Energy 31 (2006) 33–43

www.elsevier.com/locate/energy

Steps toward the hydrogen economy*

S.S. Penner
Center for Energy Research, UCSD, San Diego, La Jolla, CA 92093-0411, USA

Abstract
The hydrogen economy is defined as the industrial system in which one of the universal energy carriers is
hydrogen (the other is electricity) and hydrogen is oxidized to water that may be reused by applying an external
energy source for dissociation of water into its component elements hydrogen and oxygen. There are three
different primary energy-supply system classes which may be used to implement the hydrogen economy, namely,
fossil fuels (coal, petroleum, natural gas, and as yet largely unused supplies such as shale oil, oil from tar sands,
natural gas from geo-pressured locations, etc.), nuclear reactors including fission reactors and breeders or fusion
nuclear reactors over the very long term, and renewable energy sources (including hydroelectric power systems,
wind-energy systems, ocean thermal energy conversion systems, geothermal resources, and a host of direct solar
energy-conversion systems including biomass production, photovoltaic energy conversion, solar thermal systems,
etc.). Examination of present costs of hydrogen production by any of these means shows that the hydrogen
economy favored by people searching for a non-polluting gaseous or liquid energy carrier will not be developed
without new discoveries or innovations. Hydrogen may become an important market entry in a world with most of
the electricity generated in nuclear fission or breeder reactors when high-temperature waste heat is used to
dissociate water in chemical cycles or new inventions and innovations lead to low-cost hydrogen production by
applying as yet uneconomical renewable solar techniques that are suitable for large-scale production such as direct
water photolysis with suitably tailored band gaps on semiconductors or low-cost electricity supplies generated on
ocean-based platforms using temperature differences in the tropical seas.
q 2004 Published by Elsevier Ltd.

*
A preliminary version of this manuscript was discussed at the 2001 Porto Venere workshop on “Advances in Energy
Studies.” The current version was presented at a Panel Discussion of the 2003 International Meeting of the Society of
Automotive Engineers in Detroit, Michigan, on 03-05-03. A version of this article was also presented as an invited lecture at a
meeting of the Western States Section of the International Combustion Institute in San Diego, CA, on 03/25/02.
0360-5442/$ - see front matter q 2004 Published by Elsevier Ltd.
doi:10.1016/j.energy.2004.04.060
34 S.S. Penner / Energy 31 (2006) 33–43

1. Toward the hydrogen economy

1
Although it is the most common element in the universe, hydrogen is not found in pure form on our
planet. It is present in very large amounts combined with oxygen as water and combined with carbon and
other elements in fossil fuels and innumerable hydrocarbon compounds (about 0.9% by weight of the
earth’s surface is hydrogen, which makes it the ninth most abundant element). The idea of using
hydrogen as a fuel is very old and dates from its isolation by Henry Cavendish in 1766, to the beginnings
of fuel-cell developments by W.R. Grove in 1839, to current times. It appears in early science-fiction
writings (Jules Verne in the 1870s observed in ‘The Mysterious Island’ that “water will one day be
employed as fuel, that hydrogen and oxygen which constitute it, used singly or together, will furnish an
inexhaustible source of heat and light, of an intensity of which coal is not capable.I believe that when
the deposits of coal are exhausted, we shall heat and warm ourselves with water. Water will be the coal
of the future.”)
Since its isolation, hydrogen made by water dissociation has often been labeled as the uniquely
preferred energy carrier. The Reverend W. Cecil built a hydrogen ‘vacuum engine’ in 1820, Otto used a
fuel mixture containing 50% hydrogen during the 1870s, J.B.S. Haldane produced hydrogen by using
wind-generated electricity in 1923, Sikorski studied the use of hydrogen as a fuel for helicopters in 1938,
there were 10 good flights of the lighter than air zeppelin ‘Hindenburg’ before the 1937 disaster at
Lakehurst, and hydrogen as a liquid fuel has become a mainstay of space exploration. The costs of space
missions and military devices do, however, not relate to the hard realities of civilian market applications.
More recently (1977) the ‘Hydrogen Agreement’ at the International Energy Agency (IEA) was
implemented to advance hydrogen production, storage and use technologies through worldwide
collaboration.
The historical evolution of the world’s energy systems has been characterized by two distinct
requirements for new technologies, namely, lower or at least acceptable aggregate costs (now including
especially environmental acceptability and sustainability) for performing new and important tasks and
long time frames (typically 40 years or more to reach the 10% level for market penetration). In view of
the pervasive market value of low energy costs, it is unrealistic to develop scenarios with new
technologies before these technologies have been shown to provide competitive market costs over the
long term. With this guiding principle in mind, we restrict our discussion here to what we judge to be the
three most promising entries for near-term market penetration at acceptable costs, namely, water
splitting using nuclear heat, direct water splitting on semiconductors using solar insolation (water
photolysis or photocatalytic water splitting) and ocean-platform-based ammonia production from the
generation of electricity. Serious experimental and theoretical studies on all three of the chosen
hydrogen-production technologies began during the early 1970s. The ultimate hydrogen economy is
expected to have the following features: (i) low-cost production of hydrogen using either renewable or
inexhaustible (on the scale of the few billions of years which our planet is expected to remain habitable)
primary nuclear energy sources following the present era of using fossil fuels for hydrogen manufacture;
(ii) oxidation of hydrogen in such designs as fuel cells in fuel-cell-hybrid vehicles with very high

1
A brief hydrogen history: isolated by H. Cavendish from the reaction of iron with sulfuric acid in 1766, labeled ‘hydrogen’
(for ‘maker of water’) in 1781 by A.L. Lavoisier, identified as primarily a mixture of two molecules in 1929 by P. Harteck, D2
identified as the heavy component in 1931 by H. Urey, F.G. Brickwedde and G.M. Murphy, production of tritium by Lord
Rutherford et al from bombardment of deuterophosphoric acid with deuterons in 1935.
 S.S. Penner / Energy 31 (2006) 33–43 35

Fig. 1. Schematic of a hydrogen economy using waste heat from a fission reactor.

conversion efficiencies not capped by the Carnot limit as for conventional engine cycles; (iii) safe
operations because of rapid diffusion and in spite of the very wide flammability limits of hydrogen–air
mixtures and the associated ease of ignition and transition to detonations and explosions.
Our current industrial system relies on the use of fossil fuels as primary energy source. Based on
market selection of the lowest cost source, hydrogen was produced2 during 2001 mostly from the fossil
fuels natural gas (48% of the total), heavy oils and naphtha (30%) and coal (18%) and by electrolysis
(4%) as a by-product of chlorine production. Locations of low-cost electricity supplies provided minor
additions through electrolysis with electricity generated from fossil or nuclear fuels or at hydroelectric
power stations. The hydrogen-production procedures include steam reforming, partial oxidation, as well
as plasma pyrolysis, and generally lead to H2 at year 2000 costs of $5 to $15 per GJ of hydrogen energy,
i.e. costs of about three times or more of the cost of energy from natural gas. The largely fossil-fuel-
based economy is currently undergoing rapid changes because the need to minimize air pollution in our
urban centers has served to focus attention on the desirability of replacing gasoline- and diesel-powered
vehicles by fuel-cell hybrids with on-board reforming of fossil fuels. To the extent that this conversion
proves to be economically viable, an important step will have been taken in implementing the ideals
of the hydrogen economy. During the year 2001, world-wide hydrogen production amounted to about
5!1011 m3 under standard conditions and thus carried an energy of combustion equaling about 2% of
that supplied by fossil-fuel use. It is important to note that competitive production of electricity from
renewables (e.g. biomass, wind energy, photovoltaics, etc.) does not provide for low-cost hydrogen in
the transportation sector because of the high cost of hydrogen and oxygen production by electrolysis.
An important avenue for arriving at the hydrogen economy involves the use of nuclear energy as
the primary energy-supply source (see Fig. 1). The relatively small low-cost reserves of high-grade
uranium suggest that we must plan to use passively safe nuclear breeder reactors over the long term.
Both conventional fission and breeder reactors require cooling loops with peak temperatures around

2
See IAEA-TECDOC-1085, 03/15/02 on the Internet during 2002.
36 S.S. Penner / Energy 31 (2006) 33–43

1000–1200 K in the normal cooling cycles. Based on intensive studies first performed during the 1970s,
we are able to implement water-decomposition cycles that produce hydrogen in sequential reaction steps
and serve to implement a version of the hydrogen economy compatible with effective operation of
nuclear reactors as primary energy-supply systems. In our current economy, studies on water-splitting
cycles have not progressed to the point where hydrogen production has been demonstrated to be a
commercially preferred procedure for transportation applications. However, this deficiency may perhaps
be corrected if nuclear (breeder) reactors become the widely used primary energy-supply source in the
future.
To a hydrogen-economy idealist, the preceding observations represent heresy. The purist will not
accept the continued addition of carbon dioxide to the atmosphere as the result of fossil-fuel combustion
and will also not accept the requirements for very long storage of highly radioactive fission products or
the possible threat of facilitating nuclear proliferation. To the idealist, successful entry of the hydrogen
economy means economically viable hydrogen production using only renewable energy supplies. While
the goal of the hydrogen economy has rated very high in terms of the ecology leg of the three
components serving as the foundations of our industrialized society (E3 or energy–ecology–economy), it
continues to fail in terms of competitive costs and will continue to fail in comparison with primary
energy supplies derived from fossil-fuel combustion or nuclear fission and breeder reactors unless a new
and low-cost production technology can be developed for making hydrogen without undesirable side
effects.
In the following sections, we review briefly research efforts aimed at hydrogen production from
water-splitting cycles and from renewable energy sources. Of these, none has progressed to the point of
possible near-term commercial viability without internalizing environmental impacts of fossil-fuel use.
However, the complete list of research opportunities (including topics omitted here, such as wind-power
systems, photovoltaic power conversion, etc.) is sufficiently varied to lead us to hope for progress even
though the last 25 years of subsidized efforts have proven to be unproductive.

2. Cycles for low-temperature water splitting

Methodologies for direct water splitting have been on the drawing board for longer than a quarter
century. A vast number of low-temperature thermal cycles has been devised and, to some extent,
subjected to empirical implementation for water splitting using the sun or another heat source such as a
nuclear reactor as primary input power source. None of these has as yet entered the competitive
marketplace. The most promising methodologies are those that couple heat energy provided from
nuclear fission reactors to the hydrogen splitting cycle but even these are not likely to compete with
direct fossil-fuel utilization unless projected environmental charges are levied for carbon dioxide
additions to the atmosphere. An example of these cycles is provided by the Mark 1 cycle invented by
G. Debeni and C. Marchetti at Ispra in Italy around 1970 [1,2]. This cycle is especially compatible with
the outflow coolant temperature (w850 8C) in a gas-cooled nuclear reactor since the maximum required
cycle temperature is only 730 8C. The sequential chemical reaction steps for the Mark I cycle are
summarized in Fig. 2. The overall process is water splitting according to the process

H2 O/ H2 þ ð1=2ÞO2 :
 S.S. Penner / Energy 31 (2006) 33–43 37

Fig. 2. The MARK-1 cycle invented by G. Debeni and Marchetti around 1970.

Although the sequential reaction steps are easily performed, the uses of Hg and HBr constitute
significant potential hazards. A cycle that has been studied by German investigators [3] is shown in
Fig. 3 and two cycles currently under investigation in Japan [4] are shown in Fig. 4. Thirty years after
examination of many thousands of potential sequential water-splitting steps and some experimental
work, no commercially viable design has been devised for our fossil-fuel-dominated economic system.
Recent discussions of work of this type may be found in Refs. [3,4] but are mostly based on studies
performed during the nineteen seventies. Of special interest in this connection is the calcium–bromine–
iron cycle which has been under development in Japan since 1978 under the name UT-3 cycle with a
maximum temperature of only about 1000 K (see Fig. 4).
A crucial policy issue concerning the use of process heat from nuclear reactors to prepare hydrogen by
means of water splitting cycles is the following: Does the use of this approach guarantee the entry of the
hydrogen economy as a component of the growth of fission-energy-generated electricity? The current
answer to this question must be ‘not now but perhaps in the not too distant future.’ The answer hinges not

Fig. 3. The AGNES cycle studied in Germany.

38 S.S. Penner / Energy 31 (2006) 33–43

Fig. 4. Two cycles studied in Japan. The overall process for both cycles is H2 O/ H2 C ð1=2ÞO2 :

only on the relative costs of hydrogen production from natural gas as compared to the cost of hydrogen
from water splitting but more pertinently on the relative costs of hydrogen energy as compared with
energy from fossil fuels.
In May 2002, K. Schultz of General Atomics organized an extensive review of the current status of
water splitting using nuclear process heat. In subsequent personal communications, Schultz gave the
following cost estimates. For a reactor cost of $1120/kWe and the sulfur–iodine cycle operating at
$315/kWe, the cost of H2 becomes $1.30/kg of H2. The cost of hydrogen from NG using steam reforming
with CO2 release is approximated by H2-cost ($/kg)Z0.286!cost of NG (in $/MBtu)C0.15; thus, for
an NG cost of $4/MBtu, the cost of H2 becomes $1.30/kg of H2 and equals the estimated hydrogen cost
from a water-splitting cycle. The cost of NG has been around $2/106 Btu to $3/106 Btu if we disregard
large perturbations during the recent energy-supply crisis in California. It follows that an NG-cost
escalation by a factor of 2 to 1.3 is required before the nuclear cycle becomes competitive.3 The
preceding results are changed by the following: internalizing the value of CO2 emissions from NG
processing, recovery of values associated with graphite or oxygen production, etc. Costs lower than
those for a developed technology are generally required to introduce a new technology. It should be
noted that the costs specified in this paragraph suggest successful replacement of H2 production from NG
but not competitive introduction of H2 as a new source of energy replacing fossil fuels at competitive
costs.
The future costs of NG are difficult to estimate. If current optimism concerning the magnitude of
available NG supplies is justified, nuclear fuel use with process-heat diversion for H2 production may not
become competitive for very many years. The opposite scenario with NG deficiency and rapid price
escalation will obviously support commercial production of H2 in water-splitting cycles using nuclear
process heat.

3
NG costs reached about $6/106 Btu and more during the early part of 2003.
 S.S. Penner / Energy 31 (2006) 33–43 39

3. Direct water photolysis using a semiconductor

During the seventies of the last century, there was flurry of excitement about direct water photolysis
using an electrochemical cell (see Fig. 5) as first described [5] in 1972 for a TiO2 n-type semiconductor.
In brief, the system consisted of an n-type semiconducting TiO2 (rutile) electrode exposed to sunlight
and connected to a platinum electrode. The TiO2 was mounted on an indium plate that served as
electrode-contact material. When the TiO2 was exposed to light, electrons and holes (pC) were formed
(see Fig. 5) according to the process TiO2 C 2hn/ TiO2 C 2eKC 2pC; which led to the hydronium-ion
[H3OC(l)] and gaseous oxygen formation at the negative electrode according to the reactions

2pþ þ H2 OðlÞ/ ð1=2ÞO2 ðgÞ[ þ2Hþ ; 2Hþ þ 2H2 OðlÞ/ 2H3 Oþ ðlÞ:

The overall process is

2pþ þ 3H2 OðlÞ/ 2H3 Oþ ðlÞ þ ð1=2ÞO2 ðgÞ:

Fig. 5. Schematic of the electrochemical cell in which a TiO2 electrode is connected with a platinum electrode. The surface area
of the platinum-black electrode used was approximately 30 cm2. The symbol a refers to a TiO2 semiconductor mounted on an
indium plate, which serves as an electrode contact material. The TiO2 is exposed to light (hn). The symbol d describes an
external load across which the voltage was measured with a voltmeter V. The symbol c refers to a platinum-black electrode,
while b describes a suitable electrolyte. The following processes occur in the solid (s), gaseous (g) and liquid (l) phases:

the overall process is H2 OðlÞðC2hnÞ/ H2 ðgÞC ð1=2ÞO2 ðgÞ: Here, C2hn symbolizes absorption of 2 photons at wavelengths
where electrons and holes are produced in the semiconductor TiO2.
40 S.S. Penner / Energy 31 (2006) 33–43

The electrons pass through an external circuit and then produce hydrogen at the positive electrode
according to the reaction
2eK þ 2H3 Oþ ðlÞ/ H2 ðgÞ þ 2H2 OðlÞ:
The current densities were observed to become constant at different values for different levels of the
pH. For a 500 W xenon lamp, the observed, non-optimal quantum efficiency was much lower than
0.1 with a cell emf of 0.5 V. TiO2 is not an optimal semiconductor for water splitting because its
band gap between the valence and conduction bands is about 3 eV (corresponding to light wavelengths
l!4150 A) whereas electrochemical water splitting can be accomplished at 1.25 eV (l!10,000 A).
Thus, with an optimally chosen semiconductor, it was expected that the water-splitting efficiency could
be increased by about a factor of 4.55 for a 6000 K blackbody light source like the sun. A general
requirement for a semiconductor to split water on light exposure into both H2 and O2 is the following: the
conduction band potential must be more negative than that required for hydrogen evolution and the
valence band potential more positive than that required for oxygen evolution (Fig. 5).
Using a multi-band-gap photoelectrochemical system (PEC) in an arrangement otherwise similar to
that applied in 1972, an efficiency of 12.4% was claimed in a 1998 publication (4b). Capital costs and
O&M charges for a commercial prototype have not yet been specified. This line of investigation is
expected to be continued in the future.
A recently published study by Khan, Al-Shahry and Ingler [6] describes a process for exposing Ti to a
natural gas flame at about 850 8C, which yielded n-type TiO2 containing some carbon substitution for
oxygen in TiO2. This material absorbs sunlight below 535 nm, i.e. over a substantially wider wavelength
range than the wavelengths below 400 nm which are absorbed by pure TiO2 since the band-gap energy is
reduced from 3.00 to 2.32 eV. As the result, the modified TiO2 semiconductor yielded solar-energy-to-
hydrogen conversion (photoconversion) efficiencies of about 8.35% for illumination at 40 mW/cm2.
The partially carbon-substituted TiO2 was found to be stable over an observational period of 6 months.
Further work along the specified lines may well lead to photoconversion efficiencies above 10%, which
have been judged to be required from commercial development. Applications not directly related to
hydrogen production include the use of n-type TiO2 in wet solar cells and photodegradation of organic
pollutants under uv radiation with wavelengths less than 400 nm (see Refs. 8 to 12 cited in Ref. [6]).
Designs of improved photocatalysts with claimed water-splitting efficiencies up to 15% form the
subject of three publications (June 2003) in the International Journal of Hydrogen Energy (Volume 28,
pp. 615–23, 651–5, and 663–9).

4. Direct water photolysis in a homogeneous phase

A new approach to water splitting was published in August 2001 [7] and falls appropriately into the
category of direct water photolysis using photochemistry. The procedure is very different from that used
in the attempts to drive the water-splitting reaction by using two photons absorbed on a transition-metal
complex. With the proposed procedure, hydrogen is produced in solution by employing a mixed-valence
dirhodium compound which, on exposure to a photon flux, breaks two rhodium bonds of a core molecule
in the presence of a halogen trap to regenerate the original rhodium compound (catalyst), which may
thus be reused to react with more water and produce more hydrogen. For details referring to the
photochemical processes and the nature of the chemical rhodium complexes involved, reference should
 S.S. Penner / Energy 31 (2006) 33–43 41

be made to the original publication by Heyduk and Nocera [8]. A plot of the number of 10K4 mol of H2
produced as a function of time shows production of up to about 1.0!10K4 mol of H2 after about 5 h of
illumination using UV-to-visible light with wavelengths above 338 nm. Thereafter, the rate of hydrogen
production is reduced and about 13 h of exposure were required to reach 1.3!10K4 mol of hydrogen
output. Much work remains to be done before this type of approach merits a practically meaningful
appellation as a process for producing hydrogen from photons, not only because the efficiency of water
splitting is very low but also because the procedure produces an accumulation of halogen from a halic
acid that is required in the photochemical reaction.

5. Fuel production using ocean thermal energy conversion (OTEC)

Another solar processing procedure depends on OTEC and involves the use of this large resource
from the tropical oceans. A detailed description of OTEC has recently been published by W.H. Avery [9]
and contains an optimistic conclusion concerning sea-based production of methanol from coal on an
OTEC platform or production of hydrogen from water electrolysis followed by hydrogen reaction with
atmospheric nitrogen. Because of high hydrogen transmission costs for substantial distances, the use of
condensable fluids is preferred over sea-based production of gaseous compressed hydrogen. Since the
ultimate purpose of the hydrogen economy is the production of non-polluting fuels without carbon
dioxide addition to the atmosphere, the ammonia cycle is preferred. According to Avery [9], cost
reductions for ammonia below gasoline and diesel-fuel costs may be achieved with specified OTEC
systems after a relatively brief learning period requiring the construction of 18 40 MWe plants. The
direct use of ammonia in the transportation sector may be judged to be too hazardous and land-based
reprocessing of ammonia to hydrogen, followed by direct use of this fuel in fuel-cell systems, is likely to
be a preferred approach. It should be noted that OTEC development enjoyed support from the US
Department of Energy and from the French and Japanese governments to the extent of about $250
million until about 1995, when the development status was judged to be ready for entry by for-profit
concerns. Although this last step has not yet materialized, it is likely to occur with significant escalation
of fossil-fuel prices or with the passage of laws internalizing (i.e. charging consumers) the projected
environmental costs of continued fossil-fuel use. In summary, it is likely that OTEC production of
hydrogen-containing fuels will serve as one of several preferred approaches to commercial realization of
the hydrogen economy using only renewable energy sources.
It has been demonstrated in various parts of the world (the Caribbean, near Hawaii, off-shore Taiwan)
that OTEC systems facilitate the growth of mariculture by bringing nutrients from deep water layers to
support increased populations of fish near the sea surface. This added benefit should further reduce the
market-entry costs of OTEC-based fuel production. On the other hand, warming deep ocean waters near
the water surface will release dissolved CO2 and may lead to environmental charges for OTEC systems,
although not on the scale for fossil-fuel use.

6. Hydrogen utilization in automotive transportation

Widespread uses of hydrogen in the transportation sector clearly represent major challenges. Past
achievements include application of hydrogen in oxidation processes of various types for electricity
42 S.S. Penner / Energy 31 (2006) 33–43

generation, flight applications, home heating, etc. in spite of residual fears attributable to the
‘Hindenburg Syndrome’ that has many non-believers convinced that extensive use of hydrogen as a
working fluid is more hazardous than the parallel activity utilizing oil or gasoline or natural gas or
propane.
The design of the automotive industry in the future will inevitably be connected with the design and
configuration of low-cost fuel supplies. If hydrogen becomes readily available at low cost, we may
expect to see initially a supplementary fuel-supply system competing with the current gasoline-supply
system. Large trucks and buses are the most likely initial vehicle entries using compressed hydrogen in
fuel-cell-powered vehicles in the manner of the Ballard and Daimler–Chrysler prototypes of the late
nineties. These entries will have strong backing from environmental groups concerned about pollution
levels in urban areas, including especially residual particulate levels. A significant impediment to
continuation of current fuel-use patterns is the fear of impending long-term planetary damage caused by
escalating atmospheric CO2 concentrations.
Widespread entry of fuel-cell-powered vehicles, especially (fuel-cell)FC-B(battery) hybrids, into the
personal transportation sector will remain problematic. With the advent of (internal combustion engines
powered by gasoline) ICE-B hybrid vehicles, fuel-cell-powered vehicles are not likely to become
economically competitive unless on-board processing of conventional fuels such as gasoline becomes a
low-cost reality. Currently achievable on-board fuel storage in ICE-B hybrid vehicles corresponds to a
driving range of about 600 miles. We see no known procedure that could approach this desirable range
with gaseous hydrogen-storage systems on small vehicles. Our recent review [10] of the status of fuel-
cell-powered vehicles defines important remaining challenges that must be met for these systems in
personal-use automobiles with on-board reprocessing of conventional fuels. Only a mandate limiting the
use of fossil fuels is likely to bring personal transportation vehicles using hydrogen in fuel cells to the
market in large numbers during the next decade.
Major topics for discussion at the year 2003 International Congress of the Society of Automotive
Engineers in Detroit included discussions of hydrogen as the fuel for fuel-cell applications in
transportation vehicles. The technical difficulties for on-board reforming of gasoline or other liquid fuels
have led to a notable decline in efforts to introduce fuel cells into the transportation sector without
carrying liquid or compressed hydrogen on board and thereby requiring the enormously costly
construction of a hydrogen distribution system as a component of conversion to hydrogen-powered fuel
cells. Electrode sensitivity to trace amounts (parts per million) of CO has been well recognized for some
years and led German and Japanese developers to emphasize the direct use of hydrogen as the preferred
vehicular fuel.
Even with the selection of hydrogen as the fuel of choice and acceptance of the need to construct a
country-wide hydrogen-distribution system, formidable challenges remain, especially with regards to
vehicle costs. The final vehicular configurations will generally be fuel-cell-battery-hybrids and perhaps
fuel-cell (FC)-internal-combustion-engine (ICE)-battery (B)-hybrids. These configurations are likely to
be far more costly to construct than the recently introduced ICE-B-hybrids while yielding only relatively
small increases in the achievable miles for equivalent energy consumptions. ICE-B-hybrids yielding 40
to 60 miles/gallon of gasoline are now commercially available in good-sized cars from a number of
vendors. Replacement of 70% of the current generation of US vehicles by these ICE-B-hybrids over a
time period of 7 years will reduce US fuel consumption by about 30% in the transportation sector and
total greenhouse-gas emissions by about 9%, thereby practically meeting the originally required
greenhouse-gas-emission reductions for the US by the year 2010.
 S.S. Penner / Energy 31 (2006) 33–43 43

The long-term goals of a hydrogen economy will probably be best achieved through resolute
introduction of hydrogen fuel cells into the stationary power sector while deferring both the construction
of a country-wide hydrogen distribution system and reconfiguring of the transportation sector for about a
decade. With the experience gained from extensive use of hydrogen-powered FCs in the stationary
power sector, re-examination of the transportation sector in about 10 years should lead to a much clearer
picture on how to proceed than can be reached today.
There are inventions on the horizon (see Sec. V) which may invalidate a current conventional
assessment of future automotive markets. As an example, at-sea processes on OTEC platforms to
manufacture ammonia, followed by transmission of compressed ammonia to coastal locations and then
direct use of ammonia at very high efficiency in newly designed fuel-cell-powered vehicles may lead to
low-cost market entry of exceedingly efficient transportation systems using ammonia in fuel cells, in
spite of potential environmental hazards associated with ammonia leakage.

References

[1] Debeni G, Marchetti C. Hydrogen, key to the energy market. Eurospectra 1970;9:46–50.
[2] Marchetti G. Hydrogen and energy. Chem Economies Engng Rev (Japan) 1973;5:7–25.
[3] Walter R, Rauhut B, editors. Horizonte, die Rheinisch-Westfälische Technische Hochschule anf dem Weg ins 21.
Jahrhundert: Springer; 1999. 789 pp.
[4] Ohta T, Yürüm Y, editors. Hydrogen energy systems. Amsterdam: Kluwer Academic Publishers; 1995, p. 69–82.
[5] Fujishima A, Honda K. Electrochemical photolysis of water at a semiconductor electrode. Nature 1972;238:37–8.
[6] Khan SUM, Al-Shahry M, Ingler Jr. WB. Efficient photochemical water splitting by a chemically modified n-TiO2.
Science 2002;297:2243–5.
[7] Khaselev O, Turner JA. A monolithic photovoltaic-photochemical device for hydrogen production via water-splitting.
Science 1998;280:425–7.
[8] Heyduk AF, Nocera DC. Hydrogen produced from hydrohalic acid solutions by a two-electron mixed-valence
photocatalyst. Science 2001;293:1639–42.
[9] Avery WH. Ocean—thermal energy conversion. In: Encyclopedia of physical science and technology, vol. 11. Academic
Press; 2002, p. 123–60.
[10] Penner SS. Fuel cells for transportation vehicles. Proceedings of the International Workshop on Advances in Energy
Studies: Exploring Supplies, Constraints and Strategies, Servizi Grafici Editoriali, Via Lagrange 3-35143, Padova, Italy;
2001. p. 25–8.