The Role of the U.S. National Innovation System in the Development of the PEM Stationary Fuel Cell

NISTIR 7161
The Role of the U.S. National Innovation System in the Development of the PEM Stationary Fuel Cell

III. Drivers of Innovation in the Fuel Cell Industry

The first subsection describes the historical drivers of innovation in the fuel cell industry. Much of this information is drawn from a book about the founding and subsequent development of Ballard Power, a well-known Canadian fuel cell manufacturer. 1 The second subsection focuses on how public policy affects the development of fuel cell technology including federal energy R&D expenditures, public-private partnerships, environmental regulation, deregulation of energy markets, and the facilitation of standards. The third subsection explores the drivers of innovation for a single company, Mechanical Technology Inc. (MTI), and the subsequent creation and spin-off of Plug Power from that company. The sources for this material include interviews with three former and current Plug Power employees who were present at its founding. 2 In addition, Plug Power's 10-K Securities and Exchange Commission filings provided important financial information as well as verification of historical timelines.

As a historical driver of innovation in the fuel cell arena, the role of the government space program cannot be minimized. The space program proved the first viability of fuel cell technology, albeit for limited application and at low power levels. On the other hand, the role of government in advancing the fuel cell from an expensive item with a single mission in outer space to a less expensive item with large-scale commercial applications on earth cannot be tied to a single government program. It is undeniable that in the 1980s fuel cell technology needed government funding to advance. However, it was not a single government program, agency, or company that is responsible for the PEM fuel cell innovation that occurred during this and subsequent periods. Certain programs have made key funding decisions or produced new forms of knowledge at key moments. Often, these funds or knowledge have been fortuitous enough to provide the impetus for private actors to continue their pursuit of the technology. Since large-scale commercialization of fuel cells is yet to come, the development of this technology is still a work in progress.

A. Historical Drivers of Innovation

1. Early Fuel Cells

During the first 100 years of fuel cell development, much of the innovation originated from small groups of scientists. In 1839, Sir William Robert Grove, a Welsh judge, inventor, and physicist, created the first fuel cell. He reacted hydrogen and oxygen at catalytic platinum electrodes in the presence of an electrolyte to produce electricity and water. The invention did not produce enough electricity to be useful, however.

In 1889, Ludwig Mond and Charles Langer improved on Grove's design to make the world's first working fuel cell. They are also credited with originating the name "fuel cell." However, they decided their device had little commercial application due to the high cost of platinum.

In 1932, engineer Francis T. Bacon began his vital research into fuel cells. Early cell designers used porous platinum electrodes and sulfuric acid as the electrolyte bath. However, using platinum was expensive, and using sulfuric acid was corrosive. Bacon improved on the expensive platinum catalysts with a hydrogen and oxygen cell using a less corrosive alkaline electrolyte and inexpensive nickel electrodes. Bacon continued his work for three decades and eventually transferred his knowledge to the Pratt and Whitney Division of United Aircraft Corporation in Connecticut . This company subsequently became United Technologies Corporation (UTC), which is today one of the world's largest manufacturers of fuel cells.

2. GE and the PEM Fuel Cell

Tom Grubb of General Electric (GE) explored fuel cell research in the early 1950s, and, in 1954, he developed the first PEM fuel cell. In the late 1950s, government funding began to affect fuel cell development. U.S. Army contracts provided $1.1 million in funding for GE's fledgling PEM fuel cell program; ultimately, this led to work with the newly formed National Aeronautics and Space Administration (NASA) and the development of PEM fuel cells that powered several Gemini missions. 3

Typical of many technologies in the 1960s, fuel cell development accelerated with its selection as a technology for powering U.S. spacecraft. The GE fuel cells, using large amounts of platinum and pure gases, were very expensive to produce. Cost was not, however, an issue. Although GE spent $8.5 million of its own money over the 1960s and 1970s trying to develop fuel cells using cheaper materials, it ultimately stopped pursuing that research and instead sold the technology. During this period, the biggest improvement in GE's fuel cells came from the adaptation of a new membrane known as Nafion from the DuPont Corporation.

3. Drivers of Innovation in the 1980s

In the 1970s and early 1980s, most government and industry money in the United States was spent on the development of phosphoric acid and ceramic electrolyte fuel cells. Los Alamos National Laboratory was the only organization conducting PEM fuel cell research, and its funding was very modest. 4 In 1983, a Canadian company called Ballard Research (founded by Geoffrey Ballard; now Ballard Power), of Vancouver , British Columbia , received the first Canadian government grant for PEM fuel cell research. By this time, many of the GE patents related to PEM fuel cells had expired, thereby allowing the technology to be exploited by other companies. The advantage of PEM fuel cells over the other types was their ability to run on low-purity hydrogen. Their biggest disadvantage was their cost, particularly for the Nafion membrane and platinum electro-catalyst.

James Huff led the Los Alamos team that conducted PEM fuel cell research in the 1980s. The Los Alamos group consisted of approximately a dozen people and had an annual budget of close to $1 million. About the same time that Ballard was starting its contract to build a fuel cell for Canada , researchers at Los Alamos and Texas A&M showed that the amount of platinum required for the fuel cell could be reduced by a factor of 10. Their work convinced the people at Ballard that a PEM fuel cell might become economically feasible. For the remainder of the 1980s, Ballard and Los Alamos watched each other's progress; sometimes, breakthroughs by one group would encourage the other to keep pursuing this embryonic technology.

By the mid-1980s, Ballard had produced continuous improvements in its fuel cell. Working with limited resources, the Ballard researchers came to believe that a shoestring budget had helped, not hindered, their early progress. Sometimes, small changes in one area precipitated substantial leaps in performance. For example, the flow field design to channel how hydrogen flows through a stack of cells to produce electricity used GE's old parallel design. When Ballard changed its flow field pattern on the graphite plates that support the membrane electrode assembly, it achieved between two- and fourfold increases in performance in a few weeks' time. Geoffrey Ballard observed this about PEM fuel cell technology: "It has been a very forgiving technology.the system gets simpler, not more complex-which is the other measure of a good technology." 5

Another hurdle that Ballard needed to cross in the mid-1980s was to reduce the cost of the membrane. A breakthrough occurred when a Ballard scientist obtained a piece of membrane developed by DuPont rival Dow Chemical in 1986. Dow had developed an experimental polymer membrane of its own which had lower resistance than Nafion. The new membrane produced four times as much power as the Nafion membrane. 6 It proved to be a "eureka" moment for the Ballard scientists. When Ballard showed its progress to the Los Alamos scientists, the results convinced the latter that PEM fuel cells might be a potential power source for electric vehicles. This conclusion demonstrates the synergy of the parallel research efforts. It was Los Alamos research that had convinced Ballard that the platinum problem could be solved; now, Ballard research convinced Los Alamos scientists that the membrane could potentially produce sufficient power to operate a motor vehicle.

Thus, the innovative drivers of PEM fuel cell research in the 1980s emerged primarily from two places, Ballard Research in Canada and Los Alamos National Laboratory in New Mexico . Each group worked with relatively few resources. All of Los Alamos 's money came from public funding, as did all of Ballard's initial Canadian funding. Ballard eventually received private venture capital money in the late 1980s. 7

B. Public Policy as a Driver of Innovation

1. Government Support of Energy R&D

The rationale for government support of basic R&D is well known and widely accepted. Recent research also indicates that private companies underfund early-stage technology development R&D. 8 Overall, the amount of federal R&D dollars devoted to energy R&D has steadily declined in real terms over the last 20 years (figure 3).

Figure 3 - Federal R&D funding by budget function. Source: NSF (2002).
Figure 3 - Federal R&D funding by budget function.

Decaux (2003) finds that funding for R&D in alternative energy technologies is inadequate and inconsistent. She also finds that throughout OECD countries, overall energy R&D is declining and shifting from long-term to near-term projects. This trend has been exacerbated by utility deregulation. Her recommendations for an improved policy environment include providing direct incentives for energy R&D, targeting long-term projects, and clearly mapping the process from the technology drawing board to commercialization. 9 While the Departments of Energy (DOE) and Defense (DOD) provide direct incentives for basic research and for demonstration projects, programs such as the Advanced Technology Program support Decaux's recommendation of providing platforms for moving the technology from the drawing board to commercialization.

2. Public-Private Partnerships and the Advanced Technology Program

The U.S. Department of Commerce oversees ATP through the National Institute of Standards and Technology. ATP shares the cost of high-risk R&D projects with private companies in order to accelerate the development of innovative technologies for broad national benefit.

ATP partners with companies of all sizes, universities, and nonprofits, encouraging them to take on greater technical challenges with potentially large benefits that extend well beyond the innovators-challenges they could not or would not undertake alone. 10 ATP awards are selected through open, peer-reviewed competitions. The ATP selection process does not discriminate across technologies as long as projects meet the technical and economic criteria. This last point is important, because ATP has not favored one approach over another. It has funded research projects focused on PEM, Direct Methanol, and Solid Oxide fuel cell applications.

ATP awarded its first fuel cell project in 1997. In 1998, ATP supported the development of premium power technologies through a competition focused on that technology area. The portfolio of 13 projects and $47 million in funding that emerged from this solicitation were aimed at accelerating the development of the technologies in step with the changes taking place in the ways electricity is generated and used-distributed electric generation on site and free from the grid, and wireless portable power. Through the 1998 premium power solicitation, ATP became the first government agency to provide significant funding of R&D aimed at development of small, distributed, stationary power technologies such as fuel cells that could be used for generating electricity on site for homes and businesses. This continues to be a dominant theme in its most recent fuel cell awards. In the last two years, ATP has also emerged as a leading enabler of micro fuel cell technology that can meet the ever-increasing functionality and power needs of wireless electronics and, e.g., power a cell phone for a month before recharging. Overall, ATP has funded 25 projects in the fuel cell area. Table 2 shows the total number of awards in fuel cell projects and the total amount of government and industry cost-share funds committed.

Table 2. ATP's 25 Fuel Cell Projects, 1997-2003

Total number of active or completed projects	25
Estimated ATP share of funding	$60 million
Industry cost-share of funding	$54 million
Total funding	$114 million

The following section and table 3 provide a brief description of the knowledge gained from three early fuel cell projects from the ATP Premium Power Focused Program. These projects involved competing approaches to PEM fuel cells by Plug Power, Avista Labs, 11 and H Power-Nuvera. Each company approached the market and its PEM fuel cell design from a different perspective, with the goal of meeting different needs in the short term: Plug Power targeted residential, mass market applications; Avista targeted off-grid, backup power and power quality; and the joint venture between H Power and Nuvera targeted telecommunications. 12 The role ATP plays in the U.S. national innovation system is illustrated by these three projects. Each company is small and has limited funds to conduct longer-term research. ATP allows companies to conduct early-stage research development projects that might not otherwise be funded because the private venture capital market demands relatively short-term payback and may not adequately optimize technology development because of the extended time horizon needed for high-risk research. 13

Table 3. Fuel Cell Case Study Compilation Chart

	Plug Power	Avista	H Power
Amount of total funding	$9 737 848	$3 224 510	$6 376 772
Award period	2 years (5/99-5/01)	2.5 years (11/98-4/01)	2 years (1/99-1/01)
Research focus	High temperature membranes	Modular, self-hydrating fuel cell cartridges	Simplified one-piece membrane assembly and reformer integration
Employee growth (1998-2003)	22 to 339	7 to 45	50 to 135
Target market	3 kW to 7 kW residential, on-grid applications	Uninterrupted power source and remote, off-grid applications	Backup and primary telecom power and residential units
Initial public offering (IPO) status	IPO completed in November 1999 $58 million stock offering completed 10/13/03	Avista Labs is now a privately funded Washington corporation with several venture capital funds as investors	Plug Power buys company on 5/15/03
Strategic partners	DTI, General Electric Power Systems, Vaillant, Advanced Energy Systems, Engelhard, SRI, Polyfuel, Celanese	UOP, Black & Veatch, 3M, Airgas	Nuvera, formerly Epyx, is focused on distributed and vehicle uses of fuel cell power systems

a. Plug Power

Plug Power of Latham, New York , was founded in 1997, and in six years has grown from 22 to over 300 employees. Plug Power's early investors formed the company as a joint venture between Detroit Edison (DTE) Energy Company, Michigan 's largest electric utility, and Mechanical Technology Inc., an early developer of fuel cell technologies. Now a stand-alone, publicly traded company, Plug Power added General Electric Company and Sempra Energy, a subsidiary of Southern California Gas Company, as major investors in 2000. 14 Plug Power has built and deployed over 400 PEM fuel cell systems to date, including GenSys prime power and GenCore backup power 5 kW units.

The ATP-funded $9.7 million project allowed Plug Power and its partners (Polyfuel/SRI and Celanese) to attack the problem of carbon monoxide poisoning of the catalyst head-on by pursuing higher temperature polymer membranes, as well as by designing the fuel cell stack for higher operating temperatures. 15 This is desirable because increased temperatures reduce the poisoning effect of carbon monoxide in the fuel cell systems. PEM fuel cells produce their fuel-hydrogen-by reforming common fuels such as natural gas and propane. However, this reformate fuel stream is contaminated with trace levels of carbon monoxide because the decomposition of hydrocarbons from the original fossil fuels is not 100 % complete. Fuel cells operated at lower temperatures with the current generation of polymer (i.e., Nafion) membranes are less robust, with as little as 50 µL/L of carbon monoxide shutting the system down. With the Celanese high-temperature, non-fluoropolymer membrane, which is chemically different from Nafion, Plug Power has succeeded in operating a PEM-based fuel cell system at temperatures above 150 °C, demonstrating 20 000 µL/L carbon monoxide tolerance over a period of more than 5 000 h. This breakthrough should result in more reliable operation with simpler reformers and would also allow for more efficient use of the leftover heat and reduce the complexity and size of the overall system.

b. Avista

Avista Labs was created in 1996 as a full subsidiary of Avista Corporation, formerly Washington Water Power of Spokane, Washington, to commercialize new energy technologies. Making quick progress on its new responsibility, Avista Labs unveiled its fuel cell generator prototype at the International Fuel Cell Seminar in 1998. By March 2000, Avista was issued a comprehensive patent covering 162 claims for its PEM fuel cell power system. The summer of 2000 brought the first alpha test units to the field. Avista's rapid development speaks to the need for innovation and timely financial support. Avista currently has fuel cells installed in over 80 locations in the United States and abroad, and counts among its customers major telecommunications providers, uninterruptible power system providers, government communication sites, utilities, and railroad suppliers.

ATP funding came at a crucial time for Avista Labs. In 1997, the Fortune 500 parent company, originally Washington Water and Power (renamed Avista Corporation), was in the midst of a transition from the status quo guard of regulated power markets to the new deregulated markets. In preparation for deregulated power markets, Avista was involved in several research projects aimed at diversifying its value-added business units, including fuel cells. The new unit tasked with fuel cell development was Avista Labs, which started with only seven scientists and engineers, and a concentration on the design of modular PEM fuel cells. In 1998, Avista applied for and won an ATP award in the Premium Power Focused Program. The 2.5-year project began in November 1998 and allowed Avista Labs to pursue its unique technical approach toward the development of a commercially viable PEM fuel cell at a faster pace and with more flexibility than it otherwise would have been afforded. Avista Labs president Kim Zentz commented that "ATP allowed us to bring prototypes to decision makers much more quickly. Without ATP it would also have been much harder to prove the modular design concept."

Investors, including internal decision makers, require a "proof-of-concept" for new technologies before they are willing to fund the more costly product development phase. ATP funding helped Avista Labs prototype design ideas rapidly and speed up the cycle time of developing a new modular approach to PEM fuel cells. As a result of the more timely achievement of a proof-of-concept of modular fuel cells, the introduction of this technology to the residential market was accelerated by three years. ATP support filled the gap left by internal funding constraints and a lack of outside equity sources. 16

Avista Labs engineers opted to take a different approach to PEM fuel cells using a modular, cartridge concept instead of a stack of cells. 17 Cartridges can be replaced without interrupting generation and can use the water produced in conjunction with fan-forced air to promote cooling and self-hydration. These features simplify system performance by reducing balance-of-plant to a minimum. The advanced cartridges incorporate embedded control functions to protect the membrane electrode assemblies from conditions leading to loss of life or failure. A simple, retractable ballpoint pen was the inspiration for the hydrogen valves, which where crucial to making the cartridge design workable. Using injection-molded, extremely cheap parts for the valves, Avista engineers developed cartridges that click into place and safely keep the volatile hydrogen fuel within the cartridge. The alpha test units incorporate these features and utilize leftover water from the reactions in the fuel cell to keep the membrane wet. This self-hydrating technology increases reliability, reduces maintenance, and is ultimately more consumer-friendly because cartridges last longer. Estimated to last for five years or more, the individual cartridges are easy to change and require no more skill than replacing a book on a bookshelf.

products require a few watts to 1 kW of power. Sub-kilowatt systems are also suitable for many mobile applications involving light utility vehicles and auxiliary power units used in conjunction with other power-generating systems, e.g., on-board battery chargers in electric vehicles.

The core innovation that H Power developed for the ATP project is a one-piece membrane electrode assembly which eases assembly and lowers cost. A new bipolar plate design improves the distribution of hydrogen fuel to the membrane and incorporates channels for cooling and hydrating the cell. The metallic plate design is projected to further reduce manufacturing costs and increase the reliability of the fuel cell stack. Work continues to improve the electrical performance and the corrosion resistance of the bipolar plates.

H Power received two patents for work done during the ATP-sponsored project. These advances have allowed H Power to move more quickly into the alpha test phase for larger PEM fuel cell units, thereby reducing the time to market. H Power had success with smaller output units as demonstrated by a New Jersey Department of Transportation contract to replace solar-powered variable message highway signs with fuel cell power systems. The systems provide backup power for the state's fleet of variable-message road signs. More than 60 of the systems have been deployed and operate daily.

Nuvera Fuel Cells was formed in April 2000 when ATP-funded Epyx Corporation (a former subsidiary of Arthur D. Little, Inc.) merged with Italy 's De Nora Fuel Cells, SpA, and Amerada Hess Corporation, a leading U.S. East Coast energy company. Nuvera combines Epyx's reformer process to produce hydrogen feed for fuel cells with De Nora's leadership in electrochemical, membrane, and fuel cell technology. A 5 kW fuel cell power system running on propane and specifically tailored to the needs of the telecommunications industry was designed and built. In March 2001, this first-of-a-kind unit was delivered to Verizon to test for powering a remote cell telephone tower. Such systems could be used in the future to provide primary or backup power for cell towers and telecom switch nodes that could benefit from on-site, reliable generation of DC electricity.

3. Environmental Regulations

Although the expected service life requirements and reformers in automotive and stationary residential applications are different, the PEM fuel cells at the heart of the systems are quite similar. Together, automotive and stationary applications of fuel cells have the potential to reduce U.S. carbon dioxide emissions by two-thirds. 18

c. H Power and Nuvera

The H Power Company of Belleville , New Jersey , was founded in 1989 with the goal of commercializing PEM fuel cells. In 2000, the company employed approximately 70 people at its Belleville R&D and manufacturing facilities, and 15 in its Clifton , New Jersey , administrative offices; it also maintained a project office at McClellan Air Force Base in Sacramento . H Power has a subsidiary company, H Power Enterprises of Canada, in Ville St. Laurent , Quebec, which employs an additional 50 people. On March 25, 2003 , Plug Power purchased H Power in a stock-for-stock exchange.

H Power was attempting to commercialize sub-kilowatt fuel cell systems for a variety of telecommunications and backup power applications. Typically, these

When powered directly by hydrogen, fuel cells produce only water vapor as a byproduct. Even when using fossil fuels, PEM fuel cells result, because of their greater energy conversion efficiency, in reduced organic and nitrous oxide emissions that cause air pollution, which is a leading cause of lung-related diseases and ozone depletion. Therefore, environmental regulations targeted at both automobiles and electricity generators have significant positive impacts on firms' incentives to invest in fuel cell R&D. In fact, California 's zero emissions law has led virtually all car companies, even those initially skeptical of fuel cells, to consider PEM fuel cells. 19 Notwithstanding these insights, as discussed above, there remains under-investment in energy-related R&D due to a high level of technical and business risk, as well as appropriability issues and shortened investor time horizons. 20

Although the U.S. administration's position on the reduction in greenhouse gases is uncertain, environmental regulatory policies may have relatively little direct impact on private investment in traditional electric utility production. Using event study analysis to examine 22 milestones, J. David Diltz found that events leading to the passage of the Clean Air Act Amendments in 1990 did not have a significant negative impact on investors' perceptions of electric utilities. 21 Combined, these factors suggest that environmental regulation has a positive impact on the demand for funds for investment and does not curtail the supply of investor funds. This is an important finding, as an increase in demand for funds accompanied by a decrease in supply could drive up investment costs and potentially lead to lower overall levels of investment.

An area where the utility industry has been responsive is air quality regulation, in that almost all of the new-generation capacity built in the 1990s uses large gas turbines running on cleaner natural gas. The adoption of these large gas turbines by utilities was aided by the fact that the technology matured and the costs dropped around the same time that the regulations became effective. It is a matter of conjecture as to whether utilities might have been more willing to resist the regulations had a relatively inexpensive substitute not been available. This point notwithstanding, environmental regulations mandating a certain level of compliance are always subject to political pressures unless reasonable substitutes are available and the costs are not prohibitive. In terms of fuel cells, if, several years from now, there is increasing political pressure to reduce greenhouse gases and the price of delivering emission-free fuel cells is still substantially more than current emission-emitting technologies, how much will the body politic be willing to subsidize fuel cells, or other emission-reducing technologies, in order to reduce emissions?

An unintended consequence of the switch to natural gas turbines is that shortages in natural gas are occurring. The North American gas supply is now declining. 22 This circumstance is troublesome, because many of the fuel cells being developed-including Plug Power's-reform their hydrogen from natural gas.

Although environmental policy has, on balance, spurred investment in fuel cells, fuel cell sales are unlikely to be similarly aided. Any environmental benefits offered by PEM fuel cells are a public good for which consumers are reluctant to pay an extra premium. Scott Samuelsen of the University of California-Irvine and director of the National Fuel Cell Research Center confirms the difficulty of using the environmental benefits of fuel cells as selling points. "That [fuel cells] have no plumes of smoke, no turning parts, no 700-foot-high dams, and no noise to upset people is bound to go unnoticed." 23

4. Utility Deregulation

The confluence of technological development and changes in economists' thinking about so-called "natural monopolies" has led to a reevaluation of electricity regulation over the past 30 years. 24 Of all the federal environmental and regulatory changes, deregulation of electric and natural gas markets has had the most significant impact on the economic viability of stationary PEM fuel cells. 25 Such efforts are likely to have wide-ranging effects including a lowering of the nation's core rate of inflation. Although such regulatory changes enable small businesses to compete against the large power companies, they have at the same time increased the degree of competition with alternative fuel technologies and thereby limit the total overall R&D dollars in the energy sectors because companies believe they cannot appropriate their full investment.

The two most significant pieces of federal legislation in this regard are the Public Utilities Regulatory Policies Act of 1978 and the Energy Policy Act of 1992. The former promoted the use of cogeneration in independent power and industrial projects by exempting generators from federal and state regulatory control. This regulatory change was made in response to technological changes that allowed smaller scale generators to be combined with other applications that use the waste heat and energy from the generation process. The latter act promoted sweeping changes in the generation and transmission of electricity. When coupled with Federal Energy Regulatory Commission orders enacted in response to this piece of legislation, the number of independent generators exempt from regulation was expanded, all interstate transmission lines were made available at the same cost for all power generators, and all generators were required to make information available regarding their generating capacity. Although federal legislation affected interstate electric markets, individual state deregulation efforts promoted consumer choice and competition. Over half the states in the country have enacted retail competition plans to emulate those pioneered by Massachusetts , Rhode Island , and California . 26

One impact of this deregulation is that utilities are no longer guaranteed to recover their fixed investment costs, which has prompted utilities to become more risk-averse and choose smaller investments. For example, before deregulation, the typical utility owned the power plants that generated the electricity for the customers in their defined service area. When utilities prepared long-range R&D budgets, they knew that if they wanted to reduce the costs of power, they needed to improve the productivity at their own power plants. Now that utilities can purchase cheaper power on the open market, the incentive to innovate their own power infrastructure is diminished. Decaux finds that this has contributed to the shift to short-term R&D projects. 27 Deregulation may also have had a negative impact on utilities' investments in emission-reducing technologies as well. Roberts reports that by 1997 utilities had invested $894 million in energy-efficiency programs as compared to 1992 projections of $2.7 billion. 28

5. Development of Standards

In order to assess the total costs of fuel cells relative to current utility costs, consumers need a better understanding of the energy efficiency and potential savings of small residential fuel cell systems across the range of environmental and seasonal conditions. To help meet this goal, Plug Power is participating in a NIST testing program outside of the ATP project to investigate the energy content of various fuel sources, the electrical power and electrical energy generated by the fuel cell, and the thermal output of the fuel cell under a variety of different load-demand conditions. This program will provide purchasers with a realistic estimate of annual electrical energy output, thermal energy output, and fuel usage. 29 In addition to electricity generation, the heat generated by this 5 kW unit will be captured to provide estimates of the space and hot water heating requirements that can be met by this system. These efforts continue NIST's history of collaborating with DOE and industry to develop metrology and standards that define the energy efficiency of heat pumps, water heaters, gas furnaces, and other household appliances.

A remaining regulatory hurdle with a major impact on the commercialization prospects of grid-connected stationary fuel cells is "net metering." Typically in residential electricity generation, utilities meter the amount of electricity supplied to the grid separately from that pulled from the grid. Federal legislation requires only that utilities purchase electricity from independent generators at an "avoided cost" basis, which facilitates utilities' use of separate prices for electricity flowing from house to grid and for electricity flowing from grid to house. In essence, owners of stationary fuel cells sell any excess power to the grid at wholesale prices. Yet should they need to purchase supplemental power from the grid, they must pay retail prices.

C. Plug Power

1. Drivers of Innovation for Fuel Cells at MTI and the Founding of Plug Power

In 1961, two entrepreneurial engineers founded Mechanical Technology Inc. For the next three decades, this small company pursued government research contracts in the area of precision instruments and energy-related research such as flywheels. In the late 1980s, the company was carrying out research on the Automotive Stirling engine for DOE in a project managed by NASA. 30 As this contract neared its conclusion, MTI tried to find new areas of system- and energy-related work to pursue, particularly since this program had employed many people. Because MTI's Stirling engine used hydrogen as a working gas and MTI was involved in non air-breathing underwater propulsion studies, PEM fuel cells were seen as a promising next phase of research that utilized MTI's core system expertise.

In 1992, MTI received a $160 000 grant from the New York State Energy Research Development Authority (NYSERDA) to deliver a PEM fuel cell for hybrid electric vehicles. William D. Ernst, current vice president and chief scientist at Plug Power, developed and directed the project. Dr. Ernst originally led the team working on the Stirling engine projects and flywheel research. While MTI had significant systems and mechanical engineering skills, it lacked critical electrochemical fuel cell expertise. To remedy this, MTI, during the 1992-94 time frame, attempted to purchase a fuel cell stack from Texas A&M, H Power, and other places, but could not find a stack that was large enough for its purposes and that would perform to its specifications. Ballard would sell them a stack, but the price was prohibitive. At this point, Dr. Ernst decided to have MTI develop its own fuel cell stack and, in early 1995, hired Wayne Huang. Dr. Huang brought to MTI critical expertise in electrochemistry and fuel cells. Previously, he had worked at the Los Alamos National Laboratory on fuel cell technology, where he learned the decal printing method from its inventor, Mahlon Wilson. Eventually, MTI licensed the decal printing technology from Los Alamos and started to build its own fuel cell stack.

In 1995, the project work for NYSERDA led to a 30-month, $2 million contract with DOE to develop and build a stand-alone, proof-of-concept 50 kW PEM fuel cell system. Power of this magnitude would be comparable to a small gasoline internal combustion engine and would be an important first step in demonstrating that PEM fuel cell stacks could produce sufficient power for electric vehicle applications. This project was followed by several smaller demonstration projects with NYSERDA to continue work on fuel cells. As a result of these programs, Dr. Ernst received the Partnership for a New Generation of Vehicles award as part of a government-industry team for demonstrating the first fuel cell system that was run on reformed gasoline. Other pertinent awardees were from Epyx (Nuvera), Los Alamos National Laboratory, and Argonne National Laboratory.

Other developments were concurrently unfolding in the energy industry to drive innovation in fuel cells. In particular, deregulation of energy markets was about to hit the electric utility industry. Deregulation meant that energy markets that were previously restricted to one regulated monopolistic supplier would now be opened up to competition. However, only markets for power generation were opened up to competition; power distribution was still considered a natural monopoly in a centrally distributed world. In theory, utility companies could no longer pass on bad business decisions to customers for reimbursement but instead would be held accountable for these decisions due to increased competition. In practice, problems emerged in certain deregulated markets, especially in California . 31 During the summer of 2001, the prices paid by utilities in that state for power surged in the newly deregulated spot power generation market, while those same utilities were prevented from raising prices enough to cover costs in the still-regulated power distribution market. Once revenue shortfalls threatened the utilities' ability to borrow, the California governor intervened with a rescue plan-thereby demonstrating that electric power may be too much of a public good to ever be traded in a completely private market with little government intervention.

Another effect of deregulation was to motivate companies to think of alternative business models. In particular, deregulation allowed utilities to sell power outside of their traditional service areas. Certain executives at Detroit Edison, a public utility, saw deregulation as an opportunity to expand business opportunities and convinced their board to pursue alternative energy strategies. One executive in particular, Gary Mittleman, saw MTI and its fuel cell capabilities as an attractive new business venture. This led to DTE's investing in MTI and to the formation of Plug Power, LLC, with Gary Mittleman as its first CEO.

The new management team embarked on a serious reorganization of MTI. In 1997, Plug Power was formed as a separate limited liability joint venture spin-off company owned by DTE Energy and MTI. Plug Power's mission was to design, develop, and manufacture on-site electric power generation systems utilizing PEM fuel cells for stationary applications. 32 Later, in 2000, MTI further diversified its fuel cell technology, creating another company called MTI MicroFuel Cells to develop small fuel cells for the portable electronics market, e.g., laptops, cell phones, and PDAs.

2. Public-Private Partnerships and Their Impact on Accelerating Plug Power's Fuel Cell Development

Project managers at Plug Power understood the funding hurdle in acquiring the breadth and depth of R&D investment commitments needed to bring PEM fuel cells to market. The enormous expense, high technological risks, and long time horizons posed huge challenges for Plug Power. Federal funding opportunities that recognize that the development of new, high-payoff technologies is a risky proposition were particularly valuable. Through two cooperative agreements with ATP, Plug Power has been able to pursue next-generation PEM fuel cell technology and projects carrying more technical risk than private investors were willing to absorb. Dr. Wayne Huang, senior research engineer at Plug Power, commented that the research undertaken as part of the ATP project would have been set back by a "number of years" were it not for the awards. Private investors demanded that Plug Power spend a large share of its capital on the development of a manufacturing line for more proven, first-generation technology and on making incremental improvements on initial products. The amounts the government expends on R&D projects like this are 20 to 50 times smaller than the amounts needed to establish even the initial manufacturing infrastructure.

Under its first ATP award, Plug Power developed and demonstrated a PEM fuel cell stack running at above 150 °C using a novel high-temperature membrane. The advances in membrane electrode assemblies and fuel cell stacks solved the carbon monoxide poisoning problem, allowing for operation with 200 times higher than today's carbon monoxide tolerances. This singular breakthrough will enable Plug Power to produce simplified PEM fuel cell systems. Such systems incorporate simplified reformers, simplified thermal and water management, and simplified systems integration. Through a new ATP award, Plug Power is developing a simpler fuel processor to work with the high-temperature stack developed under the first cooperative agreement. Since the processor is a separate component from the fuel cell stack, this ATP award is a completely separate project from the high-temperature membrane project.

Through its acquisition of H Power, Plug Power has acquired the research outputs of a third ATP-funded research project. A small company focused on small PEM fuel cells for remote telecommunications applications, H Power used ATP support to engineer an integrated, simplified membrane electrode assembly for PEM-type stacks running on propane. Subsequent to this acquisition, Plug Power has introduced one of the first fuel cell products for the telecommunications market. The GenCoreTM5T is designed to provide extended-run backup power specifically for the telecommunications industry in the demanding outside plant market. This application is a direct hydrogen fuel cell enabling potential future products in the cable broadband and uninterruptible power supply industries. 33 Plug Power has additional ongoing work with DOD, NIST, NYSERDA, the Texas Railroad Commission, and other government agencies. In addition to performance of research, the outputs of these projects include consumer testing and development of standards and validation.

___________________
1Koppel (1999). The material presented by Koppel was supplemented by interviews with Plug Power employees as well as the expertise and experience of one of the authors of this paper, Gerald Ceasar.

2These employees are William D. Ernst, current Vice President and Chief Scientist; Glenn Eisman, former Chief Technology Officer; and Wayne Huang, former Director of Chemistry and Materials.

3Koppel (1999), p. 46.

4Koppel (1999), p. 49.

5Interestingly, James Huff was, in 1959, head of catalysts for the Allis-Chambers farm equipment company, which produced the first tractor to run on a fuel cell.

6Koppel (1999), p. 86.

7Ibid, p. 90. Dow Chemical and Ballard eventually collaborated on a joint venture, which is discussed more completely in section IV.

8Koppel (1999), pp. 114-15.

9Branscomb and Auerswald (2002). Early-stage technology development R&D is characterized as research that is more commercially advanced than basic research, but not yet at the point of product development.

10Decaux (2003).

11ATP (2003).

12Avista Labs is an independent company, of which Avista Corp. is a minority shareholder.

13Plug Power prospectus and company interviews, spring 2000.

14David Morgenthaler, cited in Branscombe, Morse, and Roberts (2000), p. 107.

15PRNewswire, Plug Power press release, March 16, 2000 .

16Interview with Wayne Huang, Plug Power, April 2000.

17Interview with Kim Zentz, President, Avista Labs, April 11, 2000 .

18In fuel cell architecture, there are two basic configurations for holding the membranes: the stack and the cartridge. In both designs, hydrogen must flow to one side of the membrane and air to the other, and multiple membranes are required. In stacks, the membranes are stacked together to achieve a particular output and power density. The membranes are usually separated by a complex set of precisely machined plates, made with minute channels that direct hydrogen, air, humidification, and cooling fluids to the membranes. This requires compressors, fans, and other balance-of-plant equipment. In the event of a failure of any one of the stack seals, equipment components, or membranes, the entire stack ceases to operate.

19In the Avista Labs cartridge system, the PEM membranes are housed within an inexpensive plastic cartridge that is air-cooled and self-humidifying. There are no pumps or compressors to fail. Hydrogen enters one side of the cartridge; air the other. The only moving part is a high-efficiency fan. If a cartridge fails, the system shunts around it and continues to operate as before. The system automatically bypasses the cartridge with the problem and continues to provide power to the load. Cartridge replacement takes only a few seconds and requires no tools.

20Lovins and Williams (1999).

21Avista Labs is the only fuel cell developer that uses hot-swappable, modular power cartridge architecture. This patented technology offers high reliability. In part this is because it requires less balance-of-plant (Avista fuel cells have only one moving part, a high-efficiency fan) and also because any service needs can be accomplished quickly and easily by simply swapping out one cartridge for another, while the unit continues to create reliable power.

22The Economist (2001).

23Appropriability refers to the ability of private firms to appropriate profits from their innovations.

24Diltz (2002).

25www.simmonsco-intl.com/files/IP%20Week%20-%20London.pdf.

26The Economist (2002).

27Schiller (2001).

28Berry (2000).

29Interview with Mark W. Davis and A. Hunter Fanney, Heat Transfer and Alternative Energy Systems Group, NIST.

30The Stirling is an external combustion engine and, in that respect, is similar to a steam engine. Fuel is not critical; it can run on anything that produces heat. It was invented in 1816 by Dr. Robert Stirling, a Scottish minister, and for many years competed with the steam engine. The program was the result of federal concerns over energy availability.

31According to a Federal Energy Regulatory Commission investigation of Western energy markets, the staff "concluded that an underlying supply-demand imbalance and flawed market design combined to make a fertile environment for market manipulation" (FERC 2003).

32Plug Power, Inc. (2003a).

33Plug Power, Inc. (2003b).

Go to next section or return to Table of Contents.

Date created: March 29, 2005
Last updated: August 3, 2005

ATP website comments: webmaster-atp@nist.gov / Technical ATP inquiries: InfoCoord.ATP@nist.gov.

NIST is an agency of the U.S. Commerce Department
Privacy policy / Security Notice / Accessibility Statement / Disclaimer / Freedom of Information Act (FOIA) /
No Fear Act Policy / NIST Information Quallity Standards / ExpectMore.gov (performance of federal programs)