Third, the presence of CO and C0 2 in the input fuel stream poses significant problems for the fuel
cell stack and removing these gases is relatively difficult. The result is that system efficiency and
specific power and specific energy will be reduced so that the net fuel efficiency of the vehicle
may not be much better than would be achieved with a diesel engine. The use of hydrogen derived
from methanol reduces stack efficiency by 4 to 5 percent, and balance of plant efficiency could be
reduced by another few percent. Simulations by Argonne National Lab suggest that a realistic
system efficiency range for a methanol-based fuel cell is 38 to 47 percent at full load,^103
substantially under the 60 percent often quoted for the fuel cell. Part load efficiency could be
higher or lower and is dependent on system design and “balance-of-plant” efficiency at different
load factors. For systems using partial oxidation reformers and burning diesel or gasoline, overall
system average cycle efficiencies could be less than 40 percent.^104
Given the fact that the PEM fuel cell is just emerging from the basic research stage, it is
difficult to estimate costs of a commercial model, as cost could vary greatly depending on the
success in reducing platinum loadings; developing lower-cost membranes; reducing the size and
cost of methanol reformers, or developing low-cost, high-energy-density onboard hydrogen
storage; shrinking fuel cell “balance of plant;” and other R&D needs.^105 Researchers at Los
Alamos National Laboratory estimated that current designs could cost $1,800/kW (manufacturer’s
cost) in volume production, but their most optimistic projection with future technology
improvements was $40/kW (without methanol reformer).^106 GM/Allison has estimated that a total
system cost of fuel cell and reformer could be $65/kW in volume production,^107 and some
industry analysts hope to reduce costs still further. Some PEM cell manufacturers, however,
suggest costs could come down by a factor of 5 (i.e. to $400/kW for the fuel cell system without
hydrogen storage or methanol reformer).^108 Box 3-2 presents some basic arguments presented by
fuel cell advocates in favor of their conclusion that fuel cell costs can be reduced to levels that will
be competitive with internal combustion engines.
It is difficult to evaluate these cost estimates, because even those that present detailed costs for
individual components cannot describe how the fuel cells will be manufactured and end up
basically guessing what cell manufacture will cost; further, the bases for the component costs
generally are unclear. Some of the estimates of low costs appear to be based on relatively rapid
progress in achieving early cost and size reductions, but high rates of progress at this early stage
of development are not unusual, nor do they guarantee continuation of this rate of progress. The
rate of progress made by the Japanese in utility scale fuel cells, backed with hundreds of millions
of dollars of research, probably should yield caution in assuming that attaining cost levels well
below $100/kW is likely. Consequently, in OTA’s view, the most optimistic estimates of future
fuel cell cost--fuel cells at well below $65/kW--may be possible, but they require a substantial
degree of good fortune in the R&D effort and are by no means inevitable.
(^103) R.
Kumar et al., “Modeling of Polymer Electrolyte Fuel Cell Systems, paper presented at the Automotive Technology Development
Contractors Coordination Meeting, U.S. Department of Energy, October 1993. 104
l05Allison Gas Turbine, see footnote 95.
l06C.Borroni-Bird, see footnote 100.
T. Springer et al., "PEM and Direct Methanol Fuel Cell R&D," paper presented at the Automotive Technology Development Contractors
Coordination Meeting, U.S. 107 Department of Energy, October 1994.
108 Allison Gas Turbine, see footnote95.
Ballardrepresentative, personalcommunication, October 1994.