Review of the Research Program of the FreedomCAR and Fuel Partnership: Second Report (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Hydrogen Production, Delivery, and Dispensing PROGRAM OVERVIEW As discussed in Chapter 1, the FreedomCar and Fuel Partnership in DOE’s Office of Energy Efficiency and Renewable Energy (EERE) includes the hydro- gen production, delivery, and dispensing program, which is, in turn, part of the Hydrogen, Fuels Cells, and Infrastructure Technologies (HFCIT) program (see Appendix A for an EERE organization chart). This program addresses a variety of means of producing hydrogen, including by biomass gasification and steam reforming of bioderived liquids. The manager of HFCIT is the overall DOE hydrogen program manager. There are three fuel technical teams (see Figure 1- 1): fuel pathway integration, hydrogen production, and hydrogen delivery, with participation from DOE and the five energy companies that joined the Partnership 3 years ago. The technical teams report to the Fuels Operations Group, consist- ing of energy directors and DOE program managers, who in turn report to the Executive Steering Group. Other activities related to the HFCIT program are in other DOE program of- fices. The Office of Fossil Energy (FE) supports the development of technologies to produce hydrogen from coal and related carbon sequestration technologies. The Office of Nuclear Energy (NE) supports research on the potential use of nuclear heat to produce hydrogen, and the Office of Science (SC) supports fundamental work on new materials to store hydrogen, catalysts, and fundamental biological or molecular processes for hydrogen production, as well as work potentially af- fecting other areas of the FreedomCAR and Fuel Partnership. Work on growing, harvesting, transporting, and storing biomass is carried out in EERE but is not part of the Partnership. 81 

82 review of the freedomcar and fuel partnership TABLE 4-1  Funding Levels for Hydrogen Production, Delivery, and Dispensing Activities in the Partnership Funding (millions of dollars) DOE Office FY06 FY07 FY08 Request EERE/HFCITa 8.4 34.6 40 Fossil Energy 94.9 123.6 91.6 (HFI and CCS) Nuclear Energyb 25 19.2 22.6 Sciencec 12.6 13.5 13.7 Total 140.9 190.9 167.9 NOTE: HFI, Hydrogen Fuel Initiative; CCS, carbon capture and sequestration. a The request for FY08 for EERE/HFCIT includes $17 million for work focused on production, delivery, and dispensing in the transition period. Expenditures include conversion of biomass to hy- drogen but not growth, harvesting, storage, or transportation of biomass prior to conversion. b Nuclear Hydrogen Initiative; excludes funding for the next-generation nuclear plant (NGNP). c For hydrogen production. SOURCE: DOE, Answers to questions from committee, pp. 2-9, ������������� June 7, 2007. The budget for the areas relating to hydrogen production, delivery, and dispensing is given in Table 4-1 (DOE, 2007a). In reviewing the hydrogen pro- duction, delivery, and dispensing areas, the committee considered whether it is appropriate for the federal government to be involved and, without exception, concluded that these activities are appropriate for federal involvement. As will be shown in this chapter, DOE through its HFCIT program has made substantial progress on hydrogen production, ensuring that hydrogen can be made available to meet the needs of fuel-cell-powered vehicles as they emerge. However, success in work still under way is needed to minimize cost and to make feasible the production of this hydrogen without increasing carbon dioxide emis- sions or natural gas imports. HYDROGEN FUEL PATHWAYS The hydrogen fuel/vehicle pathway integration effort is charged with looking across the full hydrogen supply chain from well (source) to tank. Specifically, the goals of this integration effort are to (1) analyze issues associated with complete hydrogen production, distribution, and dispensing pathways, (2) provide input to the Partnership on goals for individual components, (3) provide input to the Partnership on needs and gaps in the hydrogen analysis program, and (4) foster full transparency in all analyses. This involves source to vehicle tank analysis, including costs, energy use, safety, and carbon dioxide (CO2) emissions. Accomplishment of these goals is overseen by the fuel pathways integration technical team (FPITT) with representation from DOE, the five energy companies, and the National Renewable Energy Laboratory (NREL). FPITT’s expertise sup- 

HYDROGEN PRODUCTION, DELIVERY, AND DISPENSING 83 ports the analysis efforts of the Partnership, coordinates fuel activities with the vehicle systems analysis technical team, recommends additional pathway analy- ses, provides input from industry on practical considerations, and acts as honest broker for the information generated by other technical teams. DOE has made important progress toward understanding and preparing for the transition to hydrogen fuel. An effort to develop a transition strategy was estab- lished, several workshops to develop scenarios for the transition were held, and a program with three parallel approaches to hydrogen production at fueling stations has been implemented. Program target dates call for completion of the program in about 2017, consistent with the President’s goal of enabling large numbers of Americans to choose vehicles powered by hydrogen fuel cells by 2020. Clearly, there is uncertainty about the time frame in which transitional hydro- gen will be required, economically sustainable hydrogen-powered vehicles can be achieved, and a well-developed hydrogen fuel infrastructure can be put in place. Given this uncertainty, the committee believes that DOE needs to incorporate in its studies a time frame for the transition to and subsequent emergence of a mature industry. Thus far, DOE has focused on the transition through 2025, but market sustainability might not be established until 2035 or later. It will take more than a decade to move from the manufacture of a few thousand vehicles per year, when transitional quantities of hydrogen will first be needed, to a mature indus- try that supports a mature hydrogen production/supply system with centralized production and pipeline distribution. The amount of transitional hydrogen needed over that period will change dramatically. To illustrate this point, the number of hydrogen-fueled vehicles could increase from an assumed 10 million in 2025 to 40 million in the following 10 to 20 if there is a growth rate of 7 to 15 percent annually. Obviously, hydrogen supply would have to grow similarly. Even 40 million vehicles might not be sufficient to stimulate the development of a self- sustaining, mature industry, so transitional methods might be needed eventually to supply even more vehicles and to provide fuel in remote areas. The potential roles of the different transitional hydrogen supply paths need to be viewed from the perspective of this uncertainty. For instance, while transitional hydrogen for 10 million cars might be produced from natural gas without increasing the cost of the natural gas, transitional hydrogen for 40 million cars produced from natural gas would most likely increase the natural gas cost significantly. Thus different energy sources could become important at different points in the transition. Recommendation.  DOE should continue its studies of the transition to hydro- gen, extending them to 2030-2035, a transition period during which the number of hydrogen vehicles in use could increase rapidly and use the results of these studies as a basis for evaluating the potential roles of different transitional sup- plies of hydrogen fuel as demand increases substantially, including both forecourt production at the fueling station and centralized production using the most cost effective means of distributing the hydrogen. 

84 review of the freedomcar and fuel partnership HYDROGEN PRODUCTION The hydrogen production goals are based on the premise that no single energy source is likely to meet all energy needs in the long term and that U.S. energy se- curity will be enhanced by producing hydrogen from a diverse set of feedstocks.  The hydrogen production technical team facilitates the development of commer- cially viable production technologies. The energy sources under consideration for hydrogen generation, in addition to grid-based electrolysis, are natural gas, coal, biological systems, nuclear heat, wind, and the Sun. Side-by-side comparisons of the cost of producing hydrogen with these different energy resources are not included in this chapter, for two important reasons. First, as described below, the reliability of the estimates varies substantially. Estimates for coal and natural gas are based on actual commercial operating experience, but other estimates, such as those for biomass-derived hydrogen, are based on assumptions yet to be verified. Second, the availability of the resources in this country varies. For example, while the United States has ample supplies of coal for the foreseeable future, natural gas is already being imported to meet current demands. Thus comparison of the different approaches is complex and beyond the scope of this review. The hydrogen production program includes both long-term hydrogen supply from large, centralized production plants with pipeline distribution and supply during a transition when pipelines will not yet be in place. While it is clear that centralized plants will eventually provide most hydrogen at lower cost than other options, these plants and the necessary pipeline distribution system will not be available initially, when the number of hydrogen-fueled cars in operation will be small, although growing. In addition, once transitional hydrogen supply ap- proaches are in commercial use, it may be economical for the mature industry to continue to rely on them to supply some of the hydrogen needed, particularly in remote areas. Presently, there is no standard specifying a grade of hydrogen fuel that is acceptable for use in proton exchange membrane (PEM) fuel cell vehicles, and there is no all-inclusive list of maximum acceptable levels of contaminants in hydrogen. While a specification guideline has been issued (Ohi and Hewett, 2005), changes are likely to be made as more data become available. The levels of impurities that are acceptable could significantly affect the cost of hydrogen DOE has calculated that if 300 million vehicles with fuel economy of 60 mpg require hydrogen in 2040, 20 percent of that requirement could be provided by 2 trillion cubic feet (TCF) per year of natural gas, an increase of about 9 percent over the consumption in 2004 (DOE, 2005a). Likewise, producing 20 percent from biomass would require 140 to 280 million metric tons (MMT) (dry) of biomass compared with the 512 to 1,300 MMT currently available potentially from various sources. Similarly, 200 GW of installed wind power would be needed for hydrogen production by electrolysis compared with about 7 GW currently installed; 200 GW of photovoltaics compared with 5,400 MWe currently installed; and 80 GW of nuclear power compared with about 100 GW currently installed (DOE, 2005a). 

HYDROGEN PRODUCTION, DELIVERY, AND DISPENSING 85 and the overall efficiency of its production. DOE is well aware of this issue, and efforts are under way to resolve it. Production Technologies For centralized plants, the DOE hydrogen production program includes coal gasification, biomass gasification, electrolysis of water with wind energy or off-peak electricity, high-temperature water splitting with nuclear heat, and longer term approaches, including solar electrochemical and biological means. Existing commercial technologies can be used to convert natural gas or coal to hydrogen, and work currently under way at DOE, including the FutureGen program on coal with carbon sequestration, should reduce their costs moder- ately. Centralized production is visualized for each of these technologies and for natural gas reforming, distributed generation as well. All costs presented here exclude fuel sales taxes. Hydrogen Production from Coal Coal is an important potential resource for producing hydrogen because it is cost competitive and relatively abundant in the United States. At current pro- duction rates, the nation has over 200 years’ supply (see <http://gasprices-usa. com/coal.htm>). Efforts to develop and demonstrate hydrogen production from coal, including coal gasification and CCS, are managed by the DOE’s FE. The carbon sequestration subprogram is focused on developing, by 2012, technolo- gies that separate, capture, transport, and sequester carbon, increasing the cost of electricity by less than 10 percent. By that date, the program also plans to have developed a methodology capable for predicting CO2 storage capacity in a geologic formation to within ±30 percent. This program also has a number of regional partnerships, which include large-scale field tests, site evaluation work, site characterization R&D, collection of information to satisfy National Environ- mental Policy Act reviews, and other site-related activities to evaluate a variety of geologic formations for sequestration. The technologies developed by the carbon sequestration work will be used to benefit the existing and future fleet of fossil fuel power-generating facilities and provide key technologies and protocols for the FutureGen facility as it looks to capture, transport, store, and monitor the CO 2 injected in geologic formations. This arrangement—an important part of the program carried out in another program office—presents both management and technology challenges to the Hydrogen Fuel Initiative (HFI) and hence to the Partnership. This divided responsibility will require close liaison between the managers at For additional information on carbon capture and sequestration technologies and the research program, see <http://www.fossil.energy.gov/programs/sequestration/capture/>. 

86 review of the freedomcar and fuel partnership HFCIT and FE. In the Phase 1 report, the committee recommended strengthening this liaison. DOE concurred and has improved its communication and coordina- tion by taking a number of actions: • Setting up a hydrogen coordinating group composed of representatives from EERE, FE, NE, Basic Energy Sciences (BES), and the Department of Transportation (DOT); • Establishing an interagency working group to address hydrogen coordi- nation issues among federal agencies; and • Using its systems analysis capabilities to illuminate the implications for the Partnership of any cost and schedule issues that might arise in the FE program. The committee appreciates the value of these actions but notes that such mechanisms add value only insofar as they are used. It urges continued attention to building a highly effective liaison through these coordination arrangements. It believes as follows: • That CCS will pace the use of coal to produce hydrogen, and • That the technical and economic feasibility of capturing by-product CO2 and shipping it to permanent underground storage while producing electricity and hydrogen from coal will have to be demonstrated before significant commercial investment can materialize. A demonstration is being planned through FutureGen, a 275-MW, $1.8 billion integrated gasification combined cycle (IGCC) plant that will gasify the coal to produce electricity and hydrogen and sequester the resulting CO2. The committee believes that the general technology and system concepts embodied in FutureGen now offer the most promising way to produce hydrogen from coal while minimiz- ing CO2 release. And since FutureGen is now the principal platform for demon- strating a practical, commercial CCS system, its implementation will determine when the large-scale production of hydrogen from coal is introduced. To the extent that this project is delayed or fails to provide evidence accept- able to the public that CCS affords adequate protection at an acceptable cost, hydrogen production from coal will suffer corresponding delays. These delays could have multiple causes—for example, • Simple slippage in the FutureGen project schedule, a possible conse- quence of underfunding, unforeseen technical problems, siting difficul- ties, and so forth, • Issues arising from ambiguity surrounding regulatory authority, • Liability concerns, • The inability of FutureGen to provide a model for the commercial de- 

HYDROGEN PRODUCTION, DELIVERY, AND DISPENSING 87 velopment of CCS, much as the Power Reactor Demonstration Program of the 1950s failed to provide such a model for nuclear energy, or • Other difficulties, unforeseeable now but arising over the course of the project. Whatever their source, delays to CCS pose a risk to the Partnership and HFCIT goals for hydrogen production from coal. Recommendation.  DOE should conduct a systematic review of the CCS program as it affects the schedule for and program assumptions about hydrogen production from coal. This review should identify indicators of incipient program slippage and, through systems analysis, the program consequences of possible delays, leading to recommendations for management actions that would compensate for these delays. Hydrogen Production from Nuclear Heat NE seeks to demonstrate by 2017 the commercial-scale production of hy- drogen using heat from a nuclear energy system. Some advanced nuclear reactor designs operate at very high temperatures, making them well suited for thermally driven hydrogen production processes. These high-temperature reactors remain in early development by the Generation IV Nuclear Energy Systems Initiative (Generation IV) and could provide the low-cost heat necessary to produce low- cost hydrogen. The nuclear hydrogen program is managed under three technology thrusts:  • Thermochemical water-splitting cycles. Thermochemical cycles convert water to hydrogen and oxygen using chemical catalysts at high tempera- tures. These processes have the potential for high-efficiency hydrogen production on a large scale, but the technology remains in a very early stage. • High-temperature electrolysis. Also called steam electrolysis, this tech- nology uses electricity to produce hydrogen from steam instead of from liquid water. It promises higher efficiencies than standard electrolysis, which might be used at the forecourt of fueling stations during a hy- drogen transition. This, too, is in an early stage, and the chief technical challenges include the development of high-temperature materials and membranes. • Reactor/hydrogen process interface. The interface between the nuclear reactor and the hydrogen production system presents severe challenges to any working system—long heat-transfer paths at elevated temperatures; The program is summarized at <http://www.hydrogen.energy.gov/nuclear.html>. 

88 review of the freedomcar and fuel partnership heat exchangers that are subject to elevated temperature and a corrosive chemical environment; new safety and regulatory issues; and support systems for chemical processes and hydrogen and oxygen storage. In ad- dition, systems studies seek to focus this complex program and improve coordination. In all of these research areas, much basic work must be completed before a development and demonstration program can be properly contemplated. Nev- ertheless, nuclear production of hydrogen remains an important option—it is potentially lower in cost and could be a hedge against delays in CCS technologies or against coal shortages. Recommendation.  Like the hydrogen production from coal option, the Hydro- gen, Fuel Cell and Infrastructure Technology (HFCIT) program should actively employ the liaison mechanisms put in place since the Phase 1 review. However, the exploratory nature of the programs for nuclear production suggests that, unlike the coal option, a detailed systems analysis of schedule delays would be premature at this time. Instead, systems analyses should focus on the complex interactions among program components, especially between the research elements of the nuclear and chemical processes, to ensure that technical progress in each distinct area leads ultimately to a practical system. Hydrogen from Electrolysis The electrolysis of water, though energy intensive, is one of the few options under consideration for distributed, on-site, point-of-use production and delivery of hydrogen when conventional sources and processes are not available. When coupled with a renewable power generation scheme such as for wind or solar power, the overall advantages are considerable, especially when carbon emissions are taken into account. However, the relative siting of the power generation and electrolysis devices is an issue since a location suitable for, say, a wind farm might not be suitable for the hydrogen generator. Centralized electrolysis processes are also under development to reduce operating and capital costs. DOE recognizes the importance of electrolysis and has been engaged in facilitating new concepts, advances, demonstrations, and analyses. Clearer insight into the challenges of the program cost targets—for distributed generation, $3.70/gallon of gasoline equivalent (gge) in 2012 and <$3.00/gge in 2017, assuming grid electricity costs $0.05/kWh and units that produce 1,500 kg H2/day; for centralized generation with wind energy, $3.10/gge in 2012 and <$2.00/gge in 2017, excluding delivery DOE, Hydrogen Production Technical Team, Presentation (Slide 27) to the committee on March 2, 2007. 

HYDROGEN PRODUCTION, DELIVERY, AND DISPENSING 89 costs—has been gained from the recent analysis efforts. It is still too early to predict the probability of meeting the longer-term targets. The budgets for electrolysis R&D have been increasing ($1.6 million in FY05, $3.5 million in FY07). Although funding has more than doubled, it is not enough, as will be discussed. Conventional water electrolysis technology is more mature than other processes, in part because extensive operating knowledge, sys- tems design and engineering, and applications have been in place for decades, spe- cifically in military applications (submarines). Large commercial processes have also been available. As a result, proof-of-concept programs are not warranted, but there are significant matters—cost, systems integration, analyses, and field trial results—that need to be better understood. The NREL wind-electrolysis demon- stration (Harrison, 2007), as well as its source-to-wheels analyses, has made a significant contribution, and its results have led to the refinement of targets. The technology is challenged primarily by costs, in particular the cost of electrical power to split the water. The fundamental energy requirements for this process will not change, but overall system costs are addressable. Sensitivity and trade-off analyses will be needed to delineate the most attractive scenarios, in part because two distinctly different technologies (membrane and alkaline) are under consideration. Because power requirements are nearly fixed, the potential for capital cost reduction for each technology will be an important outcome of these sensitivity analyses. Both technologies have extensive histories: Membrane electrolysis offers the advantage of high hydrogen generation rates and efficien- cies and the promise of further enhancing these rates (thereby reducing capital outlay), whereas alkaline systems have track records for lifetime, reliability, and lower capital costs. Both technologies have demonstrated high-pressure genera- tion capability, and both lend themselves to minimizing downstream cleanup, storage, and compression. It is too early to predict the outcome of solid oxide electrolysis. There are approximately eight funded PEM, alkaline, and solid oxide elec- trolysis projects, as reported by General Electric at the DOE 2007 merit reviews.  These projects are engaged in analyses, component development, and demonstra- tions. Additional ongoing research looking at longer-term possibilities such as the photoelectrolysis of water is basic in nature. The efforts are aimed at reducing the capital costs of the hardware and the number of parts and at finding new membranes, catalysts, and materials of construction. Although such initiatives are appropriate, the costs of electrolysis will always be dictated by the power require- ments for splitting water, which limits what is achievable by reducing hardware costs. To date, the mass manufacture of electrolyzers has not been addressed in the commercial sector because the market is still small. Once manufacturing ac- R.Bourgeois, GE, “Advanced alkaline electrolysis,” Presentation to the DOE 2007 Annual Merit Review in May 2007. Available on the Web at <http://www.hydrogen.energy.gov/pdfs/review07/ pdp_16_bourgeois.pdf>. 

90 review of the freedomcar and fuel partnership tivities begin to take off, they will probably contribute significantly to reducing costs. In addition, the considerable overlap between membrane electrolyzers and membrane fuel cell components enhances supply chain strengths and capabilities and facilitates the development of new materials (e.g., membranes). Recommendation.  The DOE should continue to promote electrolysis that uses renewable power integrated with electrolysis systems and to support analyses and demonstrations. High-temperature electrolysis activities within the Office of Nuclear Energy should be closely monitored. Recommendation.  The Partnership should increase funding for electrolysis pro- grams to advance the technology, demonstrations, and systems integration. Recommendation.  Basic Energy Sciences should support, as appropriate, fundamental research in the area of catalysts, membranes, coatings, and new concepts. As mentioned earlier, a population center where distributed hydrogen pro- duction would be needed is not usually in a place where significant electricity is generated from the wind or the Sun. As a result, it is not clear, based on current understanding, how a generator of power from the wind should be situated rela- tive to a generator of hydrogen to maximize the benefits. For example, depending on the specifics of the location, it might be more efficient to generate electricity in a wind farm and transport it over the grid to distributed electrolyzers than to cogenerate power and hydrogen and transport the hydrogen to the fueling site. Likewise, the extent to which wind power could be generated at the site of a distributed hydrogen generator is unclear. Recommendation.  DOE should undertake a systems study to determine how best to use wind power–electrolysis combinations to generate hydrogen, considering overall cost and efficiency. Hydrogen Production from Biomass Biomass is a renewable and potentially sustainable source of liquid fuels and hydrogen. A comprehensive study jointly sponsored by DOE and the Department of Agriculture (DOE, 2005b) concluded that the United States has sufficient land resources to sustain production of biomass to supply 30 percent or more of the nation’s current consumption of liquid transportation fuel. To achieve that level of production, it is assumed that three times more forest biomass than today will be collected; that crop yield will be increased by 50 percent and recovery of crop residues by 75 percent; that 55 million acres will be dedicated to the production of perennial bioenergy crops; and that other non-farm-use residues will be converted 

HYDROGEN PRODUCTION, DELIVERY, AND DISPENSING 91 to biofuels. In addition, a cost-effective and energy-efficient process to convert cellulosic material to biofuel will be needed, which requires innovation at various stages, including crop production and biomass degradation. DOE has set goals to supply 20 percent of liquid transportation fuel by 2017 and 30 percent by 2030. These are ambitious goals, and DOE is investing sub- stantial resources in R&D to achieve them, including a FY08 request for $375 million. The FreedomCAR and Fuel Partnership includes only the technology for conversion of biomass to hydrogen. The committee did not attempt a comprehen- sive review of the program and budget on biomass growth, harvesting, collection, and transportation to conversion plants. However, it is clear that there are signifi- cant hurdles to overcome, and it will be a stretch to achieve these goals (DOE, 2006a). Work to date has not established how much can be recovered sustainably at target costs and by target dates. To make this estimate involves resolving vari- ous technical barriers that DOE has identified relating to biomass availability and cost (DOE, 2003), as well as land and water use issues and competition for both resources. The committee believes that the impact of biomass on the supply of hydrogen cannot be reliably estimated until programs relating to biomass produc- tion, harvesting, collection, storage, preprocessing, and transportation can define commercially viable pathways from crops or other biomass sources to hydrogen production plants and until the specific government-sponsored incentives become clear, along with land and water use policies that may be required to stimulate wider use of this option. Resolving these issues will require the involvement of other government departments, including the Department of Agriculture. The committee believes that early definition of government-sponsored commercial incentives and land and water use policies would help facilitate the later develop- ment of appropriate government actions in these important areas. Nonetheless, the Partnership is anticipating the possibility of a significant increase in the use of bioethanol in transportation fuel (say, E85 or E15) in the near future, from about 1.7 percent today (DOE, 2006b), and the potential im- pact on combustion engine technology of such an increase. In response to this, DOE initiated solicitation DE-PS26-07NT43103, which includes development of engines for flexible-fuel, light-duty vehicles (FFVs) optimized for operation on ethanol-gasoline blends up to 85 percent ethanol by volume. This is in line with Recommendation 3-3 of the Phase 1 report. Hydrogen can be produced by gasifying the biomass feedstock directly or by gasifying one of the conversion products, such as ethanol. These conversion tech- nologies are known, but the current cost of production is high, $7.00/kg hydrogen (H2) at a feedstock cost of $53/ton. DOE projects that a lower feedstock cost, more energy-efficient production, and cost reduction due to a larger scale of opera- tion will result in a cost of $3.50/kg H2. The committee believes that $3.50/kg is These estimates are taken from The Hydrogen Economy and are for hydrogen delivered by tank trucks from a high-pressure oxygen gasifer producing 24,000 kg H 2/day (NRC/NAE, 2004). 

92 review of the freedomcar and fuel partnership a stretch goal that could be achieved only after overcoming significant technical and policy hurdles, as described earlier. Thus the extent to which biomass will become a source of hydrogen is highly uncertain, and DOE should continue to investigate a broad portfolio of hydrogen production technologies. As part of the effort to reduce gasification costs, DOE is now considering a 155,000 kg H2/day Battelle biomass gasification plant as opposed to the 24,000 kg H2/day high-pressure gasifier that was assumed in The Hydrogen Economy (NRC/NAE, 2004). Neither operation has yet been fully demonstrated. The much larger plant needs 2,425 tons of biomass/day with current technology and 2,125 tons/day with projected future technology compared with 442 tons/day biomass for current technology and the smaller plant. The larger plant needs 307 square miles to support it with current technology and 180 square miles with future technology. The number of sites that could support such a large biomass plant and still have acceptable delivered biomass cost and delivered hydrogen cost needs to be determined. DOE estimates there are roughly 50 potential sites throughout the country with current biomass yields and upwards of 100 sites that could yield biomass with future crop technology. This needs to be confirmed. Biomass gasification technology is promising, but much remains to be done to put it on a solid basis. How the different types of biomass feedstock must be prepared for ease of delivery and reliable processing in the gasifier needs to be determined. Bench-scale, pilot plant, and semicommercial-scale work are needed to have a firm basis for scale-up to a 2,125 tons/day plant or larger. Also, gas cleanup and separation into pure hydrogen needs to be demonstrated while deal- ing with contaminants and tar. The committee judges that it will be difficult to achieve this technology by 2017, and several years more may be needed. However, if successful, hydrogen supply from biomass gasification could supplement other sources of hydrogen, and the committee continues to believe that this program is a very important element in the portfolio of hydrogen production technologies. The impact of biomass on future hydrogen supply is difficult to evaluate, in part because there are so many alternative paths. In addition to gasification of bio- mass to hydrogen there are other pathways that may be reasonable. For instance, gasifying biomass to make an alcohol mixture and then reforming the alcohol to hydrogen at distributed locations would eliminate the need to distribute hydrogen itself. Distributed reforming of cellulosic ethanol, aqueous glucose, or aqueous lignins is another possibility. Additionally, biological methods discussed in The Hydrogen Economy (NRC/NAE, 2004) but still in the basic science phase, could one day be significant. Finally, cofeeding biomass with coal in a coal gasifica- tion process might eventually be attractive. With the current state of knowledge, the committee believes it is not yet possible to identify the preferred approaches and encourages DOE to focus its program efforts on studies that will enable this identification. The committee notes that while DOE has involved the agribusiness partners in the biomass program, the energy partners are involved primarily in the conversion 

HYDROGEN PRODUCTION, DELIVERY, AND DISPENSING 93 or gasification step and not in the harvesting and so on of the biomass feedstock. The committee believes that it would be important to bring the commercial per- spective of the energy partners to all aspects of the biomass program, since these partners have long experience with creating and supplying transportation fuels from natural resources and have been involved in converting coal and other solid energy sources to transportation fuels. Recommendation.  The committee recommends that DOE projections of future hydrogen production for hydrogen-powered vehicles include scenarios in which the timetable for commercial quantities of these fuels is delayed, perhaps by as much as a decade. Recommendation.  DOE should give priority to completing process development on biomass gasification, including any needed demonstration projects. Recommendation.  DOE should undertake a systems study to assess the relative importance of barriers to biomass production, availability, transportation, and con- version to hydrogen; to identify the areas that are most important to commercial viability; and to give them priority. This study should address technical barriers already identified, including impact on the environment, and help define policies for land and water use and government-sponsored commercial incentives that would stimulate commercial expansion of the biomass options. Recommendation.  DOE should involve the energy partners in all biomass pro- grams related to conversion to hydrogen or hydrogen carriers as quickly as possible. Recommendation.  Given the large number of potential ways of using biomass to supply hydrogen, DOE should identify the most promising approaches so it can focus on options that could have the greatest impact on hydrogen supply. Special Production Considerations During Market Transition No one knows just how hydrogen will be supplied during the transition period when fuel-cell-powered cars first become available. However, it is reasonable to expect that it will first be supplied to fueling stations from existing centralized sources and distributed as a gas by tube trailer or as a liquid by carrier, since a pipeline distribution system similar to the system for natural gas will not initially be available. These supplies are likely to be supplemented increasingly with time, by facilities at forecourts to produce hydrogen locally, with no need for distribu- tion, using steam reforming of natural gas from the existing supply system or electrolysis with electricity from the grid. As the demand for hydrogen grows, its distribution by pipeline from centralized sources to forecourts will become 

94 review of the freedomcar and fuel partnership economically attractive in highly populated areas, and it will be used more and more. However, in some remote areas such a pipeline system may never become economical, owing to low demand, and the sources of fuel used at the beginning of the transition could continue to be used. Three approaches to the forecourt generation of hydrogen are being studied. They involve the reforming of natural gas taking advantage of the existing natural gas distribution system; the reforming of ethanol or other bioderived liquids; and the electrolysis of water. Dehydrogenation of a carrier liquid that is subsequently returned to a refinery or chemical plant for rehydrogenation is another option. DOE has made substantial progress in understanding the transition to a sus- tainable market, defining requirements for forecourt production systems based on natural gas reforming that meet initial cost targets and advancing other options for onsite generation. Natural gas reforming is well-established technology, and program efforts have been directed toward the specific requirements of practicing it at fueling station sites in relatively small units (e.g., 1,500 kg H2/day). DOE has established that hydrogen can be produced from natural gas in an integrated system for the target cost of $3.00 per gallon gasoline equivalent (gge) and has a cost goal of $2.50/gge by 2010. A study carried out for DOE has estimated that the natural gas required in 2025 to fuel 11.6 million hydrogen vehicles in the 27 largest U.S. cities would increase gas demand on average by 2.1 percent over the demand in 2004, although this percentage would vary from city to city, and the cost of additional gas transmission lines to transport that gas might cost $1.0 billion to $1.5 billion (Energy and Environmental Analysis Inc., 2006). There is much uncertainty regarding the effect of small increases in natural gas demand on price, but higher demand would likely increase natural gas imports. Given that 11.6 million vehicles would be only about 15 percent of the census vehicle population in those 27 cities, natural gas price and supply could become an important issue as the penetration of hydrogen cars increases in those 27 cities and expands into smaller communities, assuming all the hydrogen is made from natural gas. The foregoing discussion highlights the importance of having other ways of making hydrogen for the transition. Important issues remain in the design of reformer systems for forecourt use. These issues could substantially constrain all approaches to distributed hydrogen generation at forecourts, probably limiting hydrogen availability in the early years of its commercial introduction. The current design requires 6,500-7,000 square feet of space, and local regulations will require additional setbacks from adjacent structures. Based on a sample of 120 existing automobile fueling sta- tions in New York, Los Angeles, and Dallas, DOE concluded that it would not be feasible to place onsite reforming in 40 percent of them (49 stations) and would Assumes the manufacture of 500 reformer units per year, each with a production of 1,500 kg H 2/ day. This assumes natural gas at $5 per million Btu, a capacity factor of 70 percent, and a production efficiency (energy content of H2/input energy content of natural gas) of 69 percent. 

HYDROGEN PRODUCTION, DELIVERY, AND DISPENSING 95 be clearly feasible in only 13 percent (16 stations). The committee believes that the DOE estimate of 16 clearly feasible sites is optimistic because of the inher- ent difficulties of retrofitting an existing site with new facilities and that an even lower percentage of sites would, in fact, be feasible with present design and safety considerations. DOE is aware of the importance of reducing the space require- ment and is considering alternative designs, including smaller reformers and less hydrogen storage. Because the budget request for FY08 includes no funds for distributed natural gas reforming, additional funding will be needed to complete this important work. The committee believes that the target cost of $2.50/gge for distributed gen- eration is very optimistic based on current technological routes, particularly in view of the need to reduce the size of the generator to minimize forecourt space requirements. It believes, accordingly, that DOE needs to reevaluate this target taking into consideration the constraints and approaches available for improve- ment as well as the latest gasoline price outlook. Recommendation.  DOE should put more emphasis on the space require- ments for forecourt hydrogen generation by studying ways to minimize these requirements. HYDROGEN DELIVERY, DISPENSING, AND TRANSITION SUPPLY Overview Unlike the traditional petroleum delivery system, whereby gasoline and diesel fuel are delivered from refineries to fueling stations and stored there at relatively low cost and low energy consumption, the system for delivering hydrogen from central production to a refueling station with subsequent storage and dispensing to vehicles will be a significant factor in hydrogen fueling. Similarly, in a fully developed hydrogen economy, delivery, storage, and dispensing at a high pres- sure will probably cost as much as production and will consume more energy. Distribution costs are even more of a concern during the early transition years, when there is a lack of hydrogen demand, particularly where central production will be the source of hydrogen. Identifying central hydrogen supply during the transition, whether excess capacity or dedicated supply, could provide significant opportunities to ease the transition to a hydrogen economy. It is likely that some of the needed hydrogen would be supplied from existing facilities. To achieve the lower sulfur levels of conventional fuels (e.g., gasoline and diesel) when refining heavier crude oils, the industry will have to increase its hydrotreating capacity from 14 million barrels per day in 2004 to over 27 million DOE, Answers to questions from committee, p. 59, received April 17, 2007. 

96 review of the freedomcar and fuel partnership in 2030. At least part of this hydrogen will be generated by gas supply companies for over-the-fence sales to refineries, and these gas companies, which already pro- duce and deliver gaseous and liquid hydrogen, are likely to see hydrogen vehicles as an attractive additional market. There are five main ways to deliver hydrogen from central supply to refuel- ing stations: • Pipeline delivery. This requires storage at the production site, laying pipeline to the forecourt, and onsite storage as a gas at the forecourt. It is the lowest cost route and utilizes pipeline delivery technology that is well known and has been in commercial use for decades. Energy losses with a pipeline are less than those with the other delivery methods (see Table 4-2), but pipeline delivery probably will take the longest time to implement because of permitting and rights of way. Production and forecourt storage costs should be lowest of all supply modes since the pipeline itself is the vessel that stores the surge capacity. • Liquid delivery. This involves liquefaction at the production site, delivery of the liquid by truck to the forecourt, liquid storage at the refueling site, and dispensing as a high-pressure gas to the vehicle after vaporization and compression. Liquefaction increases the cost of hydrogen production and decreases its efficiency by approximately 16 percent (see Table 4-2). However, the high density of the liquid allows the highest payload by weight of hydrogen to be moved by truck. Furthermore, storage at the site of high-pressure gas can be a significant forecourt issue because of the space required, and liquid minimizes this. Since the basic technology has been practiced for a long time, only incremental improvements are expected in liquefaction and distribution, and these costs will probably continue to be about 50 percent more than pipeline costs. • Gaseous delivery. This requires high-pressure gas storage at the produc- tion site, delivery by high-pressure tube trailers, and high-pressure gas storage at the forecourt. Currently, the lower delivery density makes this option impractical for stations that experience high demand for hydro- gen. As a result, it takes 12 to 15 high-pressure tube trailers to deliver the same payload as one liquid hydrogen trailer. In addition, because today’s gaseous delivery technology costs several times as much as the technology for pipeline delivery depending on volume, it may not be cost-effective. New technology developments focus on increasing the payload using cryogenic gaseous hydrogen storage—that is, storage as a gas at low temperatures and/or higher pressures. • One-way liquid carrier delivery. This requires methanol, ethanol, or a similar hydrocarbon and would require reforming at the forecourt to DOE, Answers to questions from committee, p. 31, received April 17, 2007. �������������������������������������������������������������������� 

HYDROGEN PRODUCTION, DELIVERY, AND DISPENSING 97 TABLE 4-2 Delivery and Dispensing Energy Efficiencya Efficiency (%) Well (Source) to Fuel/Delivery Mode Well (Source) to Tank Tank to Wheels Wheels Overall Gasoline 81 17 14 H2/pipeline 64 41 26 Liquefaction 48 41 20 aCalculations are based on the Argonne National Laboratory’s Greenhouse Gases, Regulated Emissions, and Energy Use in Transportation (GREET) model. For analyses of a variety of fuels, see results at <http://www.transportation.anl.gov/pdfs/TA/273.pdf>. generate hydrogen. Commercial methanol processes exist, and ethanol reforming processes are being developed. While methanol and ethanol reforming onsite may be more expensive than natural gas reforming, they could have some advantages (costs and efficiency) from a delivery stand- point, particularly during the transition. The higher cost of reforming at the forecourt could more than offset the high costs for liquid or gaseous hydrogen delivery, on a source-to-tank basis. For example, ethanol is a dense liquid containing 12 percent hydrogen that can probably be trans- ported with existing infrastructure. • Two-way liquid carrier delivery. Such liquid carriers could be hydro- genated to more than 13 percent hydrogen content, transported to the forecourt, dehydrogenated, and returned to the hydrogenation site for rehydrogenation. With current technology, they can be hydrogenated to 7-8 percent hydrogen. This is a long-term option that requires R&D and systems analysis to develop an understanding of energy and cost issues. For example, if the liquid could also serve as onboard storage for hydrogen, the overall system would look something like the gasoline system but would increase vehicle complexity and cost. With delivery and dehydrogenation to only the forecourt, there could be some advan- tages, particularly where eventual pipeline delivery might be difficult or costly. The DOE Program DOE’s plan for delivery, storage, and dispensing is robust and was developed with aggressive cost targets (Table 4-3). The goal is to reduce the cost of delivery plus dispensing to less than $1.00 per kilogram hydrogen by 2017. This compares to the current costs of $3-$4/kg H2 at low volume and $2-$3/kg H2 at high volume. Given that hydrogen pipeline, truck delivery, compression, and storage technolo- gies have been practiced for decades by the gas industry, the committee questions whether it will be possible to reduce costs by a factor of 2 or 3. 

98 review of the freedomcar and fuel partnership TABLE 4-3 Cost Targets for Hydrogen Delivery and Dispensing (dollars per kilogram of hydrogen) Activity 2010 2012 2015 2017 Delivery from central <0.90 <0.60 plant to refueling gate Dispensing at refueling <0.80 <0.40 sitea aIncludes compression and storage. SOURCE: Based on J. Kegerreis and M. Paster, DOE, delivery technical team, Presentation to the committee on March 1, 2007. The program for delivery, storage, and dispensing has been slow to start up, especially on the delivery side, due to congressionally directed funding in the overall HFI. This very important program has been consistently underfunded since HFCIT started, with $16.9 million budgeted but only $8.3 million funded in the 2004-2007 period. It appears that the DOE is working well with gas companies and the Partnership’s delivery technical team (DTT). While the program is at risk as a result of past underfunding, some very important analysis work has been ac- complished. The development of the Lighthouse concept for market penetration has been a significant accomplishment and helps the DTT team focus on specific delivery, dispensing, and early supply options and issues. Also, the completion of the H2A model, whose components and submodels define delivery and dis- pensing, is significant and will help to better delineate and evaluate scenarios for getting hydrogen to fuel cell vehicles during the transition and later. The H2A model has also played a significant role in developing the research plan. Finally, analytical work on forecourt issues has progressed—for instance, the very excel- lent analysis of overall U.S. natural gas supply and demand and of the issues involved in getting natural gas to the refueling station for on-site steam methane reforming (SMR). The future program is built around the DTT Roadmap, which identifies the following key challenges: • Pipelines. Metal embrittlement, capital cost reduction, urban distribution issues, composite materials for construction, use of existing natural gas pipelines. • Compression. Reliability/durability, new technologies, and the energy efficiency and size of the refueling station. • Liquefaction. Dramatic cost reduction, dramatic increase in energy ef- ficiency, boil-off. • Off-board storage. Lower capital costs, cryogas, other hydrogen carriers, suitability of geologic storage. 

HYDROGEN PRODUCTION, DELIVERY, AND DISPENSING 99 TABLE 4-4 Budgets for Hydrogen Delivery Activities (millions of dollars) FY04 FY05 FY06 FY07 FY08 Budget request 1.0 3.8 5.9 6.2 8.0 Pipelines 1.8 Compression 0.7 Storage 0.8 Liquefaction 1.2 Carriers 1.0 Analysis 0.7 Expenditures 0.4 2.8 1.1 4.0 (spend rate) SOURCE: J. Kegerreis and M. Paster, DOE, “Hydrogen delivery,” Presentation to the committee on March 2, 2007. • Gaseous tube trailers. Increase capacity fourfold with higher pressure, cryogas, or other hydrogen carriers. • Carriers. Liquid one-way and two-way carriers; solid carriers. The DOE budget for the program is shown in Table 4-4. Overall, a lot has been accomplished, but much more progress is needed. The Partnership’s DTT has identified important R&D areas for improving the cost and energy efficiency of the delivery and dispensing of hydrogen. However, in light of how important delivery is to both the market transition and sustained market penetration time frames, it deserved more funding and attention. The program is most likely un- derfunded even at the FY08 $8 million level requested. Recommendation.  DOE should increase funding for the delivery and dispens- ing program to meet the market transition and sustained market penetration time frames. If DOE concludes that a funding increase is not feasible, the program should be focused on the pipeline, liquefaction, and compression programs, where a successful if only incremental short-term impact, could be significant for the market transition period. Recommendation.  DOE should, with the guidance of an independent out- side committee, evaluate the achievability of the program’s 2012 delivery and dispensing cost goal, $1.00/kg H2, particularly with 700 bar (10,000 psi) gas dispensing. Home Refueling One path that could reduce or even eliminate the need for a hydrogen distribu- tion and delivery infrastructure is home refueling, which would allow consumers to refuel their vehicles at home. Furthermore, if additional benefits could be at- 

100 review of the freedomcar and fuel partnership tained with such a device, then the value would be much greater—if, for example, the system could provide both onsite heat and power. So far, two approaches have been proposed for refueling hydrogen vehicles at home: (1) integrating the hydrogen generation and delivery operations with a stationary fuel cell system that generates electricity for the home and is fueled by a natural gas (propane) or by a liquefied propane gas (LPG) reformer or (2) a home-based water electrolysis unit. To demonstrate these approaches to delivering hydrogen to fuel cells, Honda (<http://world.honda.com/FuelCell/FCX/station/>) has units running in Califor- nia and in New York, while GM (<http://www.usatoday.com/money/autos/2006- 09-24-gm-hydrogen-usat_x.htm>) and others have announced plans for similar devices in the years to come. To provide fuel to the car, a slipstream of the hy- drogen used to generate home electricity is diverted to downstream purification and compression stages, followed by storage and dispensing to the vehicle. The fuel cell brings the added value of distributed heat and electrical power to the site while enabling the generation, storage, and dispensing of hydrogen at the convenience of the home owner. From a technology perspective, many of the hydrogen subsystems of the devices (generation, delivery, storage) are based on the same technologies under development in other programs currently funded by DOE, including the fuel cell itself. Water electrolysis home refueling is based on a conventional electrolytic process followed by purification, compression, and storage stages. Although ap- parently simpler from a hydrogen generation perspective, the limitations of the water electrolysis process (it requires relatively high investment and quantities of electricity) must be taken into account along with the challenges of purifica- tion, compression, and storage, as in the previous case. One more factor to be considered in this case is the cost of power to electrolyze water. As in the afore- mentioned case of fuel cells, DOE is supporting selected aspects of electrolysis technologies. Regardless of the path chosen—electrolysis or reformation/fuel cell—the home refueling device does not require additional component development ini- tiatives, as it will continue to benefit from advancements made in such areas under the existing efforts. Additional funding is therefore not required from this perspective. Integration, demonstrations, and siting will need to be addressed as will safety, permitting, and codes and standards of a hydrogen-based process in residential environments. Recommendation.  DOE should consider supporting advanced systems engineer- ing, integration, and demonstrations for home-based refueling systems, which should bring substantial learning value for such systems. This program should include careful consideration of operation and maintenance procedures that home owners are willing and able to perform. 

HYDROGEN PRODUCTION, DELIVERY, AND DISPENSING 101 REFERENCES DOE (Department of Energy). 2003. Roadmap for Agricultural Biomass Feedstock Supply in the United States, DOE/NE-ID-11129, Revision 1, November 2003. DOE. 2005a. Multi-Year Research, Development and Demonstration Plan: Planned Program Activi- ties for 2005-2015. Section 2.3, Domestic Resources for Hydrogen Production. Washington, D.C.: Office of Energy Efficiency and Renewable Energy. Available on the Web at <http:// www1.eere.energy.gov/hydrogenandfuelcells/mypp/>. DOE. 2005b. Biomass as Feedstock for a Bioenergy and Bioproducts Industry: The Technical Feasi- bility of a Billion-Ton Annual Supply. April, DOE/GO-102005-2135, Robert D. Perlack, Lynn L. Wright, Anthony F. Turhollow, Robin L. Graham, Bryce J. Stokes, and Donald C. Erbach. DOE. 2006a. Breaking the Biological Barriers to Cellulosic Ethanol: A Joint Research Agenda. DOE/SC-0095. DOE. 2006b. Annual Energy Outlook 2006. Washington, D.C.: Energy Information Administration. Energy and Environmental Analysis, Inc. 2006. Initial Look at Potential Natural Gas Infrastructure Constraints Related to Transition to Hydrogen Transportation Fuels. September 13. Harrison, K. 2007. Renewable Electrolysis Integrated System Development and Testing. Annual Merit Review Proceedings. Arlington, Virginia, May. Available on the Web at <http://www.hydrogen. energy.gov/pdfs/review07/pd_8_harrison.pdf>. Ohi, J.M., and R. Hewett. 2005. The Department of Energy’s Hydrogen Safety, Codes and Standards Program: Status Report on the National Templates. Society of Automotive Engineers Paper SAE J2719, November. National Research Council/National Academy of Engineering (NRC/NAE). 2004. The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs. Washington, D.C.: The National Academies Press.