis. The cost of hydrogen is, however, highly dependent on the size The maximum practicable HFCV case starts with a few of the reformer (in kilograms per day) and on the cost of thousand HFCVs sold in 2012 and mass production sales natural gas. The hydrogen cost increases with a smaller-size (from 500,000 unit production systems) beginning several reformer system and with higher-cost natural gas. years later (after 2015). The questions about hydrogen To illustrate this point, the natural gas price used for esti- production then are the following: (1) Will the distributed mating a $2.60/kg cost just mentioned is $6.10 per million technologies be ready for commercial use about 2012 to Btu. If the natural gas price is increased to $10 per million 2015? (2) Will the centralized technologies be ready for Btu then the hydrogen cost is increased to $3.30/kg. commercial decisions about 2020 to 2025? The committee The ability to use distributed natural gas reformers in reviewed the readiness of many technologies and includes various sizes could be very useful during the initial transi- the following technologies as meeting the criteria of furthest tion to hydrogen to keep capital costs as low as possible—in developed and lowest cost: (1) distributed natural gas reform- ing, (2) distributed water electrolysis, (3) centralized coal gasification with carbon capture and sequestration (CCS), 1At a natural gas cost of $6.10/MBtu. All of the costs of making hydrogen (4) centralized natural gas reforming, and (5) centralized in this chapter are based on the assumptions discussed in Chapter 6 unless noted in the text. A kilogram of hydrogen has approximately the same biomass gasification. The discussion below addresses the energy content as a gallon of gasoline and is sometimes referred to as a technologies in their current state of development (current gasoline gallon equivalent (gge). To compare the cost of hydrogen with the state) and their potential for further improvements (future cost of gasoline one must also take into account the mileage of the vehicles. state). Costs for the current state and future state technolo- In the Chapter 6 analysis, the fuel cell vehicle achieves two times the fuel gies are based on DOE’s H2A model, which the committee economy of the gasoline ICE on a gge basis. In this analysis, an HFCV us- ing 1 kg of hydrogen will then travel the same distance as a gasoline ICEV reviewed and accepted. using 2 gallons of gasoline. Thus, hydrogen purchased at $2.60/kg for an HFCV yields the same cost per mile as gasoline purchased for a conventional vehicle at $1.30/gal.

OCR for page 31
 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A FOCUS ON HyDROGEN essence, matching the size of the reformer to the localized hence the cost of electricity vary considerably by region, demand for hydrogen. This could, however, lead to higher state, and in many cases locally in the United States, the hydrogen costs because cost increases as reformer size cost to make hydrogen varies widely. For example, in 2004, decreases. For example, if the 1,500 kg/d reformer men- industrial electricity averaged 9.5 cents/kWh in California, tioned above is reduced to 500 kg/d size, then the hydrogen 8 cents/kWh in Vermont, and 3.9 cents/kWh in Wyoming. cost increases from $2.60/kg to $3.50/kg. Further reducing This translates to the electricity cost component of making the reformer size to 100 kg/d increases the hydrogen cost hydrogen varying from $2.10/kg hydrogen in Wyoming to $6.20/kg. to $5.00/kg hydrogen in California at 74 percent electro- Significant progress has been made in developing distrib- lyzer efficiency. At a design hydrogen production rate of uted natural gas reforming such that it could be ready for 1,500 kg/d and an average industrial electricity cost of 8.0 commercial use in the early transition years (2015 or earlier). cents/kWh, the cost using current technology is $7.20/kg Even with this success, still further efficiency increases and hydrogen. Future technical improvements are evolutionary cost reductions are possible. in nature, which could result in a total hydrogen cost of Distributed natural gas reforming (DNGR) will release $5.60/kg hydrogen (Fletcher, 2007). some CO2 into the atmosphere. It is not feasible to use The sensitivity to electricity cost is about $0.50 in hydro- CCS technologies in this distributed process to capture and gen cost for each 1 cent/kWh in electricity cost, so for 2 sequester the CO2 because of the large number of sites and cents lower electricity cost than the average, the hydrogen the small size of each of them. However, on a WTW basis, cost would be $1.00/kg lower. For the hydrogen cost from an HFCV using hydrogen from DNGR units will emit less electrolysis to be in a competitive range with other feasible than half that emitted by a conventional ICE vehicle using sources of distributed hydrogen supply (about $3/kg), elec- gasoline, as shown in The Hydrogen Economy report (NRC, tricity would have to be available at an unrealistically low 2004). cost of 2 cents/kWh (Levene, 2007). Well-to-wheels CO2 The greatest potential challenges to the use of DNGR emissions from distributed water electrolysis (DWE) could technology in the hydrogen transition are identifying and be higher than for DNGR if the electricity is generated with permitting land use for the refueling sites and the cost and the current U.S. feedstock mix. availability of natural gas. The greatest challenge to the use of DWE for hydrogen is cost. However, electrolysis of water still may be used where lower-cost methods are not available (natural gas distributed Water electrolysis reforming) or where environmental advantages are possible Electrolysis of water is now a common way to produce through the use of solar or wind power for electricity. In small amounts of hydrogen. Scaling up an electrolyzer to full selected areas with no other source of hydrogen, policies size (about 1,500 kg of hydrogen per day) has been demon- could be put in place to help supply during a growing strated, and commercial alkaline electrolyzers are available hydrogen transition. in various sizes. For instance, Norsk Hydro reliably oper- ates a 600-700 kg/d alkaline electrolyzer. Water electrolysis central coal Gasification with carbon capture and could play a role in the early to mid stages of the hydrogen sequestration transition because of the advanced stage of technology development, the widespread availability of electricity, and Commercial large-scale high-pressure gasification plants the relatively simple operation of an electrolyzer. have been used for many years to produce a syngas (carbon In alkaline electrolysis, water under an applied voltage monoxide [CO] and hydrogen) for either power generation dissociates into hydroxyl ions and hydrogen on one side of or further conversion to chemicals. These plants run on heavy a wetted mat of sodium or potassium hydroxide solution. oils, petroleum coke, or coal. In this sense the basic core The hydroxyl ions traverse the wetted mat and form oxygen technology is technically and economically ready. on the other side. Electrodes are inexpensive nickel. This Although the core gasification technology is commer- is different from the polymer electrolyte membrane (PEM) cially available, the overall plant design and integration of process considered in many studies, including The Hydrogen the gasification module with the downstream processing Economy (NRC, 2004). That process involves transport of modules can still be improved through further development hydrogen ions across a polyelectrolyte membrane under an to lower the production cost (Litynski, 2007). Co-producing applied voltage using noble metal-based electrodes. The power and hydrogen at a large coal gasification facility is alkaline process is favored over the PEM process primarily likely to be the lowest-cost method for making hydrogen because it does not require the costly membrane and plati- (NRC, 2004). This concept also has not yet been demon- num-based electrodes of the PEM process and scales up more strated at a large scale although all of the processing modules economically at scales above 25 kg/d (Harg, 2007). needed are commercially available. Electricity is the largest cost component of hydrogen The largest single challenge for widespread use of coal production by electrolysis. Because the primary source and to make hydrogen is to lower CO2 releases to acceptably

OCR for page 31
 HyDROGEN TECHNOLOGy low levels. In the case of an oxygen-supplied gasification, real technical barriers to its expanded use to make hydrogen capturing CO2 is relatively inexpensive. This CO2 must for transportation. The current cost of producing hydrogen at then be piped to a CO2 sequestration site and injected into a 380-tonne-per-day plant is $1.60/kg at the plant gate for a underground storage. In a future with constraints on carbon natural gas cost of $6.10/MBtu (H2A model runs). This cost emissions it is likely that if coal is to be used to make large is similar to that of hydrogen from a coal gasification plant. amounts of hydrogen, CO2 sequestration must be adequately Adding on delivery and dispensing costs could increase the demonstrated and commercially available. final supply cost to about $2.90/kg. Because steam methane Large-scale CO2 sequestration from an integrated coal reforming is such a well-developed commercial technology, gasification plant with CCS has not yet been adequately dem- little future cost improvement is expected. As shown in Table onstrated. Several partnerships have been formed to further 3.2 the future plant gate cost is anticipated to decrease to develop and demonstrate CCS technology involving DOE, $1.50/kg at the same natural gas cost. 41 states, and more than 400 organizations. Seven sequestra- There are two primary issues with using significant natu- tion injection tests have been awarded funding through these ral gas resources to make hydrogen for transportation. One partnerships (Litynski, 2007). The goals of this work are to is CO2 releases, and the other is the availability and cost of verify that the costs of CO2 capture and sequestration are natural gas. The CO2 associated with hydrogen production not high (less than 10 percent increase in power or hydrogen and use in an HFCV is about half of that associated with a costs) and that any sequestered CO2 will remain so (less than gasoline ICEV. From this perspective, using natural gas to 1 percent leakage after 100 years). Since it will take many make hydrogen results in lower CO2 emissions. As with coal years to complete this program, success may not be known gasification, this CO2 can be captured and sequestered, and prior to the start of the hydrogen transition. if done, the CO2 releases will be very small. The cost of hydrogen produced at a central gasification The largest barrier to widespread use of central natural plant is estimated to be $1.50/kg at the plant gate (Joseck, gas reforming technology is the availability and price of 2007) for a 2,400-ton-per-day plant that makes 306,000 natural gas. Since the incremental natural gas supply to the kg/d. This does not include costs for carbon sequestration, United States is currently from imported liquefied natural gas which should be low as a percent of the total hydrogen (LNG) and imports are expected to grow in the future, using delivered costs (NRC, 2004)). Additional pipeline delivery natural gas to make hydrogen will increase the requirement and dispensing costs could increase the final delivered cost for imports. Although natural gas pricing is a complex issue, to $2.60/kg (Paster, 2007a). Table 3.2 summarizes hydrogen it is likely that a significant number of HFCVs using natural plant gate production costs for the key central production gas will increase natural gas demand enough to affect natural technologies. The cost of making hydrogen from coal is gas prices, which in turn would have a great impact on the somewhat insensitive to the price of coal, because most of power sector (a heavy user of natural gas). If natural gas is the overall cost is related to the large capital cost of the plant. the source of hydrogen for 10 million HFCVs, the demand For instance, tripling the coal cost, from $27 to $81 per ton for natural gas in the United States would increase by about 2 increases the hydrogen cost by about $0.50/kg. percent, probably not enough to have a significant impact on The greatest challenge to the use of coal gasification for prices. However, 50 million cars, less than 20 percent of the hydrogen production is demonstration of the costs, capacity, light-duty vehicles expected in the country in 2020, would safety, and risks of long-term carbon capture and sequestra- increase natural gas demand by 10 percent, which probably tion. Although coal gasification is a commercially available would increase prices. technology, to reach the future cost estimates shown in Table 3.2, some further development is needed. Standardization centralized Biomass Gasification of plant design, gas cooler designs, process integration, oxygen plant optimization, and acid gas removal technol- Biomass may consist of waste by-products of agricul- ogy shows potential for lowering costs. Other areas that ture, such as corn stover or wheat straw, or forest residues; can have an impact on future costs include new gasification it also can be expressly cultivated as an energy crop, for reactor designs (entrained bed gasification) and improved example, switchgrass. It could be gasified in the same type gas separation (warm or hot gas separation) and purification of high-pressure oxygen-fed gasifier used for coal, but that technologies. These technologies need further R&D before alternative may not be the best for gasifying biomass to make they are commercially ready. hydrogen. High-pressure gasifiers tend to be less economical at small scales than at a large scale. Much of the feedstock and some of the process issues with biomass gasification are central Natural Gas reforming similar to those for biofuels discussed in Chapter 4. Steam methane reforming of natural gas is the predomi- Because biomass is produced in relatively small quantities nant method of making large amounts of hydrogen for the per farmed acre in a dispersed agricultural system, a biomass chemical processing industry and the refining industry. It is a plant is likely to be much smaller than other central produc- well-developed and inexpensive commercial process with no tion methods such as a coal gasification plant. The size of

OCR for page 31
8 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A FOCUS ON HyDROGEN any biomass plant will be determined by its proximity to a million dry tons in 2025, based on DOE estimates reviewed large acreage of land that has available agriculture biomass, with this committee (Hess, 2007). If all of the 500 million that is appropriate to grow energy crops, or that contains for- dry tons were converted into hydrogen, this would be about est residue, while also being near large population centers. 37 billion gallons of gasoline equivalent, or 26 percent of today’s gasoline market.2 Biomass starts out with an inherent environmental advantage versus coal or natural gas since CO2 is cycled from the atmo- Biomass gasification is promising, but much remains to be sphere in its growth and cycled back to the atmosphere in its done to put it on a solid basis. Crop yield assumptions need processing, which can result in little net CO2 emission to the to be demonstrated. Preparation of the different types of bio- atmosphere without the need for carbon sequestration if land mass feedstock for ease of delivery and reliable processing in use issues are kept to a minimum. the gasifier needs to be determined. The true number of sites Given the economic reasons for smaller plants, a different that can actually supply large amounts of biomass without gasification technology could prove to be a better choice for incurring large land use problems has to be understood. making hydrogen from biomass. There have been several Significant research, development, and technology dem- small-scale (100 to 300 tons per day of biomass feed) bio- onstration is required before the future costs shown in Table mass gasification test plants operated using a low-pressure, 3.2 can be achieved. Although individual parts of the biomass indirectly heated air gasification process, for example, a gasification process have been demonstrated, the entire pro- 300 tons per day Battelle gasifier unit producing fuel gas cess has not been demonstrated. Bench-scale, pilot plant, and for electric power generation in Vermont. In this process the semicommercial-scale work is needed to have a firm basis biomass is gasified with air at high temperature and atmo- for scale-up to a 2,125-ton-per-day plant or larger. Also, spheric pressure to produce a mixture of CO and hydrogen. gas cleanup and separation technologies into pure hydrogen Because the Battelle gasifier operates at low pressure, it has need to be demonstrated while dealing with contaminants the potential to be less capital intensive than the high-pres- and tar. The committee judges that technology readiness by sure oxygen gasifier, which also requires an air separation 2015 will be difficult to achieve; several years more may be unit for oxygen feed. needed. However, if successful, hydrogen supply from bio- DOE bases its future technology plans on the indirect- mass gasification could phase in with other supply sources heated low-pressure air gasifier type plant at 155,000 kg of hydrogen. hydrogen per day. The biomass feed rate required for a plant this size is 2,125 tons per day (Mann, 2007), and the land hYdroGeN FUel cell Vehicle TechNoloGies needed to supply such quantities is about 180 square miles for an energy crop such as switchgrass. The number of avail- The HFCV is an all-electric automobile. It differs from able sites must be determined that could support such a large previous generations of electric cars in that the power is plant and still have acceptable delivered biomass cost and provided by a hydrogen fuel cell on an as-needed basis. delivered hydrogen cost with limited land use issues. The Previous generations of electric vehicles stored energy only DOE estimates roughly 50 potential sites throughout the in batteries whereas the HFCV stores its energy primarily in country at current biomass yields and upward of 100 sites a hydrogen tank. Most major automakers have demonstrated with future crop technology (Joseck, 2007). different iterations of HFCVs, with each version demonstrat- With future technology, DOE targets biomass crop yields ing new concepts to achieve performance that is approaching per acre increasing 50 percent because of applied plant genet- that of today’s gasoline vehicles. Some HFCV prototype ics and crop rotation practices. This is based on a model that vehicles also include some degree of energy storage in bat- assumes switchgrass, a 45 percent component of projected teries using the same technology that is in today’s gasoline biomass availability, will develop in crop yield similar to hybrid vehicles. These prototype vehicles have demonstrated the achievements with corn. However, the potential for yield significant success in overcoming difficult technological improvements in other components of biomass, for example, challenges, such as reducing the size and weight of the fuel corn stover and wheat straw, is uncertain. cell and improving operation in cold weather. However, even The total hydrogen production cost at the plant gate based with the significant improvements of the past few years, on this future technology is $1.80/kg hydrogen with a bio- mass feedstock cost of $38 per dry ton (Table 3.2). Delivery 2The biomass gasification technology discussed in The Hydrogen Econo- my is based on a conventional high-pressure oxygen gasifier at just 24,000 kg and dispensing will add additional costs of about $1.70/kg hydrogen per day, a small scale to ensure economic biomass supply (NRC, (pipeline delivery is higher than for coal plants because bio- 2004). This is in contrast to the larger plant and broader feedstock supply mass plants are smaller) to $3.50/kg (liquid truck delivery) reach envisioned in the future technology and the low-pressure indirect to the plant gate cost. Total projected biomass availability heated gasifier of the future technology discussed above. The biomass at $38 per dry ton is 200 million dry tons in 2015 and 500 feed rate for this small plant is 442 dry tons per day from a 57 square mile collection area. The total hydrogen cost estimate is $7.00/kg hydrogen at a biomass feedstock cost of $53 per ton. The total cost is particularly high since hydrogen liquefaction and tanker transport are required because pipe- line supply is not economic at such a small scale.

OCR for page 31
 HyDROGEN TECHNOLOGy there are still several areas that need further improvement. and the International Organization for Standardization (ISO) The key component technology challenges for the HFCV are (i.e., SAE TIR J2719 and ISO 14687-2, respectively). Once (1) making the fuel cell system as durable and cost-effective the fuel standards are developed (expected in 2008), fuel as today’s gasoline internal combustion engine vehicle, and companies can then begin to build a large enough number of (2) engineering a small, lightweight hydrogen storage system distribution outlets to satisfy early HFCV customers that they to provide an acceptable driving range, 300 miles or more. will be able to acquire fuel safely and conveniently. The degree of success in these areas will determine when the HFCV can be commercialized and how effectively it can hFcVs and Fuel cell Technology compete with today’s ICEVs. This section discusses these issues and provides a context for how they can be viewed Progress in fuel cell development has been rapid. Vehicle to meet the maximum practicable rate of penetration for fuel cells are much smaller and lighter than they were just HFCVs. several years ago. Fuel cell costs have decreased, while performance and durability have increased considerably. However, production-quality vehicle fuel cells are still in recent history the development stage. Several major automakers have been developing the To the best of this committee’s knowledge, no vehicle fuel HFCV in earnest since the early 1990s. By the mid-1990s cell system has yet met the full set of rigorous automotive Daimler-Benz announced that it would begin production specifications required for high-volume sales to customers. of HFCVs in about 2004. Although a small number of test In some sense, these requirements must be met before an vehicles has been produced, commercial production has not automobile company can get on the high-volume manufac- yet begun. In 2003 GM announced that production could turing (500,000 units) cost reduction curve. Furthermore, begin in 2010, but more recently GM has pushed this date there is a degree of uncertainty about the manufacturing cost back to 2011 or 2012.3 Because of the highly competitive estimates because fuel cells have not yet been manufactured nature of the auto industry, not every automaker has made on a large scale. Thus, current cost estimates are derived from its plans available to the public. detailed engineering studies, as they were in NRC (2004, This situation suggests the uncertainty and risks involved 2005, 2008), Walsh et al. (2007), and Brunner (2006). On in introducing a new vehicle technology to the market. One the other hand, all major automotive manufacturers have seen of the benefits of the FreedomCAR and Fuel Partnership enough progress that, as a group, they are spending billions (FCFP) is that all of the known areas that require further of dollars to bring fuel cell vehicles to high-volume produc- development for HFCV commercialization are included with tion. The main debate among the manufacturers appears to developmental targets and dates for completion specified. be concerned with “when,” not “if.” This permits more realistic assessments of the state of overall More recently, the NRC FCFP review committee noted, development. To date, only a few hundred HFCVs have been “Fuel cell stack life currently limits the overall demonstrated produced, with none of the advantages of mass production. powerplant durability to only about one fourth of what is Costs in high-volume manufacturing can be estimated only needed to meet the performance targets set forth by the roughly, because several major subsystems are still in the Partnership. A major reduction in stack life occurs in actual development stage and “tight” manufacturing estimates are vehicle applications because of the many stops and starts not available. In the traditional process for developing a and transients with vehicle operations, fuel composition, and new-technology power train for commercialization, once the related phenomena when compared to what is observed with technology is developed, hundreds of vehicles are put into the testing methods and conditions in laboratory develop- experimental stressful applications (such as police vehicles) ment work. In addition, as laboratory fuel cell stack lifetimes to ensure that there are no “unknown-unknowns” that could lengthen, new failure modes are surfacing and must be bet- cause premature durability problems. Following that step, ter understood and resolved. One such example is platinum small-scale volume production in the thousands can begin. catalyst dissolution, which impairs long-term performance. Since the HFCV will require a new fuel as well as a new The prompt resolution of these and new failure modes, as power train, the automotive and fuel companies are working they are discovered, is critical to achieving 2010 and 2015 together to develop standards for the vehicle-fuel interaction targets” (NRC, 2008, pp. 56-57). (e.g., the purity of hydrogen required for the vehicle and Fuel cell costs have been reduced significantly over the fueling protocol). Hydrogen fuel standards are currently the past 4 or 5 years. Cost projections for high-volume under study by the Society of Automotive Engineers (SAE) (500,000 units per year) automotive fuel cell production are estimated to be $100/kW for relatively proven technologies and $67/kW for newer laboratory-based technologies (which 3Larry Burns, GM vice president of R&D and Strategic Planning, said, “I don’t know how many of them we’ll make at the time, but we should may be compared with the DOE/FCFP commercialization have them in showrooms by early next decade, around 2011 or 2012. Post- goal for 2015 of $30/kW). The cost of platinum is 57 percent 2012, the goal is to ramp up production to about a million vehicles a year, of the fuel cell stack costs and is the greatest challenge to worldwide” (Burns, 2007).

OCR for page 31
0 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A FOCUS ON HyDROGEN FIGURE 3.1 ZEV panel vehicle market penetration estimates. SOURCE: Walsh et al. (2007). Figure3-2.eps BITMAP further cost reductions.4 Future platinum supply is a criti- ment agencies. This support must include a clear message of cal issue in forward projections of fuel cell costs. Fuel cell long term commitment to . . . FCEVs. These include adequate stack life has increased to over 1,500 hours compared to and affordable hydrogen refueling, as well as a host of sus- the DOE/FCFP 2015 goal of 5,000 hours. Focused research tainable financial incentives to help minimize the capitaliza- on problems, together with recent advances in electrode tion risks of all key stakeholders during the initial transition and membrane technology, should further reduce costs and years. Ultimately, consumer knowledge and willingness to increase stack life. buy these vehicles in high volume is required” (p. 130). Walsh et al. (2007) reports on an extensive analysis of fuel Walsh et al. (2007) contained an overall estimate of mar- cell systems, as well as competing technologies that produce ket introduction time frames for the various low-emission very low emission vehicles in the study called Status and vehicle technologies they analyzed (Figure 3.1). That ZEV expert panel’s estimate is that production of thousands per Prospects for Zero Emissions Vehicle Technology: Report of the ARB Independent Expert Panel. These analysts visited year could occur by 2009, with tens of thousands per year 10 automotive manufacturers, reviewed proprietary informa- by 2020, and then mass commercialization by 2025, with the tion, and developed the following assessment after their visits statement that “the panel remains cautiously optimistic for and data gathering: “Each of the developers believes that the fuel cell system commercialization” (p. 130). The estimates simultaneous requirements can be met but on different time just discussed along with presentations from auto manufac- schedules. For example, one major developer’s objective is to turers and information included in the other resources noted compete with the ‘upper’ segment of ICE vehicles in the year were used in developing the HFCV market penetration sce- 2020 at volumes of 100,000 units per year. Another major narios in the Chapter 6 analysis. developer’s assessment is that a commercially viable fuel cell The committee concludes that the current state of fuel system would be available in 2010, if a production rate of cell development does not yet meet all of the performance 500,000 units per year could be realized” (p. 8). The panel and cost requirements needed for large-scale commercial also noted: “There are large technical barriers that can be production. If the recent progress in size and weight reduc- solved but there are other issues that are beyond the control tion, cold-weather operation, and durability improvements of any single auto manufacturer. Widespread deployment of can be continued over the next few years, a usable fuel cell FCEVs will require continuous strong support from govern- technology may be made ready for introduction by 2015. The costs of the early fuel cells are likely to be higher than the commercial targets, but these costs can drop with continued 4The stack is the heart of the fuel cell. It contains the membrane through development and large-volume manufacturing. which hydrogen passes to react with oxygen, generating electricity.

OCR for page 31
 HyDROGEN TECHNOLOGy FIGURE 3.2 BMW assessment of on-board liquid hydrogen storage. SOURCE: Brunner (2006). hFcVs and hydrogen storage absolute zero). The BMW system has impressive insulation, but some heat still gets in and gradually boils off the hydro- Hydrogen can be stored on board a vehicle as a gas, liquid, gen. This can result in serious safety issues. The other issue or solid. All of these techniques have been demonstrated in is cost, currently on the order of $500/kWh, with a goal of HFCVs in recent years. All continue to be developed through approximately $100/kWh in the “next generation” (Brunner, government research and industry programs. At this point 2006). The eventual goal is $15/kWh. To put these numbers there is no consensus as to which storage state will be the best in perspective, the FreedomCar targets for 2010 and 2015 long-term solution. However, there is a growing consensus are $4/kWh and $2/kWh, respectively. As shown in Figure that in the short term, high-pressure gaseous storage is the 3.2, today’s system suffers from performance, durability, and most practical solution. maintenance issues in addition to the noted major cost issues. Solid hydrogen storage systems that are made up of a The figure shows just inside the rectangle the areas in which low-pressure tank filled with a solid storage material with a R&D is being performed, and the goals are on the perimeter thermal management system have the potential to be small of the 10-sided figure (e.g., keeping evaporation losses to and lightweight, which aids in overall weight savings and less than 25 percent per month for the infrequent driver). improved fuel mileage. They may present the best long- Inside this is the current status of progress toward the goal term storage solution but at present are still mainly in the (e.g., less than 50 percent of the way for evaporation loss). research stage (to identify the best solid medium for storing Because of these concerns, the committee does not believe hydrogen). Thus, it is important to develop solid hydrogen liquid storage systems will be commercially viable in the storage systems as well as to integrate them into the vehicle 2015-2020 time frame without unexpected breakthroughs in in an energy-efficient manner. For further discussion of solid liquefaction and insulation. storage technical readiness, see NRC (2008) and Walsh et al. As noted by Walsh et al. (2007), “With the exception of (2007). Based on the committee’s current knowledge that no BMW, every other OEM [original equipment manufacturer] solid storage medium has yet met all of the developmental contacted indicated that this (compressed gas) was the only targets, it is unlikely that solid hydrogen storage systems will realistic short term (5-10 years) choice available and only be production ready in 2015. Honda indicated that they intend to limit the storage pres- Liquid hydrogen currently provides the densest form of sure to 350 bar.5 All the other OEMs preferred 700 bar, storage, which means that the most fuel can be stored on which will provide storage of over 50% more fuel in the board. BMW has recently demonstrated a liquid storage system. Two formidable problems must be overcome to make liquid storage practical for widespread use. Liquid 5One bar equals one atmosphere of pressure (14.7 pounds per square inch hydrogen must be kept at about −252°C (about 20°C above [psi]), so 350 bar is about 5,000 psi and 700 bar is 10,000 psi.

OCR for page 31
 TRANSITIONS TO ALTERNATIVE TRANSPORTATION TECHNOLOGIES—A FOCUS ON HyDROGEN same space envelope and correspondingly provide almost the maximum practicable case, the committee estimates that 50% more range.” the fully learned out cost for the fuel cell drive train (the fuel The DOE goal for hydrogen storage systems is enough cell system, hybrid battery, motor, and auxiliaries) for the fuel to travel about 300 miles (a similar range to that of automaker (OEM) will be $50/kW. This corresponds to a today’s gasoline ICEV). The amount of hydrogen needed for fuel cell system cost of $30/kW plus added costs for a hybrid this depends on the fuel consumption of the HFCV. Toyota battery, electric motor, and other components. Of the $30/kW demonstrated in September 2007 a 4,145-pound, five-pas- fuel cell system cost, about half is due to the fuel cell stack senger HFCV with 700-bar compressed hydrogen storage and half to the balance of the plant. Hydrogen storage costs that traveled 350 miles in real-world on-the-road conditions the OEM $10/kWh compared to DOE 2015 goal of $2/kWh in a drive from Osaka to Tokyo. Toyota calculated that the for solid storage. The fuel cell cost is the same as the 2015 vehicle is now capable of achieving a cruising distance of 466 DOE goal, while the storage costs are higher than the DOE miles. It appears that the latest HFCV designs using high- 2015 goal because high-pressure hydrogen gas storage was pressure hydrogen storage can meet the 300-mile goal. assumed in the latter. Less progress has been made in meeting the cost targets for such a system. The 2005 NRC review of the Freedom- coNclUsioNs CAR and Fuel Partnership listed the circa 2004 cost status CONCLUSION: If appropriate policies are adopted to as $15/kWh and $18/kWh for the 350- and 700-bar systems, accelerate the introduction of hydrogen and HFCVs, respectively. The 2008 NRC review of the FCFP did not hydrogen from distributed technologies can be provided update these costs, and discussions with auto companies at reasonable cost to initiate the maximum practicable indicated that little has changed with regard to costs for case. If technical targets for central production tech- compressed hydrogen storage. nologies are met, lower-cost hydrogen should be avail- Based on these facts, the committee concludes that com- able to fuel HFCVs in the latter part of the time frame pressed hydrogen storage systems that provide practical considered in this study. Additional policy measures are driving ranges (300 miles) should be available in 2015, but required to achieve low-carbon hydrogen production in the cost will be higher than that of the current FreedomCAR order to significantly reduce CO2 emissions from central targets. There is potential to lower the costs in the future coal-based plants. through the use of lower-cost carbon fiber tanks or by using future solid storage systems. CONCLUSION: Lower-cost, durable fuel cell systems In summary, onboard hydrogen storage to achieve a 300- for light-duty vehicles are likely to be increasingly avail- mile driving range has been the greatest technical challenge able over the next 5-10 years and, if supported by strong of all in trying to develop an HFCV. The quest to identify government policies, commercialization and growth of solid storage materials to achieve the DOE-FCFP 2015 goals, HFCVs could get underway by 2015, even though all including the cost goal of $2/kWh, is in the research stage. DOE targets for HFCVs may not be fully realized. It is not clear at this time whether a suitable material will be identified that can meet these goals and timing targets, but to achieve the desired driving range between refueling stops, Considerable progress has been accomplished since The the industry is prepared to use more expensive high-pressure Hydrogen Economy (NRC, 2004) toward a commercially hydrogen storage tanks that consume more space and add to viable hydrogen fuel cell vehicle due to the concentrated vehicle weight while research progresses toward a commer- efforts of private companies and governments around the cially viable solid hydrogen storage material. world. Although considerable progress is still required in fuel cell costs, durability, and storage before commercialization can begin, the automotive industry appears committed to the Technology Basis for the scenario analysis technology for the long run. Thus, lower-cost, durable fuel The committee concludes that not all the FreedomCar cell systems for light-duty vehicles are likely to be available goals for 2015 are likely to be met, but the technology may in a growing number of vehicles over the next 5-10 years, be good enough for high-volume HFCVs to be introduced but meeting all 2015 DOE commercialization targets will then anyway. For the scenarios analyzed in Chapter 6, the be difficult. committee assumes that the hydrogen storage system will be larger and more costly than the targets but will be able BiBlioGraPhY to provide adequate driving distance. The fuel cell system Bereisa, J. 2007. Energy Diversity: The Time Is Now. Presentation to the will be more costly than the target initially but will provide committee, June 25. the necessary performance expected of an early commercial Brunner, T. 2006. BMW Clean Energy—Fuel Systems. Presented at the vehicle. Although the initial costs will be high, there is CARB ZEV Technology Symposium. Sacramento, California. considerable scope for continued cost improvement through Burns, L. 2007. Quoted in Reuters dispatch, May 17. technology improvements and high-volume production. For

OCR for page 31
 HyDROGEN TECHNOLOGy Cicero, D. 2007. The U.S. DOE’s Hydrogen from Coal Program. Presenta- Levene, J. 2007. Electrolysis: Electricity Is the Key. Presentation to the tion to the committee, April 19. committee, April 19. Duleep, K. 2007. The Hydrogen Transition and Competing Automotive Litynski, J. 2007. Summary of DOE’s Carbon Sequestration Program. Technology. Presentation to the committee, April 18. Presentation to the committee, April 19. Fletcher, K. 2007. Hydrogen Transitioning to Carbon Free Energy. Presenta- Mann, M., et al. 2007. Hydrogen from Biomass Gasification: Cost Analysis tion to the committee, June 25. and Technology Status. Presentation to the committee, April 19. Giove, J. 2007. FutureGen Briefing. Presentation to the committee, Nicholas, M., et al. 2004. Using Geographic Information Systems to Evalu- April 19. ate Siting and Networks of Hydrogen Stations. Transportation Research Gronich, S. 2007. 2010-2025 Scenario Analysis. Presentation to the com- Record 1880:126-134. mittee, February 20. NRC (National Research Council). 2004. The Hydrogen Economy: Op- Harg, K. 2007. Hydrogen. Presentation to the committee, April 19. portunities, Costs, Barriers, and R&D Needs. Washington, D.C.: The Hess, R., et al. 2007. Cellulosic Biomass Feedstocks for Renewable Bioen- National Academies Press. ergy. Presentation to the committee, April 18. NRC. 2005. Review of the Research Program of the FreedomCAR and Fuel James, B. 2007. H2A Production Model Description. Presentation to the Partnership: First Report. Washington, D.C.: The National Academies committee, April 19. Press. James, B., et al. 2007. Technology Readiness: Dist. NG and Bio-derived NRC. 2008. Review of the Research Program of the FreedomCAR and Liquids Reforming. Presentation to the committee, April 19. Fuel Partnership: Second Report. Washington, D.C.: The National Joseck, F. 2007. Hydrogen Program Resource Analysis. Presentation to the Academies Press. committee, February 21. Paster, M. 2007a. Hydrogen Delivery Infrastructure. Presentation to the Kawai, T. 2007. Sustainable Mobility and the Development of Advanced committee, April 18. Technology Vehicles. Presentation to the committee, June 26. Paster, M. 2007b. Hydrogen Delivery Models. Presentation to the commit- Lasher, S., et al. 2007. Direct Hydrogen PEMFC Manufacturing Cost Es- tee, April 19. timation for Automotive Applications. Presentation to the committee, Walsh, M., et al. 2007. Status and Prospects for Zero Emissions Vehicle April 25. Technology: Report of the ARB Independent Expert Panel. Prepared for State of California Air Resources Board, Sacramento, California.