Effectiveness of the United States Advanced Battery Consortium as a Government-Industry Partnership (1998)

The USABC began by developing objectives and goals at several levels. The goals and objectives were then used as the basis for the selection and management of projects. The objectives and goals, and their underlying assumptions and implications, are described below.

The USABC established four overarching objectives (Thorpe, 1997):

• to establish a capability for an advanced battery manufacturing industry in the United States
• to accelerate the market potential of EVs through joint research on the most promising advanced battery alternatives
• to develop electrical energy systems capable of providing EVs with ranges and performance levels competitive with petroleum-based vehicles
• to leverage external funding for high-risk, high-cost R&D on advanced batteries for Evs

Operational targets based on the CARB mandates were (1) the development of a midterm battery, with production of a pilot-plant prototype in 1994 and (2) the demonstration of a feasible design for a long-term battery in 1994. The USABC's initial intention was to advance existing development-level projects to pilot-plant status. Once in pilot-plant production, advanced batteries could begin to be introduced into the EV market. Leveraging was achieved by cost sharing (roughly 50/50) between DOE and industry. Industry costs were typically split among the three automakers and the USABC contractors. If industry decided to produce and commercialize advanced batteries, they would bear all the costs.

No known chemistry can be integrated with a practical battery structure that can store enough energy per unit volume to power an EV with the performance and range of an ICE vehicle. Substituting batteries for a gas tank on a volume-for-volume basis would be even more difficult. In the literal sense, the USABC's goal of developing an electrical energy system, excluding fuel cells, with range and performance levels that are fully competitive with ICE vehicles is doomed to failure on the basis of energy content and weight penalties. Preserving performance levels requires sacrificing vehicle range. The critical question is whether EVs will have a long enough range so that users do not feel restricted.

The USABC marketing data showed that 60 percent of new retail buyers desired a range of 100 to 125 miles under full load, with services such as heater, air conditioner, and lights (R. Davis, 1997). Only 30 percent of the buyers were interested in a vehicle with a range of 75 to 100 miles. The 125-mile range was based on a ''need'' of 50 miles per day, a "want" of 50 miles per day for contingencies, and the equivalent of 25 miles per day for accessories (e.g., a heater or air conditioner). The analysis did not provide an estimate of the percentage of customers that would actually purchase an EV. Also, the USABC data did not indicate the market share that could sustain the business.

The USABC market analysis revealed that customers would accept only a modest price increase (no more than a few thousand dollars) for an EV in comparison to an equivalent gasoline-fueled ICE vehicle. Thus the EV market is price-sensitive, which was borne out in the poor showing of lease-marketed vehicles in California in 1997. For example, GM has leased only a few hundred vehicles in Los Angeles, San Diego, Tucson, and Phoenix combined (Leclair, 1997.)

On the basis of the market analysis and experience with vehicle markets, the USABC set seven customer requirements for a sustainable market:

• reliable and safe operation under both routine and extreme conditions
• a range of more than 100 miles in actual customer use
• life-cycle costs comparable to those of an ICE vehicle
• equivalent performance in hill climbing to an ICE vehicle
• battery lifetime equal to the lifetime of the vehicle
• minimal routine maintenance
• operating capability in all climates

When the USABC was established, these market objectives could not be met by an EV powered by a commercially available battery. Furthermore, no extant battery technology had the obvious potential of meeting all of the customer and USABC expectations.

The customer use and acceptance defined by CARB as a basis for its quantitative ZEV goals were somewhat different from those defined by the USABC.

The CARB assumptions were more optimistic and were focused on a particular market niche. In addition to the environmental benefits of ZEVs, CARB identified numerous other benefits, such as daily home refueling, reduced emissions testing, and quiet performance.

CARB targeted a segment of the market where the vehicle would not have to compete with the ICE vehicle in all functions. CARB anticipated that most new car buyers would have access to more than one vehicle (e.g., the two-car family) and that for one of the vehicles the miles-per-day requirement and the need to take long trips would be greatly reduced or eliminated. A range of 100 miles per day would mean the EV could be driven more than 30,000 miles per year. CARB did not specifically account for the extra load on the battery for accessories.

Throughout most of Phase I, the USABC's development strategy was based on two levels of battery performance, "midterm goals" and "long-term goals," respectively (Table 3-1). The midterm goals were designed to meet "introductory EV market requirements assuming EV subsidies." This level of performance would not support a sustainable market in terms of the seven customer requirements listed above. The long-term battery goals define a level of technology capable of sustaining a competitive EV market. The midterm performance levels were originally regarded as challenging but attainable targets that would enable automakers to respond to the CARB mandates for the late 1990s. Technical accomplishments in pursuit of the midterm goals were not necessarily expected to be useful for meeting the long-term goals. In addition to the performance goals given in Table 3-1, secondary criteria relating to characteristics, such as self-discharge, maintenance, and abuse resistance, were also established.

The USABC developed a third set of performance goals for Phase II, so-called commercialization goals, which are more demanding than the midterm goals but less ambitious than the long-term goals. Commercialization goals (also shown in Table 3-1) define conditions that the USABC judged would allow the entry of EVs into the commercial market without subsidies. The commercialization goals have been the principal focus of the program since 1996.

The USABC battery performance goals, which have played a major role in the USABC's decision making, are based on an analysis conducted in 1994 comparing the operating costs of EVs and ICE vehicles (R. Davis, 1997). Table 3-2 shows the performance values assumed in the USABC analysis; it was also assumed that the cost of the electric propulsion motors and supporting electronics for an EV would be the same as the cost of the engine, transmission, and supporting systems for an ICE vehicle. No data supporting this assumption were presented to the committee. Table 3-3 shows the calculated costs.

The USABC's analysis shows that technologies that can truly compete on a cost-per-mile basis with an ICE vehicle will have to meet the long-term performance goals. But the analysis is very sensitive to the underlying assumptions. For example, the midterm goal for total battery cost, which is really "battery price," is lower by a factor of three to five than the price that was realized in the Phase I program. If the more realistic number of $600/kWh were used, the total midterm cost would be $0.25/mile, which is much higher than the cost with other batteries. Different assumptions about infrastructure and power costs would have changed

the total cost per mile significantly. Assumptions about battery life also affected the calculated costs. The assumptions anticipate dramatic cost and performance advances in some batteries but not others. For example, if lead-acid battery life exceeds 2.5 years (as some evidence suggests), the calculated cost per mile given in Table 3-3 would be significantly lower. Other assumptions, such as driving 50 miles per day for 200 working days per year, seem to be consistent with commuter expectations.

The committee was asked to comment on the processes used by the USABC to solicit proposals, choose contractors, and make awards, for both new and continuing projects. The committee attempted to determine whether the USABC's procurement procedures were designed to select organizations with the technology (including manufacturing capabilities) and motivation most likely to result in batteries that meet USABC goals. Despite substantial technical progress (see Chapter 4), the USABC has failed to meet all of its midterm or long-term goals. Thus, the committee considered it important to determine whether the procurement process was responsible in any way for the lack of overall success of the USABC program.

The USABC's procedures for selecting Phase I collaborative R&D projects involved public announcements of the consortium's initiative, the preparation and release of a request for proposal information (RFPI), and the evaluation of proposals. The USABC mailed out 130 RFPI packages and received 59 proposals based on 16 different battery technologies (Nichols, 1992). Eight awards were made for six technologies.¹

The RFPI is a detailed document that outlines the purpose and objectives of the USABC and reviews the prospective market for EVs. Primary and secondary technical criteria for midterm and long-term batteries are included, and two major program goals are highlighted: (1) the demonstration of production capability for the midterm battery in 1994 and (2) the demonstration of design feasibility for the long-term battery in 1994.

The RFPI spells out the expectations of the USABC for potential participants, including integrated team efforts to address well defined milestones and rapid progress toward prototypes because of the very short time scales set by CARB for the introduction of ZEVs in 1998. The specification of deliverables to be tested independently by DOE's national laboratories showed that this was an engineering, rather than a science, initiative. After issuing the RFPI, the USABC management urged interested parties to form partnerships that would complement their individual capabilities, a strategy the committee endorses.

Companies responding to the RFPI were required to provide information on company background; the proposed advanced battery technology; the technology development plan; schedule, deliverables, cost, and cost sharing; and cooperative arrangements. The requirement that participants share in the costs of the program all but eliminated contractors who did not have realistic expectations of achieving their goals.

The 59 proposals received in response to the RFPI were evaluated according to selection criteria based on the potential of the proposed technology to meet performance goals; the vision of the proposed development plan; the ownership of the proposed technology; the capability for high-volume manufacturing; cost sharing; the financial viability of the company/team; experience and past performance; the probability of successful commercialization; the qualifications of key personnel; the understanding of the project requirements; previous experience with high-power batteries; the ability to comply with battery packaging constraints; and the ability to meet required time schedules. Weighting factors were assigned to each category, with some values varying between midterm and long-term goals.

In the committee's judgment, some features of the RFPI might have put some potential contractors at an unnecessary disadvantage or discouraged the participation of some organizations. These features include the absence of specific guidelines for cost sharing; an apparent lack of incentives to develop midterm batteries; the requirement that technology be in hand; and the agreements for proprietary information and intellectual property. Nevertheless, the USABC's selection of projects was not biased. In the committee's opinion, the extensive knowledge of the status of battery technologies provided by DOE and other consortium participants was an important factor in the selection of appropriate proposals.

As Phase I proceeded, some technologies that were not sufficiently developed were eliminated by the consortium, generally by mutual agreement with the

developers, and selected development projects were discontinued by individual companies (see Chapter 4). When Phase I ended and Phase II began, the only remaining candidate that could meet long-term goals was lithium polymer battery technology. To stimulate alternatives and possible competing technologies, another RFPI was issued but did not elicit any responses the USABC decided were worth funding.

The original RFPI was sent to all known U.S. and European companies that were working on or had developed battery technology. Asian companies were not invited to participate in Phase I because an Asian battery consortium was already well established (Stroven, 1997), and Asian companies already have major science, technology, and manufacturing capabilities for batteries. The USABC's decision to exclude them appears to be at odds with two of the objectives of the USABC initiative, namely, (1) accelerating the market potential of EVs through joint research on the most promising battery technologies and (2) providing electrical energy systems capable of providing EVs with range and performance levels competitive with those of petroleum-based vehicles. But including Asian companies could have conflicted with another USABC objective, establishing a capability for a United States advanced battery manufacturing industry. Thus, the exclusion of Asian participants from Phase I reflects the USABC's higher priority on fostering U.S. competitiveness than on the benefits of Asian advances in battery technology.

The committee noted that the USABC did not exclude European manufacturers in Phase I, which appears to be inconsistent with its treatment of potential Asian partners. SAFT America, Inc. (the American arm of the French company, SAFT), VARTA (a German company) and the American office of Silent Power (which was owned by the German RWE group in 1993) all participated (see Chapter 4). However, it was not clear to the committee that members of the Asian battery consortium would have been interested in working within the USABC framework because their program was already well established, and they would probably not have been willing to reveal commercially sensitive information as required by the RFPI (see below). No responses to the Phase II RFPI were received from Asian companies, even though the USABC solicited their participation in an effort to diversify its portfolio of long-term battery technologies.

The responses to the Phase I RFPI addressed 16 battery technologies, although the lead-acid system was deliberately excluded from consideration, based, in part, on the fact that Chrysler, Ford, and GM were already working individually

with lead-acid battery manufacturers in a competitive environment. ² Collaborative USABC ventures are, by necessity, precompetitive. In addition, the USABC had some doubts about the potential of lead-acid batteries to meet USABC's performance criteria.

In retrospect, the committee concluded that the exclusion of lead-acid technology from the USABC portfolio had a positive effect on the development of battery technology. In 1992, the manufacturers of lead-acid batteries and their suppliers formed their own consortium, the Advanced Lead Acid Battery Consortium (ALABC), which has an applied research program aimed at making lead-acid technology competitive with the battery technologies being developed under the USABC initiative. The ALABC made significant technical progress between 1992 and 1997 with very little government funding, especially toward achieving the ALABC targets of specific energy, cycle life, recharge time, and purchase cost (Moseley, 1997).

A number of small businesses have participated in the USABC program. Ovonic Battery Company has developed a Ni/MH (nickel metal hydride) battery, Yardney Technical Products has developed a low-cost nickel electrode, and EIC Laboratories was a subcontractor to W.R. Grace on the development of a lithium-ion-polymer battery. Other small businesses have supplied or manufactured specialty items to USABC developers (Sutula, 1997).

Nevertheless, the committee believes that, despite commendable efforts by USABC management to foster partnership agreements, some features of the procurement process may have put small companies at a disadvantage in bidding for USABC contracts. For example, the absence of specific, published guidelines for cost sharing could have discouraged the participation of small companies that did not have the resources to assume major costs. Even though in practice the amount of cost sharing was negotiable, the time pressures to get the program under way may have discouraged lengthy negotiations. As a result, the participation of some small, less financially robust companies may have been discouraged. In addition, the scoring process for proposals was heavily weighted in favor of organizations with large-scale manufacturing capabilities and demonstrated financial viability. Thus, small, relatively new organizations with promising technologies did not have much chance of participating.

The committee recognizes that the circumstances of the USABC initiative, including the very aggressive time frame for commercialization, might have precluded the participation of small companies that were unable (or unwilling) to

form partnerships with larger organizations.^3,4 However, the committee believes that participation by more small companies—particularly companies working to develop battery technology relevant to EVs—would have been useful. Historical records show that most "disruptive" technologies were begun by small companies or companies that were not in the mainstream (see, for example, Christensen, 1997). The limited markets for EVs could be attractive to smaller companies, although these markets would not be big enough to meet the commercialization requirements of the 1990 CARB mandate. Also, smaller companies could increase the market penetration, thereby helping to amortize the costs of R&D and the establishment of production facilities.

The committee questioned whether the cost-shared development of midterm batteries would be viewed by many companies as a commercially viable endeavor. Midterm batteries were not expected to capture a sustainable share of the market for EV batteries or even to contribute to the development of long-term systems, and it is not clear when midterm batteries would be displaced by long-term batteries in vehicles. (See chapters 4 and 6 for a discussion of whether midterm battery technologies would be able, after further development, to meet the long-term requirements.) Thus, the opportunities to amortize the development and commercialization costs of a midterm battery may be very limited. If the sales window for EVs with midterm batteries was very short, or was eliminated or delayed (as actually happened), contractors would have little commercial incentive to commit to developing a midterm battery.

Some companies may not have been interested in participating because the Phase I RFPIs were not confidential, despite the fact that the information requested was generally considered proprietary. The RFPI stated that "notwithstanding proposer's markings to the contrary, all information submitted in response to a consortium RFPI shall be treated on a non-confidential basis" (USABC, 1991). Companies with relatively mature battery technologies, which had been developed at considerable expense over many years, were understandably concerned about revealing proprietary information. Negotiations to handle potential problems could have been arranged but would have delayed the project selection process, which was already under severe time pressure.

From the point of view of an industrial supplier, the terms listed in Section VII (Purchase Order Terms and Conditions) of the RFPI would seem restrictive.

The seller must commit resources through cost sharing and then turn over all intellectual assets to the buyer. The seller has no guarantee that it will receive a contract to manufacture the system it has developed.⁵ (In practice, contracts and agreements were negotiated to provide better incentives for participation. The language in the RFPI should be reconsidered by any follow-on program to the USABC.) The seller's freedom to enter other markets does not fully remove this concern, because a battery that had been optimized for one purpose could probably not be used in another application without extensive and costly redesign.

The committee wishes to make two observations about the development of technology and the development of new science in the context of the USABC initiative. First, the committee acknowledges that the USABC program in Phases I and II was necessarily a technology development program with little room for new science. This was an inevitable consequence of the short CARB timetable for the production of marketable vehicles. Second, the committee strongly believes that new science is essential for the long-term health of the battery industry in the United States and probably is essential for the successful development of practical power systems for electric and hybrid vehicles. The committee does not fault the USABC for not emphasizing new science in its program but believes (1) that DOE should preserve its battery-related scientific research programs in addition to participating in the USABC and (2) that any successor program to the USABC should include relevant new scientific research.

The expectations described in the RFPI clearly indicate that the USABC program is an engineering, rather than a science, initiative, although the proposal evaluation criteria—if applied rigorously—are less definitive. Although the capability of the proposed technology to meet the applicable criteria was an important factor in the decision-making process, the USABC also gave considerable weight to other factors, such as ownership of the proposed technology, high-volume manufacturing process capability, cost sharing, and the financial viability of the proposer's company/team. A large corporation with high-volume manufacturing capabilities and a willingness to share costs with a strong patent could be given a very positive evaluation, even though the patent might cover an impractical or long-range battery technology. In practice, the project selection process was not biased, suggesting that the background knowledge of battery development supplied by DOE and others was a valuable complement to the quantitative proposal evaluation criteria.

Because of the very ambitious USABC timetable that required production of a midterm battery in less than three years from the date of the RFPI, the committee assumed that only proposals from companies that had significant technology in hand would receive serious consideration. Battery systems for which there was a scientific base but no supporting technology appeared to be doomed to failure because of the short time frame. With one exception, the committee found that the project selection process adopted by the USABC did focus on battery systems for which technology already existed. Somewhat surprisingly, however, the USABC funded W.R. Grace to develop a lithium-ion battery with a polymer electrolyte, a new system that had not reached the prototype phase of development (see further discussion in Chapter 4).

Thus, on the one hand, the technical challenge was great, and new science was not likely to result in an acceptable product within the time constraints. On the other hand, because the performance and cost requirements were so demanding, the contributions of new science could have been substantial (if the time schedules had been less stringent). If a new battery initiative is being contemplated, the committee believes it should have a better balance between new science and technology development.

Finding 1. The original objectives of the USABC program led to the establishment of goals that represented great challenges to existing battery technologies. When the targets of pilot-plant production of a midterm battery and demonstration of design feasibility for a long-term battery by 1994 were not met, commercialization goals, which are less demanding than the long-term goals, became the principal focus of Phase II of the program.

Finding 2. The assumption that an EV that fulfills the long-term goals will be fully competitive with an ICE vehicle is not realistic. In range, performance, and recharging characteristics, a battery-run ZEV cannot match an ICE vehicle.

Finding 3. Analyses of the comparative costs of EVs and ICE vehicles—such as the analysis conducted by the USABC—are very sensitive to assumptions about infrastructure and power costs. Thus, achieving the long-term goals may not result in commercially viable EVs that are competitive with ICE vehicles.

Finding 4. Battery systems developed to meet the midterm goals were not expected to gain a sustainable share of the market for EV batteries or even necessarily to contribute to the development of battery systems to meet the long-term goals. The principal purpose of developing midterm batteries was to meet the original CARB requirements for ZEV market share in the late 1990s and establish a market share for EVs. Although this marketing objective has been changed, the development of midterm batteries has been useful in providing power sources for the development of EV technology.

Finding 5. The criteria used by the USABC to evaluate proposals and award contracts were heavily weighted in favor of large organizations with high-volume manufacturing capabilities. The criteria also reflected an aggressive schedule for the commercialization of EV batteries, which left little room for the development of new science.

Finding 6. Small companies with limited resources were at a disadvantage because of the requirements in the RFPI for cost sharing, demonstrated financial viability, and high-volume manufacturing capabilities. The USABC's encouragement of partnerships among interested parties did help to counteract this disadvantage, although promising innovative technologies under development by small companies may still not have been given adequate consideration. As a result, small companies, including those working to develop battery technology relevant to EVs, have not been active participants in the USABC program, although DOE has created opportunities for small businesses to pursue R&D on batteries through the Small Business Innovative Research program.

Finding 7. Decisions to invite or exclude Asian participation in the USABC program have been inconsistent, but the committee could not determine whether there has been—or is likely to be—any resulting impact on battery development.

Finding 8. The decision by the USABC to exclude lead-acid battery technology from its program had a positive impact on the development of this technology. The Advanced Lead Acid Battery Consortium undertook its own applied research program, which has resulted in significant improvements in lead-acid battery technology.