Appendix G
Hydrogen Production Technologies: Additional Discussion

This appendix discusses in more detail the technologies that can be used to produce hydrogen and which are addressed in Chapter 8. Cost analyses for them are presented in Chapter 5. In this appendix, the committee addresses the following technologies: (1) reforming of natural gas to hydrogen, (2) conversion of coal to hydrogen, (3) nuclear energy to produce hydrogen, (4) electrolysis, (5) wind energy to produce hydrogen, (6) production of hydrogen from biomass, and (7) production of hydrogen from solar energy. The following major sections—one for each of the technologies—include a brief description of the current technology; possible improvements for future technology; refer to Chapter 5 and Appendix E (which presents spreadsheet data from the committee’s cost analyses), where applicable, for the current and possible future costs, CO2 emissions, and energy efficiencies; note the potential advantages and disadvantages of using the technology for hydrogen production; and comment on the Department of Energy’s (DOE’s) research, development, and demonstration (RD&D) plan for hydrogen.

In general, in developing estimates about future possible technologies, the committee systematically adopted an optimistic posture. The estimates are meant to represent what possibly could be achieved with concerted research and development (R&D). But the committee is not predicting that the requisite R&D will be pursued, nor is it predicting that these technical advances necessarily will be achieved, even with a concerted R&D program. Estimates were made of what might be achieved with appropriate R&D.

The state of development referred to as “possible future” technologies is based on technological improvements that may be achieved if the appropriate research and development are successful. These improvements are not guaranteed; rather, they may be the result of successful R&D programs. And they may require significant technological breakthroughs. Generally, these possible future technologies are available at a significantly lower cost than are the “current technologies” using the same feedstocks.

HYDROGEN FROM NATURAL GAS

Compared with other fossil fuels, natural gas is a cost-effective feed for making hydrogen, in part because it is widely available, is easy to handle, and has a high hydrogen-to-carbon ratio, which minimizes the formation of by-product CO2. However, as pointed out elsewhere in this report, natural gas is already imported as liquefied natural gas (LNG)1 into the United States today, and imports are projected to increase. Thus, increased use of natural gas for a hydrogen economy would only increase imports further. As a result, the committee considers natural gas to be a transitional fuel for distributed generation units, not a long-range fuel for central station plants for the hydrogen economy.

Production Techniques

The primary ways in which natural gas, mostly methane, is converted to hydrogen involve reaction with either steam (steam reforming), oxygen (partial oxidation), or both in sequence (autothermal reforming). The overall reactions are shown below:

CH4 + 2H2O → CO2 + 4H2

CH4 + O2 → CO2 + 2H2

In practice, gas mixtures containing carbon monoxide (CO) as well as carbon dioxide (CO2) and unconverted methane (CH4) are produced and require further processing. The reaction of CO with steam (water-gas shift) over a catalyst produces additional hydrogen and CO2, and after purification, high-purity hydrogen (H2) is recovered. In most cases,

1  

Importation of large amounts of LNG would require major investments to provide LNG marine terminals and related infrastructure. These would be potential targets for terrorist attacks, which would threaten the security of LNG supplies.



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 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs Appendix G Hydrogen Production Technologies: Additional Discussion This appendix discusses in more detail the technologies that can be used to produce hydrogen and which are addressed in Chapter 8. Cost analyses for them are presented in Chapter 5. In this appendix, the committee addresses the following technologies: (1) reforming of natural gas to hydrogen, (2) conversion of coal to hydrogen, (3) nuclear energy to produce hydrogen, (4) electrolysis, (5) wind energy to produce hydrogen, (6) production of hydrogen from biomass, and (7) production of hydrogen from solar energy. The following major sections—one for each of the technologies—include a brief description of the current technology; possible improvements for future technology; refer to Chapter 5 and Appendix E (which presents spreadsheet data from the committee’s cost analyses), where applicable, for the current and possible future costs, CO2 emissions, and energy efficiencies; note the potential advantages and disadvantages of using the technology for hydrogen production; and comment on the Department of Energy’s (DOE’s) research, development, and demonstration (RD&D) plan for hydrogen. In general, in developing estimates about future possible technologies, the committee systematically adopted an optimistic posture. The estimates are meant to represent what possibly could be achieved with concerted research and development (R&D). But the committee is not predicting that the requisite R&D will be pursued, nor is it predicting that these technical advances necessarily will be achieved, even with a concerted R&D program. Estimates were made of what might be achieved with appropriate R&D. The state of development referred to as “possible future” technologies is based on technological improvements that may be achieved if the appropriate research and development are successful. These improvements are not guaranteed; rather, they may be the result of successful R&D programs. And they may require significant technological breakthroughs. Generally, these possible future technologies are available at a significantly lower cost than are the “current technologies” using the same feedstocks. HYDROGEN FROM NATURAL GAS Compared with other fossil fuels, natural gas is a cost-effective feed for making hydrogen, in part because it is widely available, is easy to handle, and has a high hydrogen-to-carbon ratio, which minimizes the formation of by-product CO2. However, as pointed out elsewhere in this report, natural gas is already imported as liquefied natural gas (LNG)1 into the United States today, and imports are projected to increase. Thus, increased use of natural gas for a hydrogen economy would only increase imports further. As a result, the committee considers natural gas to be a transitional fuel for distributed generation units, not a long-range fuel for central station plants for the hydrogen economy. Production Techniques The primary ways in which natural gas, mostly methane, is converted to hydrogen involve reaction with either steam (steam reforming), oxygen (partial oxidation), or both in sequence (autothermal reforming). The overall reactions are shown below: CH4 + 2H2O → CO2 + 4H2 CH4 + O2 → CO2 + 2H2 In practice, gas mixtures containing carbon monoxide (CO) as well as carbon dioxide (CO2) and unconverted methane (CH4) are produced and require further processing. The reaction of CO with steam (water-gas shift) over a catalyst produces additional hydrogen and CO2, and after purification, high-purity hydrogen (H2) is recovered. In most cases, 1   Importation of large amounts of LNG would require major investments to provide LNG marine terminals and related infrastructure. These would be potential targets for terrorist attacks, which would threaten the security of LNG supplies.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs FIGURE G-1 Schematic representation of the steam methane reforming process. CO2 is vented to the atmosphere today, but there are options for capturing it for subsequent sequestration. Worldwide production of hydrogen is about 41 million tons per year (ORNL, 2003). Since over 80 percent of this production is accomplished by steam methane reforming (SMR), this method is discussed first. Steam Methane Reforming Steam methane reforming involves four basic steps (see Figure G-1). Natural gas is first catalytically treated with hydrogen to remove sulfur compounds. It is then reformed by mixing it with steam and passing it over a nickel-on-alumina catalyst, making CO and hydrogen. This step is followed by catalytic water-gas shift to convert the CO to hydrogen and CO2. Finally, the hydrogen gas is purified with pressure swing adsorption (PSA). The reject stream from PSA forms a portion of the fuel that is burned in the reformer to supply the needed heat energy. Therefore, CO2 contained in the PSA reject gas is currently vented with the flue gas. If the CO2 were to be sequestered, a separations process would be added to capture it. The reforming reactions are as follows: CH4 + H2O → CO + 3H2 CO + H2O → CO2 + H2 (water-gas-shift reaction) Overall: CH4 + 2 H2O → CO2 + 4H2 The reaction of natural gas with steam to form CO and H2 requires a large amount of heat (206 kJ/mol methane). In current commercial practice, this heat is added using fired furnaces containing tubular reactors filled with catalyst. Partial Oxidation Partial oxidation (POX) of natural gas with oxygen is carried out in a high-pressure, refractory-lined reactor. The ratio of oxygen to carbon is carefully controlled to maximize the yield of CO and H2 while maintaining an acceptable level of CO2 and residual methane and minimizing the formation of soot. Downstream equipment is provided to remove the large amount of heat generated by the oxidation reaction, shift the CO to H2, remove CO2, which could be sequestered, and purify the hydrogen product. Of course, this process requires a source of oxygen, which is usually provided by including an air separation plant. Alternatively, air can be used instead of oxygen and product hydrogen recovered from nitrogen and other gases using palladium diffusion. POX can also be carried out in the presence of an oxidation catalyst, and in this case is called catalytic partial oxidation. Autothermal Reforming As already indicated, SMR is highly endothermic, and tubular reactors are used commercially to achieve the heat input required. When oxygen and steam are used in the conversion and are combined with SMR in autothemal reforming (ATR), the heat input required can be achieved by the partial combustion of methane. The reformer consists of a ceramic-lined reactor with a combustion zone and a subsequent fixed-bed catalytic SMR zone. Heat generated in the combustion zone is directly transferred to the catalytic zone by the flowing reaction gas mixture, thus providing the heat needed for the endothermic reforming reaction. As will be discussed, ATR is used today primarily for very large conversion units. There are several other design concepts that combine direct oxygen injection and catalytic conversion, including secondary reforming. It has been suggested that methane conversion to hydrogen and elemental carbon might also be an attractive route, but the committee believes that this is unlikely. Such an approach would generate a large amount of carbon by-product,2 and less than 60 percent of the combined heats of combustion of the hydrogen and carbon products is associated with the hydrogen. For this approach to become a viable alternative, uses for large amounts of carbon must be found. Natural Gas Conversion Today Steam methane reforming is widely used worldwide to generate both synthesis gas and hydrogen. The gas produced is 2   On a stoichiometric basis, 3 kg C would be made per kilogram of hydrogen.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs used to make chemicals such as ammonia and methanol, to refine petroleum, metals, and electronic materials, and to process food components. More than 32 million tons per year (t/yr) H2 (80 million kg/day) are produced using natural gas SMR. Hydrogen is also made today using partial oxidation and ATR. The vast commercial experience based on this manufacturing capacity has led to many improvements in the technology, reducing costs and increasing efficiency. Perhaps the most important element is the tubular reactor in which the SMR reaction takes place. Progress has led to higher tube wall temperatures, better control of carbon formation, and feedstock flexibility.3 This progress in turn has led to lower steam-to-carbon ratios and improved efficiency. The water-gas-shift unit has also been improved, and now one-step shift can be employed to replace the former two-step operation at different temperatures. Finally, purification of the hydrogen product has been simplified by using PSA to remove methane, carbon oxides, and trace impurities in a single step. While designs today do not generally include CO2 capture, technology is currently available to accomplish this. Using a commercial selective absorption process, CO2 could be recovered for subsequent sequestration. Progress has also been made in designing and building larger SMR plants. Currently, single-train commercial plants of up to 480,000 kg H2 per day (200 million standard cubic feet per day [scf/d]) are being built, and even larger plants can be constructed using multiple trains. Units as small as 300 kg/day are also being built.4 In many cases, the units built are one of a kind, with specific features to meet the requirements of a site, application, or customer. At least one company is fabricating commercial SMR hydrogen plants as small as 300 kg/day using components of fixed design, one of the elements of mass production.5 Partial oxidation utilizing natural gas is fully developed and used commercially. In most cases today, commercial units use feeds of lower value than natural gas, such as coal, coke, petroleum residues, or other by-products, because of economics. However, natural gas is a preferred feed for POX from a technical standpoint and can be used to generate hydrogen where competitive. Oxygen-blown ATR with natural gas is used today in very large units that generate a mixture of CO and H2 for the Fischer-Tropsch process or methanol synthesis. This is attractive in part because the units can produce the hydrogen-to-carbon monoxide ratio needed in the synthesis step. Since the heat of reaction is added by combustion with oxygen, the catalyst can be incorporated as a fixed bed that can be scaled up to achieve further benefits of larger plant size in both the ATR and the oxygen plant that is required. ATR also offers benefits when CO2 capture is included. This is because the optimum separation technology for this design recovers CO2 at 3 atmospheres (atm), thus reducing the cost of compression to pipeline pressure (75 atm). In summary, all three processes (SMR, POX, and ATR) are mature technologies today for the conversion of natural gas to hydrogen. SMR is less costly except in very large units, where ATR has an advantage. SMR is also somewhat more efficient when the energy for air separation is included. POX has the advantage of being applicable to lower-quality feeds such as petroleum coke, but this is not directly relevant to natural gas conversion. Future Natural Gas Conversion Plants Given the current interest in possibilities for a hydrogen economy and the current commercial need for hydrogen, significant effort is being focused on improving natural gas conversion to hydrogen. Improved catalysts and materials of construction, process simplification, new separations processes, and reactor concepts that could improve the integration of steam reforming and partial oxidation are being investigated. Catalytic partial oxidation is also under consideration. Since steam reforming and partial oxidation are mature technologies, the primary opportunities for improvement involve developing designs for specific applications that are cost-effective and efficient. Several thousand distributed generators will be needed for the hydrogen economy, and it should be possible to lower the cost of these generators significantly through mass production of a generation “appliance.” Such appliances may be further improved by tailoring the design to the fueling application. For example, hydrogen would likely be stored at roughly 400 atm, and to the extent that the conversion reactor pressure can be increased, hydrogen compression costs would be reduced and efficiency improved. For distributed generators incorporating POX or ATR, suitable cost-effective methods for hydrogen purification need to be developed. Alternatively, in such cases there are potentially attractive opportunities to recover the oxygen needed with membranes and thus to lower the cost. Other concepts are also in the exploratory research stage. These involve new or modified ways of providing the endothermic heat of steam reforming or utilizing the heat of reaction in partial oxidation. New, lower-cost designs for distributed generation probably can be advanced to the commercial prototype stage in the next 5 to 7 years. Some of these improvements could be applicable to large plants. Economics The committee undertook cost studies as described elsewhere (in Chapter 5 and Appendix E) to identify the areas 3   J.R. Rostrup-Nielsen, Haldor Topsoe, “Methane Conversion,” presentation to the committee, April 25, 2003. 4   Personal communication from Dale Simbeck, SFA Pacific, to committee member Robert Epperly, April 30, 2003. 5   Dennis Norton, Hydro-Chem, “Hydro-Chem,” presentation to the committee, June 11, 2003.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs TABLE G-1 Economics of Conversion of Natural Gas to Hydrogen   Plant Size (kilograms of hydrogen per stream day [SD]) and Case   1,200,000a 24,000b 480c   Current Possible Future Current Possible Future Current Possible Future Investment (no sequestration), $/kg/SD 411 297 897 713 3847 2001d Investment (with sequestration), $/kg/SDe 520 355 1219 961 — — Total H2 cost (no sequestration), $/kg 1.03f 0.92f 1.38 f 1.21 f 3.51g 2.33g Total H2 cost (with sequestration), $/kge 1.22 f 1.02 f 1.67 f 1.46 f — — CO2 emissions (no sequestration), kg/kg H2 9.22 8.75 9.83 9.12 12.1 10.3 CO2 emissions (with sequestration), kg/kg H2 1.53 1.30 1.71 1.53 — — Overall thermal efficiency (no sequestration), %h 72.3a 77.9a 46.1 53.1 55.5 65.2 Overall thermal efficiency (with sequestration), %e, h 61.1 68.2 43.4 49.0 — — aIncludes compression of product hydrogen to pipeline pressure of 75 atm. bIncludes liquefaction of H2 prior to transport. cIncludes compression of H2 to 400 atm for storage/fueling vehicles. dIncludes estimated benefits of mass production. eIncludes capture and compression of CO2 to 135 atm for pipeline transport to sequestration site. fBased on natural gas at $4.50/million Btu. gBased on natural gas at $6.50/million Btu. hBased on lower heating values for natural gas and hydrogen; includes hydrogen generation, purification, and compression, and energy imported from offsite, as well as distribution and dispensing. that could have the greatest impact on the introduction of hydrogen fuel. For hydrogen production from natural gas, plant sizes of 1,200,000 kg per stream day (kg/SD), 24,000 kg/SD, and 480 kg/SD were studied (see Table G-1).6 For each plant size, a current case representing what can be done today with modern technology and a future case representing what might be possible in the future were included. The possible future case for the 480 kg/SD plant includes the estimated benefits of mass production. For the two larger plants, options were included to capture CO2 and to compress it to pipeline pressure (75 atm) for sequestration offsite. Capture was not included for the smallest plant, since the cost for collection of CO2 from distributed plants was considered to be prohibitive, in that forecourt sequestration of CO2 added $4.40/kg H2 to the cost (DiPietro, 1997). As shown in Table G-1, current investments vary with plant size, from $411 to $3847/kg/SD as size is decreased from 1.2 million to 480 kg/SD. While improved technology visualized in the possible future cases lowers investment by 20 to 48 percent, plant size has a more pronounced effect (see Figure G-2). For the two larger plants, CO2 capture increases investment by 22 to 35 percent. As illustrated in Figure G-3, hydrogen cost7 in the largest plant with no CO2 capture is $1.03/kg of hydrogen with current technology and $0.92/kg with future technology. This cost increases to $1.38/kg and $1.21/kg in a midsize plant, and to $3.51/kg and $2.33/kg in the smallest plant. CO2 capture adds 11 to 21 percent, depending on the case. Table G-1 shows overall thermal efficiency8 for the largest plant to be 72.3 to 77.9 percent without CO2 capture (for current and possible future technology, respectively), and 61.1 to 68.2 percent with CO2 capture (for current and possible future technology, respectively). Efficiency for the smallest plant is 55.5 to 65.2 percent.9 Without capture, the CO2 emissions are 8.8 to 12.1 kg CO2 per kilogram hydrogen. Capture lowers these emissions to 1.3 to 1.7 kg CO2 per kilogram of 6   All plant capacities are in kilograms of hydrogen per stream day. 7   Hydrogen costs are based on a natural gas price of $4.50/million Btu for the two larger plants and $6.50/million Btu for the smallest one. 8   Based on lower heating values of natural gas and hydrogen; includes production. 9   The thermal efficiencies for the midsize plant are 43.4 and 46.1 percent with current technology (with and without CO2 capture, respectively) and 49.0 and 53.1 percent with possible future technology (with and without CO2 capture, respectively). These numbers are lower than might be expected, because it is assumed that hydrogen from these plants would be delivered to fueling stations as a liquid. These cases include the liquefaction of hydrogen.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs FIGURE G-2 Estimated investment costs for current and possible future hydrogen plants (with no carbon sequestration) of three sizes. FIGURE G-3 Estimated costs for conversion of natural gas to hydrogen in plants of three sizes, current and possible future cases, with and without sequestration of CO2.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs hydrogen. Emissions and thermal efficiency estimates include the effects of generating the required electricity offsite in state-of-the-art power generation facilities with 65 percent efficiency with 0.32 kg CO2/kWh of electricity. The DOE states that its goal by 2010 is to reduce the cost of the distributed production of hydrogen from natural gas and/or liquid fuels to $1.50/kg (delivered, untaxed, without sequestration) at the pump, based on a natural gas price of $4/million Btu (DOE, 2003a). The committee’s analysis indicates that this goal will be very difficult to achieve for the distributed-size hydrogen plants and will likely require additional time. The possible future case for distributed generation, which already incorporates the estimated benefits of mass production of SMR units, yields a hydrogen cost of $1.88/kg with $4/million Btu of natural gas. Achievement of the DOE goal would require additional thermal efficiency improvements and investment reductions. The goal could be met if, for example, the SMR thermal efficiency were further increased from 70 to 80 percent (excluding the compression of product hydrogen to storage pressure) and the SMR investment was cut by 35 percent, assuming that the benefits of mass production have been appropriately included. The committee did not study the likelihood of achieving these additional improvements. It is also important to note that the committee’s cost estimates are based on the assumption that distributed generators operate throughout the year at 90 percent of design capacity. As a consequence, units would have to operate at or near design capacity 24 hours per day, or else the actual cost of hydrogen from such units would be higher than calculated.10 Achieving a 90 percent capacity factor would require careful integration of the design rate of the hydrogen generator, hourly demand variations at fueling stations, and onsite storage capability. The committee believes that there is considerable uncertainty regarding the future cost of hydrogen from small hydrogen plants. This uncertainty is further increased by the need for high reliability and safe operation with infrequent attention from relatively unskilled operators (i.e., customers and station attendants). In the committee’s view, the DOE program should address these issues on a priority basis, as discussed below. Hydrogen cost using steam methane reforming is sensitive to the price of natural gas, as shown in Figure G-4. Based on current technology cases, an increase in natural gas price from $2.50 to $6.50/million Btu increases hydrogen cost by 97 percent in a 1.2 million kg/SD plant and by 68 percent in a 24,000 kg/SD unit. For the 480 kg/SD unit, an increase from $4.50 to $8.50/million Btu raises hydrogen cost by 28 percent. These numbers highlight the importance of focusing research on improving efficiency in addition to reducing investment. TABLE G-2 U.S. Natural Gas Consumption and Methane Emissions from Operations, 1990 and 2000 Consumption/Emissions 1990 2000 Natural gas consumption (Tcf)a 18.7 22.6 Methane emissions (Gg)b 5772 5541 aSee EPA (2002). bU.S. Department of Energy, Energy Information Sheets, “Natural Gas Consumption,” May 12, 2003, Washington, D.C. Other Environmental Impacts Natural gas is lost to the atmosphere during the production, processing, transmission, storage, and distribution of hydrogen. Since methane, the major component of natural gas, has a global warming potential of 23,11 this matter deserves discussion. Methane is produced primarily in biological systems through the natural decomposition of organic waste. Methane emissions include those from the cultivation of agricultural land and the decomposition of animal wastes. The Environmental Protection Agency (EPA) estimates that 70 percent of methane emissions result from human activities and the balance from natural processes.12 Less than 20 percent of total global emissions of methane are related to fossil fuels, including natural gas operations (IPCC, 1995). The EPA reports that 19 percent of the anthropogenic emissions of methane in 2000 came from natural gas operations, and 25 percent of that came from distribution of natural gas within cities, primarily to individual users (EPA, 2002). Perhaps the most compelling statistic is that between 1990 and 2000, methane emissions from natural gas operations decreased even though natural gas consumption increased (Table G-2). Clearly, improvements are being made to reduce losses from natural gas operations. For example, the EPA says that a voluntary program with industry, the Natural Gas STAR Program,13 has reduced methane emissions by 216 billion cubic feet (Bcf) since its inception in 1993. As already pointed out, the advent of hydrogen-powered cars would increase natural gas consumption significantly. 10   Based on the committee’s model, a reduction of on-stream time from 90 to 70 percent would increase the cost of hydrogen in a 480 kg/SD unit by 11 to 15 percent. 11   The Intergovernmental Panel on Climate Change (IPCC) has defined global warming potential as follows: “An index, describing the radiative characteristics of well mixed greenhouse gases, that represents the combined effect of the differing times these gases remain in the atmosphere and their relative effectiveness in absorbing outgoing infrared radiation. This index approximates the time-integrated warming effect of a unit mass of a given greenhouse gas in today’s atmosphere, relative to that of carbon dioxide” (IPCC, 2001a). 12   See Environmental Protection Agency (EPA), “Current and Future Methane Emissions from Natural Sources.” Available online at http://www.epa.gov/ghginfo/reports/curr.htm. Accessed December 10, 2003. 13   Information on the U.S. Environmental Protection Agency’s STAR Program is available online at http://www.epa.gov.gasstar/. Accessed November 15, 2003.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs FIGURE G-4 Estimated effects of the price of natural gas on the cost of hydrogen at plants of three sizes using steam methane reforming. Costs based on current technology. NOTE: SD = stream day. However, this increase would not necessarily increase losses from the natural gas system. Advantages and Disadvantages There are several advantages to generating hydrogen from natural gas. Feedstock availability is quite widespread, since an extensive pipeline distribution system for natural gas already exists in the United States and natural gas is available in most populated areas of the country. Further, there is extensive commercial experience, and natural-gas-to-hydrogen conversion technology is widely used commercially throughout the world and is at an advanced stage of optimization in large plants. If centralized, large-scale natural gas conversion plants are built, CO2 can be captured for subsequent sequestration, although its separation and capture are probably not economically feasible with small, distributed hydrogen generators. Furthermore, the committee believes that small-scale reformers at fueling stations are one of the technologies most likely to be implemented in the transition period if policies are put in place to stimulate a transition to hydrogen for light-duty vehicles. The primary disadvantages of using natural gas are that it is a nonrenewable, limited resource, and increasing amounts are projected to be imported in the future to meet U.S. market needs—which runs counter to the DOE’s goal of improving national security. Also, natural gas prices are volatile and are very sensitive to seasonal demand. Over the past 12 months, for example, the price has varied from $2.70 to more than $9.50/million Btu,14 and there has been an upward trend in the U.S. wellhead gas price since 1998. This variability becomes even more important given that SMR economics are sensitive to natural gas price. Research Needs and the Department of Energy Program Distributed generation of hydrogen from natural gas in fueling facilities could be the lowest-cost option for hydrogen production during the transition. However, the future cost of this option is uncertain, given the technical and engineering uncertainties and special requirements that demand priority attention in the DOE program, as it is advanced by contract research organizations. Distributed generation of hydrogen as envisioned has never before been achieved because of two particular requirements: (1) the mass production of the thousands of generating units, incorporating the latest technology improve- 14   See NYMEX Henry-Hub NATURAL GAS PRICE, available online at http://www.oilnergy.com/1gnymex.htm#year. Accessed December 10, 2003.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs ments, needed to meet demand, minimize cost, and improve efficiency; and (2) unit designs and operating procedures that ensure the reliable and safe operation of these appliances with only periodic surveillance by relatively unskilled personnel (station attendants and consumers). Currently, there is a market for such units in the merchant industrial sector, which accounts for about 12 percent of the total hydrogen market in the United States (ORNL, 2003). It is clear that the DOE must provide the impetus for the program. In contrast, centralized generation of hydrogen in one-of-a-kind, medium-sized and large plants is widely practiced, and as a result there is extensive commercial experience in this area. Given the commercial market for hydrogen, the committee believes that suppliers will continue to search for ways to improve the technology and make it even more competitive for medium- and large-scale plants. Publications from the DOE hydrogen program indicate that the program on distributed generation will include demonstration of a “low-cost, small-footprint plant” (DOE, 2003a, b). However, it is not clear whether the program gives priority to distributed generation or includes an effort to demonstrate the benefits of and specific designs for mass production in the specified time frame of the program. The needed designs would involve concomitant engineering that would create designs for manufacturing engineering, to guide research and to prepare for mass production of the appliance, and would also develop a system design for a typical fueling facility, including the generation appliance, compression, high-pressure storage incorporating the latest storage technology, and dispensers. With today’s technology, such ancillary systems cost 30 percent as much as the reformer. The committee believes that these costs can be reduced by over 50 percent and that efficiency can be improved through system integration and the incorporation of the latest technology. Compression and high-pressure storage are examples of systems in which significant improvements are expected. The DOE hydrogen program is positioned to stimulate the development of newer concepts, such as membrane separation coupled with chemical conversion, and this seems appropriate to the committee. However, most of the effort in this area appears directed toward POX and ATR. The committee believes that SMR could be the preferred process for this application, and that it should also be pursued in parallel with the effort involving POX and ATR. HYDROGEN FROM COAL This section presents the basics of making hydrogen from coal in large, central station plants. The viability of this option is contingent on demand for hydrogen large enough to support an associated distribution system, a large resource base, competitive uses of coal, the environmental impacts of production and transportation, and the technologies and the associated costs for converting coal into hydrogen. Many of the issues and technologies associated with making hydrogen from coal are similar to those of making power from coal. These subjects are closely linked and should be considered in concert—particularly with respect to clean coal technologies. These technologies will be required for making hydrogen, and they also offer the best opportunity for low-cost, high-efficiency, and low-emission power production. The lowest-cost hydrogen coal plants are likely to be ones that coproduce power and hydrogen.15 Coal is a viable option for making hydrogen in large, central station plants when the demand for hydrogen becomes sufficient to support an associated, large distribution system. The United States has enough coal to make hydrogen far into the future. A substantial coal infrastructure already exists, commercial technologies for converting coal to hydrogen are available from several licensors, the cost of hydrogen from coal is among the lowest available, and technology improvements are identified that should reach future DOE cost targets. The major consideration is that, because of the high carbon content in coal, the CO2 emissions from making hydrogen from coal are larger than those from any other conversion technology for making hydrogen. This underscores the need to develop carbon sequestration techniques that can handle very large amounts of CO2 before the widespread implementation of coal to make hydrogen should occur. Coal Transportation If coal is to be a major source for future hydrogen production, the infrastructure for delivering it to the future hydrogen plants will need to be expanded enough to handle these future requirements. Based on the assumptions used by the committee, the current production and delivery infrastructure capacity would need to be increased by 11 percent to meet the 2030 hydrogen demand, and by 57 percent to meet the 2050 hydrogen demand. Coal is a viable option for making hydrogen in large, central station plants when the demand for hydrogen becomes large enough to support an associated transport, storage, and distribution system. Most bulk coal transportation is by rail, with trucks used for local transport. For reasons of economics, most of the world’s coal consumption is in power plants located nearby coal mines, which minimizes the necessity for long-distance transportation. More than 60 percent of the coal used for power generation worldwide is consumed within 50 km of the mine site. In the United States, the average distance that coal is shipped by rail is farther, at about 800 miles. That distance has increased in recent years owing to the move toward greater use of coals with lower sulfur content (found mainly in the West) to meet sulfur oxide emissions standards in plants located mainly in the South and the East. As coal is currently shipped over great distances in the United States, 15   David Gray and Glen Tomlinson, Mitretek Systems, “Hydrogen from Coal,” presentation to the committee, April 24, 2003.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs delivery to broad geographic areas should not be a barrier to the use of coal to make hydrogen for at least the next 30 years, since demand will not be much different from current trends. Environmental Impacts of Coal Consumption and Transportation Using more coal to produce hydrogen will have a number of environmental consequences. Coal mining itself causes numerous environmental issues, ranging from widespread land disturbance, soil erosion, dust, biodiversity impacts, waste piles, and so forth, to subsidence and abandoned mine workings. Once coal has been extracted, it needs to be moved from the mine to the power plant or other place of use. The main pollutants resulting from conventional combustion of coal are sulfur oxides (SOx), nitrogen oxides (NOx), particulates, CO2, and mercury (Hg). SOx is dealt with through lower-sulfur-content coal as well as flue gas desulfurization (FGD). Approximately 30 percent of U.S. coal power generating equipment had some sort of FGD or SOx reduction technology at the end of 1999, according to data gathered by DOE’s Energy Information Administration.16 Newer processes for power generation, such as integrated gasification combined cycle power generation, which involves a conversion rather than a combustion process, is more effective at reducing criteria pollutants than existing pollution control technologies are (East-West Center, 2000). Potentially the most significant future issue for coal combustion is CO2 emissions, since on a net energy basis coal combustion produces 80 percent more CO2 than the combustion of natural gas does, and 20 percent more than does residual fuel oil, which is the most widely used other fuel for power generation (EIA [2001], Table B1). Likewise, the CO2 emissions associated with making hydrogen from coal will be larger than those for making hydrogen from natural gas. Using currently available technology, the CO2 emissions are about 19 kg CO2 per kilogram of hydrogen produced, compared with approximately 10 kg CO2 per kilogram of hydrogen manufactured from natural gas. Atmospheric emissions from coal-fired generating plants are of concern to various bodies—national (criteria pollutants [CO, particulates,17 O3, NO2, SO2, and Pb], are defined and regulated by the EPA under the National Ambient Air Quality Standards) and international (greenhouse gases, considered under the UN Framework Convention on Climate Change, are mainly CO2, CH4, N2O, hydrofluorocarbons, perfluorocarbons, and SF6). Since the 1970s, the U.S. electricity industry has made considerable progress in reducing SO2, NO2, and particulate emissions, despite a large increase in coal consumption, through the use of FGD, filtration, electrostatic precipitators, and selective catalytic reduction (SCR). To the extent that new emission control technologies can be applied to existing plants and that new generating technologies can be used, further progress is expected in overall emissions reductions (Ness et al., 1999). Current Coal Technologies Conventional coal-fired power generation uses a combustion boiler that heats water to make steam, which is used to drive an expansion steam turbine and generator. Various designs of coal combustion boilers exist, the most modern and efficient of which use pulverized coal and produce supercritical (high-pressure/high-temperature) steam. Overall efficiencies are typically in the 36 to 40 percent range. Although a staple for power generation for decades, this conventional combustion technique is not suitable for making hydrogen. Hydrogen-making technologies employ a conversion process rather than a combustion process. These conversion processes, such as gasification, are suitable for making power and/or hydrogen. Clean Coal Technologies Clean coal technologies use alternative ways of converting coal so as to reduce plant emissions and increase plant thermal efficiency, leading to an overall cost of electricity that is lower than the cost for electricity from conventional plants. Systems under development include low-emission boiler systems (LEBSs), high-performance power systems (HIPPSs), integrated gasification combined cycle (IGCC), and pressurized fluidized-bed combustion (PFBC) (Ness et al., 1999). The goal is to attain thermal efficiencies in the 55 to 60 percent range (higher heating value [HHV]) (Ness et al., 1999). With the exception of the IGCC systems, all of the others rely on increasingly sophisticated emissions control systems; IGCC uses a different conversion system to reduce emissions at the outset. It is this gasification technology that is best suited to making hydrogen from coal. Gasification Technology Gasification systems typically involve partial oxidation of the coal with oxygen and steam in a high-temperature and elevated-pressure reactor. The short-duration reaction proceeds in a highly reducing atmosphere that creates a synthesis gas, a mix of predominantly CO and H2 with some steam and CO2. This syngas can be further shifted to increase H2 yield. The gas can be cleaned in conventional ways to recover elemental sulfur (or make sulfuric acid), and a high-concentration CO2 stream can be easily isolated and sent for 16   Energy Information Administration, Form EIA-767, “Steam-Electric Plant Operation and Design Report”; Form EIA-860A, “Annual Electric Generator Report-Utility”; and Form EIA-860B, “Annual Electric Generator Report—Non-utility.” 17   Two sizes are considered criteria pollutants, PM10 and PM2.5. The 2.5 mm particles result from combustion; the larger, 10 mm particulates typically take the form of airborne dust. Both can penetrate the lungs and are known to cause long-term damage resulting in respiratory and bronchial diseases.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs disposal. The use of high temperature and pressure and oxygen minimizes NOx production. The slag and ash that is drawn off from the bottom of the reactor encapsulate heavy metals in an inert, vitreous material, which currently is used for road fill. The high temperature also eliminates any production of organic materials, and more than 90 percent of the mercury is removed in syngas processing. Syngas produced from current gasification plants is used in a variety of applications, often with multiple applications from a single facility. These applications include syngas used as feedstock for chemicals and fertilizers, syngas converted to hydrogen used for hydro-processing in refineries, production, generation of electricity by burning the syngas in a gas turbine, and additional heat recovery steam generation using a combined cycle configuration. There are currently at least 111 operating gasification plants running on a variety of feedstocks. These include residual oils from refining crude oil, petroleum coke, and to a lesser extent, coal. The syngas that is generated has typically been used for subsequent chemicals manufacture; making power from IGCC systems is a more recent innovation, successfully demonstrated in the mid-1980s and commercially operated since the mid-1990s. Gasification is, therefore, a well-proven commercial process technology, and several companies offer licenses for its use. Oxygen-Blown Versus Air-Blown Gasification Gasification plants exist that use either air-blown or oxygen-blown designs. Air-blown designs save the capital cost and operating expense of air separation units, but the dilution of the combustion products with nitrogen makes the separation of CO2, in particular, a much more expensive exercise. In addition, the extra inert nitrogen volume going through the plant increases vessel sizes significantly and increases the cost of downstream equipment. Oxygen-blown designs do not introduce the additional nitrogen, so once the sulfur compounds have been removed from the syngas, what is left is a high-purity stream of CO2 that can be more easily and cheaply separated. Because of the need to consider CO2 capture and sequestration for future hydrogen generation plants, only oxygen-blown designs are feasible for consideration. Estimated Costs of Hydrogen Production and Carbon Dioxide Emissions Most gasification plants produce syngas for chemical production, and often for steam. IGCC plants then burn the syngas to produce power. The flexibility to polygenerate multiple products to suit a given situation is one of the strengths of the gasification system. Thus, relatively few gasification plants are dedicated to producing hydrogen only (or indeed any other single product). The future large-scale hydrogen generation plant will likely also generate some amounts of power because of the advantages provided through polygeneration. It is necessary therefore to preface any remarks concerning the costs of producing only hydrogen or the costs of sequestering CO2 with this caveat. All of the technology needed to produce hydrogen from coal is commercially proven and in operation today, and designs already exist for hydrogen and power coproduction facilities. However, technology advances currently in development will continue to drive down the costs and increase the efficiency of these facilities. Hydrogen-from-coal plants combine a number of technologies including oxygen supply, gasification, CO shift, sulfur removal, and gas turbine technologies. All of these technology areas have advances under development that will significantly improve the plant’s capital and operating costs and thermal efficiency. Examples of these pending technology advances include Ion Transport Membrane (ITM) technology for air separation (oxygen supply); advances in gasifier technology (feedstock preparation, conversion, availability); warm gas cleanup; advanced gas turbines for both syngas and hydrogen; CO2 capture technology advances; new, lower-cost sulfur-removal technology; and slag-handling improvements. It is estimated that today a gasification plant producing hydrogen only would be able to deliver hydrogen to the plant gate at a cost of about $0.96/kg H2 with no CO2 sequestration. If CO2 capture were also required, it would cost $1.03/ kg H2. This pricing reflects costs for producing hydrogen from very large, central station plants at which hydrogen will be distributed through pipelines. In these plants a single gasifier can produce more than 100 million scf/day H2. It is envisioned that a typical installation would include two to three gasifiers. The economics of making hydrogen from coal is somewhat different from that for making it from other fossil fuels, in that the capital costs needed per kilogram of produced hydrogen are larger for coal plants, but the raw material costs per kilogram of produced hydrogen are lower. Coal is inexpensive, but the coal gasification plant is expensive. If the coal price is changed by 25 percent, the hydrogen cost is changed by only $0.05/kg. If the cost of the plant is changed by 25 percent, the hydrogen cost is changed by $0.16/kg. This should lead to a very stable cost of hydrogen production that can be lowered through future improvements in technology. In addition to the CO2 produced from making the electricity consumed in producing hydrogen, CO2 emissions result from the carbon in the coal. The emissions depend on the type and quality of coal, but for typical Western coal with 2 percent sulfur and 12,000 Btu/dry lb, approximately 18.8 kg CO2 are emitted per kilogram of hydrogen produced. With a CO2 capture system in place, it is estimated that this figure could be reduced by as much as 80 to 90 percent, the exact amount depending on capital efficiency and cost-benefit analysis. Although the economics of hydrogen production from coal does vary somewhat with the quality of coal being gasified, essentially any coal can be gasified to produce hy-

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs drogen. Coals with ash content greater than 30 percent are already being gasified. The main effects of coal-quality variance on hydrogen production are the amount of by-products produced (primarily slag and elemental sulfur) and the capital cost, which would be affected mostly by the amount of additional inert material in the coal that has to be handled. For a gasification plant producing maximum hydrogen from coal, the variance in potential feed coal quality is estimated to produce a variance of less than 15 percent in the amount of CO2 generated per ton of hydrogen produced. The lower-quality coals generate lower amounts of CO2 per ton of hydrogen. Other effects of coal quality are less significant. Research and Development Needs In terms of its stage of development, coal gasification is a less mature commercial process than coal combustion processes and other hydrogen generation processes using other fossil fuels, especially in the aspects of capturing CO2 and providing flexibility in hydrogen and electricity production. In that sense the potential for improvement through technology development is significant. The main issues are capital cost and reliability (the latter is usually addressed through including standby equipment). Both are major reasons why IGCC technology has not been widely adopted for power generation, which is a very competitive business. The flexibility to vary between hydrogen production and power production will cost extra capital, which has to be recovered. For the commercial processes available from several different licensors, the R&D needs should be directed at capital cost reduction, standardization of plant design and execution concept, gas cooler designs, process integration, oxygen plant optimization, and acid gas removal technology. The potential efficiency and capital cost improvements in these areas could combine to lower the overall cost of hydrogen from coal by about 10 to 15 percent from today’s costs. Since many parts of the coal-to-hydrogen process are the same as for coal-to-power processes, similar improvements in power costs from IGCC should be possible. These areas are improvements to existing technology, so they should be able to be achieved in the near term. The potential also exists for new technologies to make larger improvements in the efficiency and cost of making hydrogen from coal. For new gasification technologies, the best opportunities for R&D appear to be for new reactor designs (entrained bed gasification), improved gas separation (hot gas separation), and purification techniques. These technologies, and the concept of integrating them with one another, are in very early development phases and will require longer-term development to verify the true potential and to reach commercial readiness. Recent studies have indicated that the combined potential of these new technologies could lower the cost of making hydrogen from coal by about 25 percent. Future Costs Evolutionary improvements in current technology can lower the cost of hydrogen from coal from the estimated $0.96/kg to about $0.90/kg. The evolution of future costs will be a function of the number of units constructed over time, since each subsequent plant gives an additional opportunity to apply the experience derived from prior plants, as well as economies of scale for process unit production. The introduction of new technologies can lower costs even further. New gasification technologies along with new syngas cleanup and separation technologies hold potential for further improving efficiencies and lowering the costs of producing hydrogen to about $0.71/kg (see Chapter 5 and Appendix E). Separating and capturing CO2 will increase these costs to $0.77/kg. Department of Energy Programs for Coal to Hydrogen The DOE programs for making hydrogen from coal reside in the Office of Fossil Energy and are related to programs to make electricity from coal. The overall goal of the Hydrogen from Coal Program is to have an operational, zero-emissions, coal-fueled facility in 2015 that coproduces hydrogen and electricity with 60 percent overall efficiency (DOE, 2003c). Major milestones for reaching this goal include these: 2006—Advanced hydrogen separation technology, including membranes tolerant of trace contaminants, identified; 2011—Hydrogen modules for coal gasification combined-cycle coproduction facility demonstrated; and 2015— Zero-emission, coal-based plant producing hydrogen and electric power (with sequestration) that reduces the cost of hydrogen by 25 percent compared with the cost at current coal-based plants demonstrated. To reach these milestones, R&D activities within the Hydrogen from Coal Program are focused on the development of novel processes that include these: Advanced water-gas-shift reactors using sulfur-tolerant catalysts, Novel membranes for hydrogen separation from CO2, Technology concepts that combine hydrogen separation and water-gas shift, and Fewer-step designs to separate impurities from hydrogen. Associated coal gasification R&D programs in which success is dependent on efficiency improvements and lower cost include these: Advanced ITM technology for oxygen separation from air, Advanced cleaning of raw synthesis gas,

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs ling the current push toward a hydrogen economy, namely, reducing CO2 emissions and reducing the need for hydrocarbon imports. In addition, it is the most affordable renewable technology deployed today, with expectations that costs will continue to decline. Since renewable technologies effectively address two of the major public benefits of a move to a hydrogen energy system, and wind energy is the closest to practical utilization with the technical potential to produce a sizable percentage of future hydrogen, it deserves continued, focused attention in the DOE’s hydrogen program. Although wind technology is the most commercially developed of the renewable technologies, it still faces many barriers to deployment as a hydrogen production system. There is a need to develop optimized wind-to-hydrogen systems. Partnerships with industry are essential in identifying the R&D needed to help advance these systems to the next level. Department of Energy’s Multi-Year Research, Development, and Demonstration Plan There is little mention of hydrogen production from wind throughout the entire June 2003 draft of “Hydrogen, Fuel Cells and Infrastructure Technologies Program: Multi-Year Research, Development and Demonstration Plan” (DOE, 2003b) or in the July 2003 Hydrogen Posture Plan: An Integrated Research, Development, and Demonstration Plan (DOE, 2003a). An RD&D plan for hydrogen production from wind power needs to be developed and integrated into the overall hydrogen strategic RD&D plan. Summary Energy security and environmental quality, including reduction of CO2 emissions, are strong factors motivating a hydrogen economy. These goals can both be fulfilled by wind-hydrogen systems. Thus, wind has the potential to play a significant role in a future hydrogen economy, both during the transition and in the long term. Since wind is currently the renewable technology that is most developed and lowest cost, wind-electrolysis-hydrogen systems merit serious attention. Wind-electrolysis-hydrogen systems have yet to be fully optimized. There are integration opportunities and issues with respect to wind machines and electrolyzers and hydrogen storage that need to be explored. For example, coproduction of electricity and hydrogen can potentially reduce costs and increase the function of the wind-hydrogen system. This could facilitate the development of wind energy systems that are more cost-effective and have broader utility, thereby assisting their development and deployment. HYDROGEN PRODUCTION FROM BIOMASS AND BY PHOTOBIOLOGICAL PROCESSES Two basic avenues for molecular hydrogen production by biological processes are currently being considered: (1) via photosynthetically produced biomass followed by subsequent thermochemical processing, and (2) via direct photobiological processes without biomass as intermediate. The first process is well known and intensely researched, while the second is still in the early research stage. These processes have in common the capturing and conversion of solar energy into chemical energy mediated by photosynthetic processes. In both cases, solar energy serves ultimately as the primary energy source for the production of molecular hydrogen by biological processes. In contrast to processes using fossil fuels as primary energy sources, biological processes do not involve net production of CO2. Efficiency of Photosynthetic Biomass Production In photosynthesis as carried out by plants, cyanobacteria, and microalgae, solar energy is converted into biomass in commonly occurring ecosystems at an overall thermodynamic efficiency of about 0.4 percent (see Figure G-13; Hall and Rao, 1999). This low efficiency is due to the molecular properties of the photosynthetic and biochemical machinery, as well as to the ecological and physical-chemical properties of the environment. Of the incident light energy, only about 50 percent is photosynthetically useful. This light energy is used at an efficiency of about 70 percent by the photosynthetic reaction center and is converted into chemical energy, which is converted further into glucose as the primary CO2 fixation end product at an efficiency of about 30 percent. Of this energy, about 40 percent is lost due to dark respiration. Because of the photo inhibition effect and the nonoptimal conditions in nature, a further significant loss in efficiency is observed when growing plants in natural ecosystems. Therefore, the energy content of common biomass collected from natural ecosystems contains only on the order of 0.4 percent of the primary incoming energy (see Figure G-13). Although higher yields (in the 1 to 5 percent range) have been reported for some crops (e.g., sugarcane), the theoretical maximal efficiency is about 11 percent. Generally, two types of biomass resources can be considered in the discussion on renewable energy feedstock: (1) primary biomass, such as energy crops, including switchgrass, poplar, and willow, and (2) biomass residues (primary when derived from wood or processed agricultural biomass; secondary when derived from food or fiber processing by-products, or animal waste; and tertiary when derived from urban residues).21 Biomass Availability Today about 4 percent of total energy use in the United States is based on the use of biomass, mainly in the form of 21   M.K. Mann and R.P. Overend, National Renewable Energy Laboratory, “Hydrogen from Biomass: Prospective Resources, Technologies, and Economics,” presentation to the committee, January 22, 2003.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs FIGURE G-13 Efficiency of biological conversion of solar energy (adapted from Hall and Rao, 1999). forest residues. At a cost of $30 to $40/t, available biomass can be estimated to be between 220 million and 335 million dry tons per year.22 This biomass consists mainly of urban residues, sludge, energy crop, and wood and agricultural residues. A significant fraction of this biomass, especially forest residues, is already used by industry or in other competing processes, such as energy generation directly. However, if all of this theoretically available biomass could be converted to hydrogen, the annually available amount would be on the order of 17 million to 26 million t H2. As Figure 6-3 indicates, in an all-fuel-cell-vehicle scenario in the year 2050, 112 million t H2 would be required annually. Considering this demand and the competing demands for other uses of biomass, the currently available biomass is insufficient to satisfy the entire demand in a hydrogen economy, and new sources for biomass production would need to be considered. Primary biomass in the form of energy crops is expected to have the quantitatively most significant impact on hydrogen production for use as transportation fuel by 2050.23 Estimates of energy that can potentially be derived from energy crops to produce biomass by 2050 range between 45 and 250 exajoules (EJ) per year. Bioenergy crops are currently not produced as dedicated bioenergy feedstock in the United States. Therefore, crop yields, management practices, and associated costs are based on agricultural models rather than on empirical data (Milne et al., 2002; de la Torre Ugarte et al., 2003; Walsh et al., 2000). Land Use for Additional Biomass Production In the most aggressive scenario for a hydrogen economy as considered in Chapter 6, a land area between 280,000 and 650,000 square miles is required to grow energy crops in order to support 100 percent of a hydrogen economy. The magnitude for this demand on land becomes apparent when comparing these numbers with the currently used cropland area of 545,000 square miles in the United States. Consequently, bioenergy crop production would require a significant redistribution of the land currently dedicated to food crop production and/or the development of a new land source from the U.S. Department of Agriculture’s (USDA’s) Conservation Reserve Program (CRP). Although bioenergy crops can be grown in all regions of the United States, regional variability in productivity, rainfall conditions, and management practices limit energy crop farming to states in the Midwest, South, Southeast, and East (see Figure G-14) (Milne et al., 2002; de la Torre Ugarte et al., 2003; Walsh et al., 2000). Considering all cropland used for agriculture, as well as cropland in the CRP, in pasture and idle cropland, de la Torre Ugarte et al. (2003) considered two management scenarios for profitable bioenergy crop production: one to achieve high biomass production (production management scenario, or PMS), and another to achieve high levels of wildlife diversity (wildlife management scenario, or WMS). The production management scenario would annually produce about 188 million tons of dry biomass, which would be equivalent to 15 million tons of H2, requiring 41.8 million acres of cropland, of which about 22   Mark Pastor, Department of Energy, “DOE’s Hydrogen Feedstock Strategy,” presentation to the committee, June 2003; Roxanne Danz, Department of Energy, Office of Energy Efficiency and Renewable Energy, “Hydrogen from Biomass,” presentation to the committee, December 2, 2002. 23   M.K. Mann and R.P. Overend, National Renewable Energy Laboratory, “Hydrogen from Biomass: Prospective Resources, Technologies, and Economics,” presentation to the committee, January 22, 2003.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs FIGURE G-14 Geographic distribution of projected bioenergy crop plantings on all acres in 2008 in the production management scenario (after Walsh et al., 2000). 56 percent would be from currently used cropland, 30 per cent from the CRP, and 13 percent from idle cropland and pasture. The crop would be exclusively switchgrass. In the wildlife management scenario, 96 million dry tons (dt) of biomass (equivalent to 7.6 million t H2) would be produced on 19.4 million acres of cropland, of which about 53 percent would be from currently used cropland, 42 percent from the CRP, and 4 percent from idle cropland and pasture. Land from the CRP would become a significant source for farming biomass crops. The CRP sets aside environmentally sensitive acres under 10- to 15-year contracts. Appropriate management practices must be developed before CRP lands are used. Environmental ramifications of various management practices must be examined to ensure that there is no substantial loss of environmental benefits, including biodiversity and soil and water quality. It is conceivable that a farming scenario alternating between agricultural crops and bioenergy crops on existing agricultural and CRP lands could be developed; however, those unproven cases were not considered in this analysis. Biomass Cost Bioenergy crop production is considered profitable at $40/dt, and could compete with currently grown agricultural crops (TIAX LLC, 2003; Milne et al., 2002). Based on assumed yields, management practices, and input costs, switchgrass is the least-expensive bioenergy crop to produce on a per dry ton basis. Production costs (farm gate costs) for switchgrass are estimated to range from $30/dt to $40/dt, depending on the management scenarios (WMS versus PMS) (de la Torre Ugarte et al., 2003). Adding processing and delivery costs would result in an approximate delivered biomass price on the order of $40 to $50/dt, respectively. Using these feedstock costs as well as current and projected gasifier efficiencies (50 percent versus 70 percent) in the committee’s analysis, the future costs per kilogram of hydrogen produced from biomass and delivered at the vehicle is about $3.60 (scenario MS Bio-F; see Figure 5-4 in Chapter 5). In this scenario, a reduction in biomass cost was assumed to be achieved by increasing the crop yield per hectare by 50 percent, which presents significant technical challenges. The profitability of bioenergy crop farming will vary with given field and soil types (Milne et al., 2002). Notably, the price per dry ton of bioenergy crop is predicted to increase with the total biomass produced. A shift of cropland use from traditional agricultural crops to bioenergy crops will also result in higher prices for traditional crops. Because of land ownership, management, and crop establishment, biomass production by energy crop production will be more expensive than using residue biomass. Also, regional variation in the availability of residue biomass, such as in woody areas in the northeastern United States, could make hydrogen pro-

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs duction from biomass competitive in such regions in the short term. However, such operations would be restricted to selected regions in the United States, and, in a long-term sustainable scenario, would require biomass production at the same rate as its consumption. The committee considered it to be unlikely that such localized operations would contribute significantly to the nation’s H2 supply. Therefore, such cases were not considered further in this analysis, nor were fertilizer costs and the energy required to produce, harvest, and transport biomass. Environmental Impact of Biomass Used for Hydrogen Production In the overall process of biomass production and gasification, no net CO2 is generated, except for the CO2 released from fossil fuels used for (1) harvesting and transportation of biomass, (2) operation of the gasification systems, and (3) electricity, as well as for (4) production and delivery of fertilizers in an advanced biomass system. Biomass handling alone is estimated to consume about 25 percent of the total capital costs of operation of a midsize biomass gasification plant. Furthermore, biomass production requires, in addition to land (see above), about 1000 to 3000 t of water per ton of biomass, as well as nutrients in the form of nitrogen (ammonia), phosphorus (phosphate), sulfur, and trace metals. Profitable future hydrogen production from biomass will require energy crops with increased growth yields, which translates into increased need for fertilizers, energy for production of fertilizers, and potentially water. As is the case with the production of food crops, erosion, nutrient depletion of the soil, and altered water use practices could result in potentially significant environmental impacts as a consequence of farming activities. These effects need to be carefully considered. Technologies for Hydrogen Production from Biomass Current Technologies Current technologies for converting biomass into molecular hydrogen include gasification/pyrolysis of biomass coupled to subsequent steam reformation24 (Milne et al., 2002; Spath et al., 2000). The main conversion processes are (1) indirectly heated gasification, (2) oxygen-blown gasification, and (3) pyrolysis, as well as (4) biological gasification (anaerobic fermentation). Biomass gasification has been demonstrated at a scale of 100 tons of biomass per day.25 Only a small, 10 kg/day of H2 pilot biomass plant is in operation, and no empirical data on the operation, performance, and economics of a full-scale biomass-to-hydrogen plant are available.26 The thermodynamic efficiencies of these processes are currently around 50 percent. Considering the low energy content of biomass, between 0.2 percent and 0.4 percent of the total available solar energy is converted to molecular hydrogen. Biomass gasifiers are designed to operate at low pressure and are limited to midsize-scale operations, owing to the heterogeneity of biomass, its localized production, and the relatively high costs of gathering and transporting biomass. Therefore, current biomass gasification plants are associated inherently with unit capital costs that are at least five times as high as those for coal gasification (see Figure 5-2 in Chapter 5) and operate at lower efficiency. Coproduction (biorefinery) of, for example, phenolic adhesives, polymers, waxes, and other products with hydrogen production from biomass, is being discussed in the context of plant designs to improve the overall economics of biomass-to-hydrogen conversion27 (Milne et al., 2002). The technical and economic viability of such coproduction plants is unproven and was not considered in this analysis. Several major technical challenges of biomass gasification/pyrolysis exist and include variable efficiencies, tar production, and catalyst attrition28 (Milne et al., 2002). Moisture content as well as the relative composition and heterogeneity of biomass can result in significant deactivation of the catalyst. Recent fundamental research has identified a new, potentially inexpensive class of catalysts for aqueous-phase reforming of biomass-derived polyalcohols (Huber et al., 2003). In contrast to residue biomass, the use of bioenergy crops as biomass for gasification is advantageous, as its composition and moisture content are predictable, and the gasification process can be optimized for the corresponding crop. Using anaerobic fermentation to convert biomass into hydrogen, a maximum of about 67 percent of the energy content (e.g., of glucose) can be recovered in hydrogen theoretically (calculated after Thauer et al., 1977). Considering the currently known fermentation pathways, a practical efficiency of biomass conversion to hydrogen by fermentation is between 15 and 33 percent (4 mol H2/mol glucose), although this is only possible at low hydrogen partial pressure. However, more efficient fermentation pathways could be conceived and would require significant bioengineering ef- 24   M.K. Mann and R.P. Overend, National Renewable Energy Laboratory, “Hydrogen from Biomass: Prospective Resources, Technologies, and Economics,” presentation to the committee, January 22, 2003. 25   Roxanne Danz, Department of Energy, Office of Energy Efficiency and Renewable Energy, “Hydrogen from Biomass,” presentation to the committee, December 2, 2002. 26   M.K. Mann and R.P. Overend, National Renewable Energy Laboratory, “Hydrogen from Biomass: Prospective Resources, Technologies, and Economics,” presentation to the committee, January 22, 2003. 27   M.K. Mann and R.P. Overend, National Renewable Energy Laboratory, “Hydrogen from Biomass: Prospective Resources, Technologies, and Economics,” presentation to the committee, January 22, 2003. 28   M.K. Mann and R.P. Overend, National Renewable Energy Laboratory, “Hydrogen from Biomass: Prospective Resources, Technologies, and Economics,” presentation to the committee, January 22, 2003.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs forts. These values compare with a biomass gasification efficiency of around 50 percent. The impurity of the hydrogen from biomass may be of concern, as fuel cell operations require relatively high-grade quality. Economic Analysis of Current and Future Biomass-to-Hydrogen Conversion In the past, and through funding support by the DOE, the process of biomass gasification has received most attention. Gasification technology using biomass, typically wood residues as feedstock, was adapted from coal gasification, and a few small-scale prototypes of biomass gasification plants have been built. Thus, the committee considered only the economics of biomass gasification. However, no midsize gasification facility exists to date that converts biomass to hydrogen, and no empirical data are available on the operation, performance, and economics of a midsize biomass-to-hydrogen plant, as assumed in the economic model. The assumptions made for the committee’s analysis of current technology consist of modular combinations adopted from existing technical units for coal gasification (shell gasifier, air separation unit, traditional shift), without considering the variability in chemical composition and moisture content of typical biomass. An overall gasification efficiency of 50 percent is assumed. Furthermore, the committee assumed a scenario in which 100 percent of the H2 demand would need to be met by biomass-derived hydrogen, acknowledging that in a possible future scenario, a mix of different primary energy sources is more likely. As the relative proportion of such mixes of primary feedstock is unknown, the committee considered the simplified case. Estimation of the economics of future technology for biomass-to-hydrogen conversion using gasification is more problematic and much more uncertain because of the necessary extrapolations. The committee made the following assumptions for a midsize plant: (1) advanced biomass gasifiers can be developed and will use newly developed technology, such as fluidized catalytic cracking; (2) biomass gasifiers can be modified to produce a CO and H2 syngas, as does coal gasification; (3) biomass gasification will operate at an overall efficiency of about 70 percent; and (4) through genetic engineering and other breeding methods, the growth yield of switchgrass can be increased by 50 percent. The committee also assumed that the future biomass is derived from bioenergy crops at a price of $50/dt, as opposed to coming from less expensive biomass residues, although it is possible that a mixture of bioenergy crops and residues could be used for future gasifications. With these assumptions, the current price per kilogram of hydrogen delivered at the vehicle of $7.04 (see Figure 5-2) could be reduced in the future to about $3.60. As can be see in Figure 5-4, two factors contribute to the high price: the high capital charges for gasification and the high biomass costs. Photobiological Hydrogen Production In recent years, fundamental research on hydrogen production by photosynthetic organisms has received significant attention. In photosynthesis, water is oxidized photo-biologically to molecular oxygen and hydrogen in order to satisfy the organism’s need to build biomass from CO2. This notion has prompted the idea of reengineering this process to release those equivalents as molecular hydrogen directly. Such direct production of molecular hydrogen is probably thermodynamically the most efficient use of solar energy in biological hydrogen production (theoretically about 10 percent to 30 percent), because it circumvents inefficiencies in the biochemical steps involved in biomass production, as well as those involved in biomass conversion to hydrogen (see Figure G-13). The photosynthetic formation of molecular hydrogen from water is thermodynamically feasible even at high hydrogen partial pressure. However, such biological capability does not occur in any known organism; thus, it will require substantial metabolic engineering using new approaches in molecular biotechnology. In a variation of this approach, electron flow from the photosynthetic reaction center could be coupled to nitrogenase, which also releases H2. Another mode of hydrogen production, discussed in context of photosynthetic H2 production, is dark fermentation mediated by photosynthetic microorganisms. In all cases, the reducing equivalents for producing hydrogen are derived from water, which is abundant and inexpensive. It is unclear to what extent the DOE is providing substantial funds for such research. Technologies Competing for Land Surface Area Since the primary energy of all biological processes for hydrogen production is renewable solar energy, all other technologies using solar energy, including photovoltaic and other newer processes, such as thin-film technology, are competing for (land) surface area. Wind energy is indirectly solar energy. Currently, the solar-to-electrical conversion efficiency of newer photoelectric processes is 15 to 18 percent, compared with 0.4 percent for bulk biomass formation, and about 10 percent, potentially, for direct hydrogen production by photosynthetic organisms. Because solar energy harvesting technologies are competing for land use among each other and with other societal activities, such as farming, housing, and recreation, the overall efficiency of a solar energy conversion process will be a key determinant for its economic viability. Advantages and Disadvantages of Using Biomass and Photobiological Systems for Molecular Hydrogen Production Hydrogen production from biomass is an attractive technology, as the primary energy is solar (i.e., “renewable”), with

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs no net CO2 being released (except for transport). Notably, when coupled to CO2 capture and sequestration on a larger technical scale, this technology might be the most important means to achieve a net reduction of atmospheric CO2 (see Chapter 6, Figures 6-9 and 6-10). Furthermore, different forms of biomass (bioenergy crops, residues including municipal waste, etc.) could be used in different combinations. The current concept of biomass-to-hydrogen conversion has several limitations. Biomass conversion to hydrogen is intrinsically inefficient, and only a small percentage of solar energy is converted into hydrogen. Moreover, in order to contribute significantly to a hydrogen economy, the quantity of biomass that needs to be available necessitates the farming of bioenergy crops. However, bioenergy crops obtained by farming will be intrinsically expensive. Residue biomass is less expensive but more variable and heterogeneous in composition, thus making the gasification process less efficient. In addition, significant costs are associated with the collection and transportation of dispersed, low-energy-density bioenergy crops and residues. Most importantly, large-scale biomass production also would pose significant demand on land, nutrient supply, water, and the associated energy for increased biomass production. The environmental impact of significant energy crop farming is unclear, but it can be assumed to be similar to that in crop farming and include soil erosion, significant water and fertilizer demand, eutrophication of downstream waters, and impact on biological diversity. Biomass production is also sensitive to seasonal variability as well as to vagaries of weather and to diseases, with significant demands regarding the storage of biomass in order to compensate for the anticipated fluctuations. The public acceptance of growing and using potentially genetically engineered, high-yield energy crops is also unclear. In addition, competing uses of biomass for purposes other than hydrogen production will also control the price of biomass. Overall, it appears that hydrogen production from farmed and agriculture-type biomass by gasification/pyrolysis will only be marginally economical and competitive. Biomass gasification could play a significant role in meeting the DOE’s goal of greenhouse gas mitigation. It is likely that both in the transition phase to a hydrogen economy and in the steady state, a significant fraction of hydrogen might be derived from domestically abundant coal. In co-firing applications with coal, biomass can provide up to 15 percent of the total energy input of the fuel mixture. The DOE could address greenhouse gas mitigation by co-firing biomass with coal to offset the losses of carbon dioxide to the atmosphere that are inherent in coal combustion processes (even with the best-engineered capture and storage of carbon). Since growth of biomass fixes atmospheric carbon, its combustion leads to no net addition of atmospheric CO2 even if vented. Thus, co-firing of biomass with coal in an efficient coal gasification process, affording the opportunity for capture and storage of CO2, could lead to a net reduction of atmospheric CO2. The co-firing fuel mixture, being dilute in biomass, places lower demands on biomass feedstock. Thus cheaper, though less plentiful, biomass residue could supplant bioenergy crops as feedstock. Using residue biomass would also have a much less significant impact on the environment than would farming of bioenergy crops. Photobiological hydrogen production is a significantly more efficient process and requires nutrients to a lesser extent than does biomass-to-hydrogen conversion. The objective is to engineer a (micro)organism that catalyzes the light-mediated cleavage of water with the concomitant production of hydrogen at high rates and high thermodynamic efficiency. This process does not take place in naturally occurring organisms at an appreciable rate or scale. While this approach has much potential, there are also major challenges. Substantial bioengineering efforts have to be undertaken to engineer microorganisms with a robust metabolic pathway, including improved kinetics for hydrogen production and efficiencies in light energy conversion and hydrogen production, before a pilot-scale photobiological system could be evaluated. This requires long-term, fundamental research at a significant funding level. Also, inexpensive, large-scale reactor systems need to be designed that minimize the susceptibility of the reactor system to biological contamination. In addition, the public perception of the use and possible concerns over the potential “escape” of genetically engineered microorganisms need to be addressed. The Department of Energy’s Research and Development Program According to the June 2003 draft of “Hydrogen, Fuel Cells and Infrastructure Technologies Program: Multi-Year Research, Development and Demonstration Plan” (DOE, 2003b), DOE’s Office of Energy Efficiency and Renewable Energy program has set technical targets for the years 2005, 2010, and 2015 to reduce costs for biomass gasification/ pyrolysis and subsequent steam reforming. Specific goals include the reduction of costs for (1) biomass feedstock, (2) gasification operation (including efficiency), (3) steam reforming, and (4) hydrogen gas purification. Although no specific budget amounts were reported (except at a very high level of aggregation), major funding for R&D of hydrogen production from biomass is apparently in improving gasification/pyrolysis processes. The goals are quite ambitious. The committee’s economic analysis (Chapter 5) shows that gasification and the availability of large quantities of inexpensive biomass are major economic barriers for hydrogen derived from biomass. Although listed in the draft report, the EERE program seems to support photobiological hydrogen research, but specific funding levels are unclear. The DOE’s R&D targets for increasing the utilization efficiency of absorbed light and hydrogen production are extremely ambitious, and it is unclear how realistic they are. It appears that if such molecular projects are funded, they are for small amounts.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs Summary The committee’s analysis indicates the following: Considering the assumptions for future technology, biomass-to-hydrogen conversion is unlikely to produce hydrogen at a competitive price, even when compared with hydrogen generated from distributed natural gas. The environmental impact of growing significant quantities of biomass as energy crops, including engineered, high-yield crops, will most likely place significant strains on natural resources, including water, soil, land availability, and biodiversity. Because of the inherently high cost for collecting and transporting biomass, a biomass gasification plant will be limited in size, will not make full use of the economics of scale, and will be limited to certain geographic regions in the United States. Biomass-to-hydrogen conversion is a thermodynamically inefficient path for using solar energy. The use of biomass (residues), when co-fired (e.g., with coal) and coupled to subsequent carbon sequestration, might be an important technical option for achieving zero emission and, potentially, a net reduction of atmospheric CO2. Photobiological hydrogen production is a theoretically more efficient process, but significant fundamental molecular research is needed to identify and improve the limiting factors in order to evaluate fully this approach for hydrogen production. HYDROGEN FROM SOLAR ENERGY Introduction It has been estimated that solar energy has the potential of meeting the energy demand of the human race well into the future.29 One of the methods of recovering solar energy is through the use of photovoltaic (PV) cells. Upon illumina-tion with sunlight, PV cells generate electric energy. Commercial PV modules are available for a wide range of applications. However, they represent a miniscule contribution to U.S. electric power production. The current cost of electricity from a PV module is 6 to 10 times the cost of electricity from coal or natural gas. Therefore, if PV electricity were to be used to make hydrogen, the cost would be significantly higher than if fossil fuels were used. The key for solar energy to be used on a large scale for electricity or hydrogen production is cost reduction. This would require a number of advancements in the current technology. Current State of Technology Approximately 85 percent of the current commercial PV modules are based on single-crystal or polycrystalline silicon. The single-crystal or polycrystalline silicon cells are generally of the dimension of 10 to 15 centimeter (cm) (Archer and Hill, 2001). They are either circular or rectangular. In a module, a number of cells are soldered together. Each cell is capable of providing a maximum output of 0.6 volt (V), with the total module output approaching 20 V. The output current of each cell in bright sunlight is generally in the range of to 2 to 5 amps. The single-crystal silicon cells are made from wafers obtained by continuous wire sawing of single-crystal ingots grown by the Czochralski process. Similarly, a large portion of the polycrystalline silicon cells are made from ingots obtained by directional solidification of silicon within a mold. The wafer thickness is generally in the range of 250 to 400 microns. It is worth noting that nearly half of the silicon is wasted as “kerf” loss during cutting. Polycrystalline silicon cells are also made from silicon sheet or ribbon grown by other techniques (Archer and Hill, 2001). This process avoids the cost associated with cutting silicon ingots into wafers. The silicon wafers or ribbons are then further processed to develop p-n junctions and wire contacts. The array of cells is laminated using glass and transparent polymer, called ethylvinylacetate (EVA), to provide the final PV module. The modules are known to have long lifetime (10- to 25-year warranty from manufacturers). The current technology gives about 18 percent cell efficiency and 15 percent module efficiency.30 A second type of PV technology is based on deposition of thin films. PV cells are prepared by deposition of amorphous as well as microcrystalline silicon from a variety of techniques, including plasma-enhanced chemical vapor deposition, hot wire chemical vapor deposition, and so on. Polycrystalline thin-film compounds based on group II-VI of the periodic table, such as cadmium telluride (CdTe), and group I-III-VI ternary mixtures such as copper-indium-diselenide (CIS), have been used to make thin-film solar cells (Ullal et al., 2002). The thickness of deposited layers is much less than 1 micron. As compared with crystalline silicon solar cells, the thin-film technology potentially has a number of significant advantages in manufacturing: (1) lower consumption of materials; (2) fewer processing steps; (3) automation of processing steps; (4) integrated, monolithic circuit design leading to elimination of the assembly of individual solar cells into final modules; and (5) fast roll-to-roll deposition (Wieting, 2002). It has been estimated that for crystalline silicon solar cells, the complete process involves more than two dozen separate steps to prepare and process ingots, wafers, cells, and circuit assemblies before a module is complete (Wieting, 2002). On the other hand, thin-film module 29   Nathan Lewis, California Institute of Technology, “Hydrogen Production from Solar Energy,” presentation to the committee, April 25, 2003. 30   The efficiency in this section is defined at 25°C under 1000 W/m2 of sunlight intensity with the standard global air mass 1.5 spectral distribution. Thus, 15 percent module efficiency refers to peak watt efficiency (Wp) and implies that 15 percent of the incident sunlight energy is converted to electricity.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs production requires only half as many process steps, with simplified materials handling. Thin-film technology appears to hold greater promise for cost reduction, which has led to research by several laboratories over the past two or three decades. Some of the results in efficiency improvement of small laboratory research-size cells, typically of the size of 1 cm2, are shown in Figure G-15. Research cell efficiencies as high as 21.5 percent for copper-indium (gallium)-diselenide (CIGS) are reported (Ullal et al., 2002). Similarly, high efficiency of 16.5 percent has been reported for CdTe research cells. Amorphous silicon is deposited by using silane (SiH4) and hydrogen mixtures. In laboratory-scale cells of amorphous silicon, the highest efficiencies obtained are about 12 percent. One big challenge for thin-film solar cells is to overcome the large drop in efficiency from the laboratory-scale cell to that of a real module. For example, commercial modules of CdTe and CIGS have efficiencies in the range of 7 percent to 12 percent (as compared with laboratory-scale cell efficiencies of 16.5 percent and 21.5 percent). Similarly, commercial amorphous silicon modules have efficiencies less than 10 percent (Shah et al., 1999). The drop in efficiency as cell size is increased is substantial. Attempts are being made to increase the efficiency of amorphous and microcrystalline silicon cells by making dual and triple junction cells (Yang et al., 1997). This change leads to multiple layers, each having a different optimum band gap. However, the deposition of multiple layers increases the processing steps and therefore the cost. A final note is that amorphous silicon modules, when exposed to sunlight, undergo light-induced degradation, operating thereafter at a lower, stabilized efficiency (Shah et al. 1999; Staebler and Wronski, 1977). In spite of its promise, the thin-film technology has been unable to reduce the cost of solar modules, owing to low deposition rates that have led to low capital utilization of expensive machines. The yields and throughputs have been low. These plants need better inline controls. In recent times, owing to manufacturing problems, some corporations have shut down their thin-film manufacturing facilities. Clearly, easier and faster deposition techniques leading to reproducible results are needed. Also, deposition techniques that would not result in a substantial drop in efficiency from laboratory scale to module scale are required. Today there is no one clear “winner technology.” More than a dozen firms produce solar modules. Even the largest of these firms do not have world-class, large-scale produc- FIGURE G-15 Best research cell efficiencies for multijunction concentrator, thin-film, crystalline silicon, and emerging photovoltaic technologies. SOURCE: National Renewable Energy Laboratory.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs tion facilities (greater than 100 MWp worth of solar modules per year). This size limitation does not allow the economy-of-scale benefits for the solar cell production. Many companies use multiple technologies. The current cost of solar modules is in the range of $3 to $6 per peak watt (Wp). For solar cells to be competitive with the conventional technologies for electricity production, the module cost must come down below $1/Wp. Table G-9 provides cost estimates of producing electricity as well as hydrogen calculated by the committee. In the current scenario, with a favorable, installed cost of about $3.285/Wp, the electricity cost is estimated to be about $0.319/kWh (scenario Dist PV-C of Chapter 5). For a futuristic case with all of the expected technology and production advances, the anticipated installed cost of $1.011/ Wp provides electricity cost of $0.098/kWh (scenario Dist PV-F, Table E-49 in Appendix E). While this target is attractive for electricity generation, it does not produce hydrogen at a competitive cost. Energy is consumed in the manufacture of solar modules. It has been estimated by NREL that for a crystalline silicon module, the payback period of energy is about 4 years. For an amorphous silicon module this period is currently about 2 years, with the expectation that it will eventually be less than 1 year. Future Technology Photovoltaic Cells Various developments are likely to improve the economic competitiveness of solar technology, especially for thin-film technology. The current research on microcrystalline silicon deposition techniques is leading to higher efficiencies. Techniques leading to higher deposition rates at moderate pressures are being developed (Schroeder, 2003). Better barrier materials to eliminate moisture ingress in the thin-film modules will prolong the module life span. Robust deposition techniques will increase the yield from a given type of equip TABLE G-9 Estimated Cost of Hydrogen Production for Solar Cases Case Installed Cost ($/kW) Electricity Cost ($/kWh) Hydrogen Cost with Electrolyzer ($/kg) Current (Dist PV-C) 3285 0.319 28.19 (Dist PV Ele-C) Future (Dist PV-F) 1011 0.098 6.18 (Dist PV Ele-F) NOTE: See Appendix E for definition of the symbols for the solar technology cases. See also Tables E-48 and E-49 of Appendix E. ment. Inline detection and control methods will help to reduce the cost. Some of this advancement will require creative tools and methods. The committee believes that installed costs of roughly $1/ Wp are attainable. Material costs are quite low, but substrate material, expensive coating equipment, low utilization of equipment, and labor-intensive technology lead to high overall costs. It is expected that in the next decade or two, improvements in these areas have a potential to bring the cost much below $1/Wp. World-class plants with economies of scale will further contribute to the lowering of cost. For crystalline-silicon-wafer-based technology, the raw material costs by themselves are almost $1/Wp. However, improvements in operating efficiency, the cost of raw materials, and reduced usage of certain materials are expected to bring overall cost in the neighborhood of $1/Wp. A concept that has been proposed is the dye-sensitized solar cell, also known as the Grätzel cell (O’Regan and Grätzel, 1991). A dye is incorporated in a porous inorganic matrix such as TiO2, and a liquid electrolyte is used for positive charge transport. Photons are absorbed by the dye, and electrons are injected from the dye into n-type titania nanoparticles. The nanoparticles of titania are fused together and carry electrons to a conducting electrode. The dye gets its electron from the electrolyte, and the positive ion of the electrolyte moves to the other electrode (Grätzel, 2001). This type of cell has a potential to be low-cost. However, the current efficiencies are quite low, and the stability of the cell in sunlight is very poor. Research is needed to improve performance at both fronts. Another area of intense research is that on the integration of organic and inorganic materials at the nanometer scale into hybrid solar cells. The current advancement in conductive polymers and the use of such polymers in electronic devices and displays provides the impetus for optimism. The nano-sized particles or rods of the suitable inorganic materials are embedded in the conductive organic polymer matrix. Once again, the research is in the early phase and the current efficiencies are quite low. However, the production of solar cells based either solely on conductive polymers or on hybrids with inorganic materials has a large potential to provide low-cost solar cells. It is hoped that one would be able to cast thin-film solar cells of such materials at a high speed, resulting in low cost. Regarding production costs, all of the technologies discussed so far convert solar energy into electricity and use the electricity to generate hydrogen through the electrolysis of water. Since PV cells produce dc currents, the electric power can be directly used for electrolysis. As discussed in the section above on electrolyzers, considerable cost reductions are anticipated, which will lower the cost of hydrogen from solar cells. These cost reductions will be particularly valuable for solar cell electricity because the low usage factor associated with PV modules also contributes to the low usage of electrolyzers. This contributes heavily to the cost of hydro-

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs gen produced. For example, in the committee’s analysis of costs discussed in Chapter 5 (summarized in Table G-9), for the future optimistic case the cost of hydrogen is calculated to be $6.18/kg (Dist PV Ele-F). For this case, the cost of the installed PV panels, including all of the general facilities, is estimated to be $1.011/Wp, and is used in conjunction with an electrolyzer that is assumed to take advantage of all of the advancements made in the fuel cell. The PV part is responsible for $4.64/kg, and the electrolyzer is $1.54/kg. Compared with this, the cost of hydrogen from a future central coal plant at the dispensing station is estimated to be $1.63/ kg with carbon tax (CS Coal-F). This cost implies that for a PV-electrolyzer to compete in the future with a coal plant, either the cost of PV modules must be reduced by an order of magnitude or the electrolyzer cost must drop substantially from $125/kW. A factor contributing to this is the low utilization of the electrolyzer capital. It has been proposed to use electricity from the grid to run the electrolyzer when solar electricity is unavailable. This use increases the time on-stream for the electrolyzer; however, in the long term, for solar to play a dominant role in the hydrogen economy, it cannot rely on power from the grid to supplement equipment utilization. Therefore, while electricity at $0.098/kWh from a PV module can be quite attractive for distributive applications where electricity is directly used, its use in conjunction with electrolysis to produce hydrogen is certainly not competitive with the projected cost of hydrogen from coal. Direct Production Research is being done to create photoelectrochemical cells for the direct production of hydrogen (Grätzel, 2001).31 In this method, light is converted to electrical and chemical energy. The technical challenge stems from the fact that energy from two photons is needed to split one water molecule. A solid inorganic oxide electrode is used to absorb photons and provide oxygen and electrons. The electrons flow through an external circuit to a metal electrode, and hydrogen is liberated at this electrode. The candidate inorganic oxides are SrTiO3, KTaO3, TiO2, SnO2, and Fe2O3. When successful, such a method holds promise of directly providing low-cost hydrogen from solar energy. Regarding production costs, it seems that a photoelectrochemical device in which all of the functions of photon absorption and water splitting are combined in the same equipment may have better potential for hydrogen production at reasonable costs. However, it is instructive to do a quick “back of the envelope” analysis for the acceptable cost by such a system. It is assumed that cost per peak watt for a photoelectrochemical device is the same as that for the possible future PV modules (see Table E-48 of Appendix E.) It is further assumed that this energy is recovered as hydrogen rather than as electricity. Therefore, a recovery of 39.4 kWh translates into a kilogram of hydrogen. This implies that 4729 kWe worth of solar plant in the Dist PV-F spreadsheet will produce about 576 kg/day of hydrogen (assuming an annual capacity factor of 20 percent). At the total cost of $0.813 million per year, this gives $3.87/kg of hydrogen! This cost is still too high when compared with that of hydrogen from coal or natural gas plants. It implies that photoelectrochemical devices should recover hydrogen at an energy equivalent of $0.4 to $0.5/Wp. This cost challenge is similar to that for electricity production from the solar cells. Advantages and Disadvantages of Solar Energy Solar energy holds the promise of being inexhaustible. If harnessed, it can meet all of the energy needed in the foreseeable future. It is clean and environmentally friendly. It converts solar energy into hydrogen without the emission of any greenhouse gas. Because of its distributed nature of power production, it contributes to the national security. There are certain challenges associated with the use of solar energy. The intermittent nature of sunshine, on both a daily and a seasonal basis, presents a number of challenges. A backup system, or a storage system for electricity/hydrogen, is needed for the periods when sunshine is not available and power demand exists. Furthermore, this intermittent availability means that four to six times more solar modules have to be installed than the peak watt rating would dictate. This intermittency also implies that a significant decrease in the module cost is required. Another challenge is to ensure that no toxic materials are discharged during the fabrication of solar cells and over the complete life cycle of the cell. Such questions have been raised in the context of cadmium-containing solar cells, and public perception in such cases will play a key role. Challenges and Research and Development Needs Large-scale use of solar energy for hydrogen economy will require research and development efforts on multiple fronts. In the short term, there is a need to reduce the cost of thin-film solar cells. This reduction will require the development of silicon deposition techniques that are robust and provide high throughput rates. New deposition techniques at moderate pressures with microcrystalline silicon structures for higher efficiencies are needed. Inline detection and control and the development of better roll-to-roll coating processes can lead to reductions in the manufacturing costs. Increased automation will also contribute to the decreased cost. Issues related to a large decrease in efficiency from small laboratory samples to the module level should be addressed. In the short run, thin-film deposition methods can potentially gain from a fresh look at the overall process from the laboratory scale to the manufacturing scale. The research in 31   Nathan Lewis, California Institute of Technology, “Hydrogen Production from Solar Energy,” presentation to the committee, April 25, 2003.

OCR for page 198
The Hydrogen Economy: Opportunities, Costs, Barriers, and R&D Needs this area is expensive. Some additional centers for such research in academia with industrial alliances could be beneficial. It will be necessary to collect multifunctional teams from different engineering disciplines for such studies. In the midterm to long term, organic-polymer-based solar cells hold promise for mass production at low cost. They have an appeal for being cast as thin films at very high speeds using known polymer film casting techniques. Currently, the efficiency of such a system is quite low (in the neighborhood of 3 to 4 percent or lower), and stability in sunlight is poor. However, owing to the tremendous development in conducting polymers and other electronics-related applications, it is anticipated that research in such an area has a high potential for success. Similarly, the search for a stable dye material and better electrolyte material in dye-sensitized cells (Grätzel cells) has a potential to lead to lower-cost solar cells. There is a need to increase the stable efficiency of such cells; a stable efficiency of about 10 percent could be quite useful. In the long run, the success of directly splitting water molecules by using photons is quite attractive. Research in this area could be very fruitful. Department of Energy Programs for Solar Energy to Hydrogen The current DOE target for photoelectrochemical hydrogen production in 2015 is $5/kg H2 at the plant gate. Even if this target is met, solar hydrogen is unlikely to be competitive. Therefore, beyond 2015 a much more aggressive cost target for hydrogen production by photoelectrochemical methods is needed. Since photoelectrochemical hydrogen production is in an embryonic stage, a parallel effort to reduce the cost of electricity production from PV modules must be made. A substantial reduction in PV module cost (lower than $1/Wp), coupled with a similar reduction in electrolyzer costs (much below $125/kW at a reasonably high efficiency of about 70 percent based on lower heating value), could provide hydrogen at reasonable cost. In the long run, considering the environmental issues associated with fossil fuels and considering the limitless supply of solar energy, this has a potential to be quite attractive. This option will be especially attractive if advances in battery technology are unable to substantially increase the electricity storage density (based on mass of battery) and greatly reduce the cost of batteries. Therefore, it is recommended that thin-film technologies and other emerging PV technologies that hold the promise for cost reduction be aggressively pursued. As stated earlier, it means that more efficient and robust methods for thin-film coating must be developed. Organic-polymer-based solar cells should also be funded. There is tremendous development underway in conducting polymers for light-emitting diodes and other display technologies. The potential of these materials for solar cell PVs must be actively explored. Summary All of the current methods and the projected technologies of producing hydrogen from solar energy are much more expensive (greater than a factor of 3) when compared with hydrogen production from coal or natural gas plants. This is due partly to the lower annual utilization factor of about 20 percent (as compared with say, wind, at 30 to 40 percent). This high cost puts enormous pressure on the need to reduce the cost of a solar energy recovery device. While an expected future installed module cost of about $1/Wp is very attractive for electricity generation and deserves a strong research effort in its own right, this cost fails to provide hydrogen at a competitive value. The raw material cost for crystalline silicon-wafer-based technologies is a large fraction of the $1/Wp value and is therefore less likely to provide hydrogen economically. On the other hand, thin-film technologies do not use much raw material in thin films themselves but require tremendous progress in the deposition technology. There is a need for a robust deposition method that would have a potential to reduce cost much below $1/Wp. Emerging polymer-based technologies have a potential to provide low-cost devices to harness solar energy. It is apparent that there is no one method of harnessing solar energy that is clearly preferable. However, it appears possible that new concepts may emerge that would be competitive. The benefits of such developments would be very substantial. In the future, as the cost of the fuel cell approaches $50 per kilowatt, the cost of an electrolytic cell to electrolyze water is also expected to approach a low number (about $125/kW). With such low-cost electrolyzer units, the electricity cost of about $0.02 to $0.03/kWh is expected to result in a competitive hydrogen cost. It is also estimated that for a photoelectrochemical method to compete, its cost must approach $0.04 to $0.05/kWh. The order-of-magnitude reductions in cost for both hydrogen processes are similar.