Solar Energy 78 (2005) 661–669

www.elsevier.com/locate/solener

PEM electrolysis for production of hydrogen from

renewable energy sources
Frano Barbir *

Connecticut Global Fuel Cell Center, University of Connecticut, 44 Weaver Road, Unit 5233 Storrs, CT 06269-5233, United States

Received 2 June 2003; received in revised form 30 August 2004; accepted 1 September 2004
Available online 19 October 2004
Communicated by: Associate Editor S.A. Sherif

Abstract

PEM electrolysis is a viable alternative for generation of hydrogen from renewable energy sources. Several possible
applications are discussed, including grid independent and grid assisted hydrogen generation, use of an electrolyzer for
peak shaving, and integrated systems both grid connected and grid independent where electrolytically generated hydro-
gen is stored and then via fuel cell converted back to electricity when needed. Specific issues regarding the use of PEM
electrolyzer in the renewable energy systems are addressed, such as sizing of electrolyzer, intermittent operation, output
pressure, oxygen generation, water consumption and efficiency.

2004 Elsevier Ltd. All rights reserved.

Keywords: PEM electrolysis; Hydrogen; Renewable energy

1. Introduction as safe as other common fuels (Swain and Swain,

1992; Thomas, 1996; Barbir, 1999a). However, hydrogen
Hydrogen may play an important role as an energy is not an energy source. It does not occur in nature in its
carrier of the future (Veziroglu and Barbir, 1991, elemental or molecular form. Therefore, hydrogen must
1998). Hydrogen may be used as fuel in almost every be produced. The most abundant source of hydrogen is
application where fossil fuels are being used today, but water, but water splitting requires energy, and because
without harmful emissions (Barbir and Veziroglu, of the laws of thermodynamics, energy required to split
1992), with a sole exemption of NOx emissions when water is higher than energy that can be released from
hydrogen is combusted (which, however, can be effec- produced hydrogen. Because of that, hydrogen is like
tively controlled). In addition, hydrogen may be con- electricity—an energy carrier, a convenient form of en-
verted into useful forms of energy more efficiently than ergy. From sustainability point of view, a synergy be-
fossil fuels. And despite public perception, hydrogen is tween hydrogen and electricity and renewable energy
sources is particularly interesting.
Although several methods have been and are being
developed for production of hydrogen from renewable
*
Tel.: +1 860 486 6703; fax: +1 860 486 8378. energy sources, the only one currently practical is water
E-mail address: [email protected] electrolysis. Water electrolysis is a mature technology,

0038-092X/$ - see front matter
2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.solener.2004.09.003
662 F. Barbir / Solar Energy 78 (2005) 661–669

and it is being used for hydrogen production capacities (1.482 V). Protons pass through the polymer electrolyte
ranging from few cm3/min to thousands m3/h. It is rela- membrane and on the cathode combine with electrons
tively efficient (>70%), but because it needs high quality to form hydrogen. Passage of protons through the mem-
energy (electricity) hydrogen produced by water-elec- brane is accompanied by water transport (electroos-
trolysis is expensive (>$20/GJ). There is a potential to motic drag).
generate relatively inexpensive hydrogen from hydro- A PEM electrolyzer cell is similar to a PEM fuel
power and nuclear plants during off-peak hours. Justi cell. It has a polymer membrane and porous electrodes,
(1987) proposed large PV power plants in North Africa flow fields, current collectors and separator plates, end
to generate hydrogen which could be shipped or piped to plates, bus plates, manifolds. The principle of opera-
Europe. Ogden and Williams (1989) evaluated solar tion is just reverse of fuel cell operation. However,
(PV) hydrogen option and concluded that hydrogen the materials are typically different from PEM fuel cell
could be economically produced from large PV power materials. Carbon materials, such as catalyst support,
plants if the cost of PVs could be brought to $0.2 to porous electrode structures (carbon fiber paper or car-
$0.4/Wp. Coluccia et al. (1994) analyzed option of pro- bon cloth) and bi-polar plates, commonly used in fuel
ducing hydrogen from hydropower in Zaire and ship- cells cannot be used on the oxygen side of a PEM elec-
ping it via pipeline to Southern Europe. A trolyzer due to corrosion. PEM electrolyzers therefore
Hydro-Quebec Hydro-Hydrogen Pilot Project seriously primarily use metallic components (porous structures,
considered production of hydrogen from hydropower flow fields and separator plates). The catalyst is typi-
in Canada and shipping it to Northern Europe via cally platinum or platinum alloys. Similarly to fuel
tankers (Gretz et al., 1990; Drolet et al., 1996). cells, individual electrolyzer cells may be stacked into
a stack, in order to get the desired output at a reason-
able stack voltage.
2. PEM electrolysis Besides the cell-stack, an electrolyzer must have a
power supply/voltage regulator, water supply system,
Most of the electrolyzers used today in capacities up water circulation pump, water–gas separators for hydro-
to several thousand m3/h are based on alkaline (KOH) gen and (optionally) oxygen, heat exchanger, controls
electrolyte. Another option is to use a proton exchange and instrumentation, including the safety features.
membrane as electrolyte. This perfluorosulfonic acid There are only a few PEM electrolyzer manufacturers
polymer (also known as NafionTM) has been used in today. Fig. 2 shows industrial PEM electrolyzer HO-
chlor-alkaline electrolysis and also in fuel cells. PEM GEN 40 by Proton Energy Systems. This electrolyzer
electrolysis is a process just reverse of a PEM fuel cell is intended for industrial hydrogen generation. A version
process (Fig. 1). Water is split into oxygen, protons of it, so called HOGENRE, has been developed for use
and electrons on one electrode (anode) by applying a in conjunction with renewable energy sources. Table 1
DC voltage higher than a thermoneutral voltage shows HOGENRE specifications.

Fig. 1. Schematic representation of PEM electrolysis and fuel cell processes.

 F. Barbir / Solar Energy 78 (2005) 661–669 663

Fig. 2. HOGEN PEM Electrolyzer by Proton Energy Systems (stack on the left and complete packaged system on the right).

Table 1
Specifications of HOGENRE PEM electrolyzer by Proton Energy Systems
Hydrogen output 0.5 or 1.0 N m3/h
Max delivery pressure 200 bar
Hydrogen purity >99.9% (optionally >99.999%)
Water usage 0.5 or 1.0 l/h
Water quality (min) required deionized (ASTM Type II)
Power consumption 6.6 kW h/N m3
Electrical supply required AC: 190–240 VAC, 1 phase, 50/60 Hz, 7.2 or 12 kV A DC: 60–200 VDC, 150 A (max)
Operating environment Indoor (optionally outdoor)
Dimensions 97 · 105 · 106 cm
Weight 220 kg
Installation ‘‘Plug & play’’
Controls and automation Fully automatic and unattended

3. Applications This system circumvents the problems related to the grid

connection and generation imbalance charges. The elec-
Hydrogen may be produced from PV generated elec- trolyzer will be exposed to the variable power supply,
tricity in a variety of applications, and used as a fuel di- and a DC/DC power regulator must be a part of the
rectly, or transmitted through pipelines to the users, or power conditioning and controls box in Fig. 3, in order
used to enhance the performance of the PVs by match- to match the electrolyzerÕs voltage–current requirements
ing their output to the user needs. Each of these applica- at any power with minimum conversion losses. Another
tions is briefly discussed below. option would be to select the PV panels so that their i–V
output matches the electrolyzerÕs i–V polarization curve
3.1. Grid independent solar-hydrogen generation (Spinadel et al., 2002).
If the PV installation is in remote area (and most of
Instead of connecting to the grid, a PV array may be well insolated sites are indeed in remote areas) grid may
connected to an electrolyzer to produce hydrogen, which not be available at all. In that case delivery of hydrogen
then may be used in a variety of applications (Fig. 3). through a pipeline over a long distance may be an

Fig. 3. Schematic diagram of grid independent PV-hydrogen generation.

664 F. Barbir / Solar Energy 78 (2005) 661–669

option. Some studies suggest that transmission of hydro- grid operator. In that case the excess power from the
gen through a pipeline in some cases may be more eco- PVs (above that can be absorbed by the grid in that par-
nomical than transmission of electricity over a long ticular moment) is used to generate hydrogen (Fig. 5).
distance (Justi, 1987; Coluccia et al., 1994; Oney et al., However, the problems with intermittent electrolyzer
1994; Keith and Leighty, 2002). Recently, there was a operation and its low capacity factor remain. A varia-
proposal, accompanied by a techno-economic study, to tion of this scheme is to operate the electrolyzer at con-
ship megawatts of wind power through hydrogen pipe- stant power and ‘‘dump’’ the excess power from the PV
lines from the wind farms in the Dakotas to the users power plant into the grid (off course providing there is
in Chicago—distance of some 1600 km (Keith and enough absorbing capacity in the grid).
Leighty, 2002).
3.4. Integrated PV-hydrogen utility energy system
3.2. Grid-assisted PV-hydrogen generation
One way to deliver a constant or any required load
If grid is available, one way to eliminate problems profile to the grid is to equip the PV array or PV power
with intermittent electrolyzer operation is to combine plant with an energy storage device, such as a regenera-
the PVs with an input from the grid (Fig. 4). The power tive fuel cell (a combination of electrolyzer and a fuel cell
conditioning/controls unit facilitates that the electro- with hydrogen storage), as shown in Fig. 6. The electro-
lyzer receives constant power DC input, by combining lyzer and fuel cell functions may be included in a single
the output from the PVs with the required input from stack (unitized version) or in two separate stacks (dis-
the grid. This way the electrolyzer may operate all the crete version). In order to fill the hydrogen tank, a com-
time at its design point. The capacity factor may reach pressor may be required or the electrolyzer may be
high 90s (the only down-time would be for mainte- designed to provide high pressure. Proton Energy Sys-
nance), which would significantly improve economics, tems reported operating pressures with PEM electrolyz-
i.e., reduce the cost of hydrogen. ers up to 200 bar (Anderson et al., 2002; Barbir et al.,
2002). The power conditioning and controls unit has
3.3. PVs for grid-electricity and hydrogen generation an extremely complex function in this configuration. It
must direct power from the PVs to either the grid or
An electrolyzer may be used to ‘‘peak shave’’ the out- the electrolyzer, switch to fuel cell power when there is
put from the PV power plant if that is required from the not enough power from the wind turbine, and provide

Fig. 4. Schematic diagram of grid-assisted PV-hydrogen generation.

Fig. 5. Schematic diagram of PV installation for grid-electricity and hydrogen generation.

 F. Barbir / Solar Energy 78 (2005) 661–669 665

Fig. 6. Schematic diagram of integrated PV-hydrogen utility energy system.

voltage regulation, both from the PVs to the electrolyzer attractive option in a remote location where fuel delivery
and from the fuel cell to the grid. The renewable fuel cell is not regular).
system is typically less costly than a battery bank for Similar systems have already been demonstrated in
high power/long duration storage, approximately above USA, Germany, Italy, Finland, Switzerland, Spain and
5 kW h (Barbir et al., 2002). Due to presently high costs South Korea (Dienhart and Siegel, 1992; Rosa et al.,
of electrolyzers and fuel cells and a low capacity factor 1994; Galli et al., 1994; Collier, 1994; Szyszka, 1996;
in this configuration, the cost of delivered electricity Oud and Steeb, 1996; Lehman et al., 1997; Vanhanen
would be several times higher than the cost of PV gener- et al., 1998; Hollmuller et al., 1998). The power output
ated electricity (Barbir, 1999b), even when the genera- of existing demonstration projects varies from 1 kW to
tion imbalance charges are taken into account. 350 kW.
Therefore, this configuration may be used only where This system may also incorporate a wind turbine for
the quality of delivered electricity justifies the cost. added security and versatility of power supply. Com-
bined wind and solar (PV) systems are in operation at
3.5. Grid independent integrated PV-hydrogen energy the Desert Research Institute (Reno, Nevada), and at
system the Hydrogen Research Institute (HRI), at the Univer-
site du Quebec, Trois Rivieres (Agbossou et al., 2001).
A similar system may be used for autonomous, grid These independently conceived systems use both photo-
independent power supply (Fig. 7). In this case, energy voltaics and wind turbines to generate hydrogen fuel.
storage is an imperative, and the regenerative hydrogen They are relatively small systems using wind/solar
fuel cell may be competitive with other energy storage hydrogen generation with a compressed hydrogen stor-
options (batteries, flywheel, compressed air, pumped hy- age system, 5 kW electrolyzers, and incorporating a fuel
dro, ultracapacitors, etc.). The only difference between cell which regenerates the electricity. Both use a combi-
this and the previously described system is in the power nation of battery and hydrogen energy storage, in which
conditioning and control unit, which in this case does a small battery bank acts as an energy buffer, and the
not have to synchronize with the grid, but it has to pro- bulk of the energy storage is in the form of hydrogen.
vide desired AC or DC voltage outputs. One option is to A system consisting of an electrolyzer, fuel cell, and
use PV-generated hydrogen as a fuel for cooking and hydrogen and oxygen tanks has been proposed for a high
heating in the house and/or for a fuel cell or hydrogen altitude unmanned aircraft (Mitlitsky et al., 1993). Such
combustion engine powered vehicle (this may be an an airplane would be powered by electric propellers using

Fig. 7. Schematic diagram of grid independent integrated PV-hydrogen energy system.

666 F. Barbir / Solar Energy 78 (2005) 661–669

PVs placed on the top of the wings. Excess power during power output. The problem is that at very low loads
the day would be used to electrolyze water into hydrogen the rate at which hydrogen and oxygen are produced
and oxygen, which could be stored in the structural ele- (which is proportional to current density) may be lower
ments for the wing, and then used during the night in a than the rate at which these gases permeate through elec-
fuel cell to generate electricity for aircraft propulsion. trolyte, and mix with each other. This may create haz-
This process could continue day after day without need ardous conditions inside the electrolyzer. Hydrogen
to bring the aircraft down. Such an aircraft would be flammability limits in oxygen are between 4.6% and
far less expensive than satellites. In this case, the effi- 93.9%, but the alarms and automatic shutdown of the
ciency and the mass of the electrolyzer, fuel cell and electrolyzer are set at much safer concentrations. Hydro-
hydrogen and oxygen storage are critical for accomplish- gen permeation rate at 80 C through Nafion 117, typi-
ing the mission. This energy storage system has potential cally used in PEM electrolyzers, should be less than
to be much lighter than the batteries, with specific energy 1.25 · 10
4 cm3/s/cm2 at atmospheric pressure, corre-
in excess of 400 W h/kg, and when hydrogen and oxygen sponding to a current density of 0.002 A/cm2 (Kocha
are stored in structural elements with little additional et al., 2002), which is rather negligible compared to
weight energy density may exceed 700 W h/kg. 1 A/cm2, a typical current density in PEM electrolyzers
A similar system is used in a ‘‘water rocket’’ concept at full power. However, hydrogen permeation rate in-
(Mitlitsky et al., 1999). It makes sense from safety and creases linearly with pressure, which means that at
weight points of view to carry water into the orbit, 200 bar hydrogen loss would be 0.4 A/cm2. Fig. 8
and then use PVs to generate electricity to electrolyze shows hydrogen permeation through Nafion 117 in an
water into hydrogen and oxygen. These gases would electrolyzer configuration as a function of pressure and
be stored for use as propellants for repositioning the sat- temperature. Oxygen permeation rate through hydrated
ellite and for electricity generation via fuel cell during Nafion membrane is considerably lower (Sakai et al.,
periods when the PVs are not insolated. 1986).
Another problem related to operation with a highly
variable power source is thermal management. The elec-
4. Issues trolyzer takes time to reach its normal operating temper-
ature, but due to intermittent operation it may operate
Operation of a PEM electrolyzer in conjunction with most of the time at a temperature below nominal, which
renewable energy sources, and particularly with a PV results in a lower efficiency.
panel or array, has several specific issues that will be dis-
cussed below. 4.3. Output pressure

4.1. Relative sizing of an electrolyzer and PV array A PEM electrolyzer may be designed to generate
hydrogen at elevated pressure. This includes adequate
An electrolyzer may be sized to receive all the power membrane support, appropriate gasketing and account-
generated from a PV array, but it would operate with ing for material creep. Proton Energy Systems sell PEM
the same capacity factor as the PV array, which is deter- electrolyzers that generate hydrogen at 14 bar (200 psig),
mined by the sun availability. Capacity factor is a coef- but it has reported operation at 200 bar (3000 psi) in the
ficient of utilization of installed capital, and therefore it laboratories. Pressurization during electrolysis results in
is an important factor in determining the economics of slightly higher cell voltages (as predicted by the Nernst
any power generating or energy conversion device. A
more economical option may be to size the electrolyzer
1.4
at a power lower than the PV maximum power output.
H2 Permeation ( µg H2 /cm 2s)

In that case some of the power from the PV would be 1.2

unutilized, or stored in the batteries, but the electrolyzer 1
75 C
would operate with a higher capacity factor. For any 0.8 55 C
combination of PV availability and load profiles, there 0.6 25 C
is an optimum electrolyzer capacity (Barbir, 1999b).
0.4
Economics of PV-hydrogen systems greatly depends on
the configuration of the system and its application, in 0.2
addition to the available solar insolation. 0
0 1000 2000 3000 4000
Pressure (PSI)
4.2. Intermittent operation
Fig. 8. Hydrogen permeation through Nafion 117 in an
Direct coupling of an electrolyzer with a PV array electrolyzer configuration as a function of pressure and
implies intermittent operation with highly variable temperature.
 F. Barbir / Solar Energy 78 (2005) 661–669 667

equation), which corresponds to an isothermal compres- by the current density, which in turn directly corre-
sion work. It is therefore the most efficient way of com- sponds to the rate of hydrogen production per unit of
pression. The only loss is hydrogen permeation through electrode active area. A higher voltage would result in
the polymer membrane. High hydrogen pressure is often more hydrogen production, but at a lower efficiency.
required if hydrogen is to be stored in high pressure Typically, cell voltage is selected at about 2 V, but a low-
tanks for later use, or used as a fuel in an automobile, er nominal voltage (as low as 1.6 V) may be selected, if
or shipped through pipelines. the efficiency is more important than the size (and capi-
tal cost) of the electrolyzer.
4.4. Oxygen generation Another ‘‘source’’ of inefficiency is hydrogen perme-
ation (loss) through the polymer membrane. This is typ-
The water electrolysis results in both hydrogen and ically insignificant at low operating pressures, but it may
oxygen being produced on the opposite sides of the significantly affect the overall efficiency at very high
membrane. The electrolyzer produces stoichiometric pressures (>100 bar).
amounts of oxygen, i.e., for every two moles of hydro- In addition there are power losses in voltage regula-
gen a mol of oxygen is produced as well. Oxygen is typ- tion and some power is needed for the auxiliary equip-
ically vented (unless oxygen generation is the primary ment (pumps, fans, solenoid valves, instrumentation
purpose of water electrolysis such as in submarines), and controls). Typical industrial electrolyzers have elec-
but if needed oxygen may be captured and stored for tricity consumption between 4.5 and 6.0 kW h/N m3, cor-
later use (for example in fuel cell). Storage and handling responding to the efficiency of 65–80%. Fig. 9 shows
of oxygen, particularly at elevated pressures, requires typical PEM electrolyzer efficiency curves at various
special care. pressures.
The electrolyzer efficiency is therefore:
4.5. Water consumption
1:482 i
iloss gDC
gEL ¼ ð1Þ
3 V cell i 1þn
Theoretically from 1 l of water 1.24 N m of hydrogen
can be produced. Actual water consumption is about where Vcell is the individual (average) cell potential (V), i
25% higher, since both gases leave the electrolyzer wet, is the operating current density (A/cm2), iloss is the inter-
and some water is lost from the system due to oxygen ex- nal current and hydrogen loss (A/cm2), gDC is the effi-
haust and periodic hydrogen purge. This is particularly ciency of DC/DC voltage regulator, n is the ratio
important for the systems that are supposed to work between parasitic power and net power consumed by
in a closed loop with a fuel cell. If the fuel cell uses air the electrolyzer.
instead of pure oxygen, some water is lost there due to Coupling of a source such as PV array with an elec-
air exhaust from the fuel cell. trolyzer may result in somewhat lower efficiency, due to
the losses related to power/voltage matching. DC/DC
4.6. Efficiency voltage regulators may be designed to operate with the
efficiency as high as 93–95%, but this high efficiency
The efficiency of an electrolyzer is inversely propor- may be achieved only in a very narrow power range.
tional to the cell potential, which in turn is determined In a highly variable mode of operation such as with

1
0.9
electrolyzer efficiency (HHV)

0.8
0.7
0.6
0.5
0.4
200 psi
0.3
2000 psi
0.2
3000 psi
0.1
0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
H2 generation rate (g/h/cm²)

Fig. 9. Electrolyzer stack efficiency for different delivery pressures.

668 F. Barbir / Solar Energy 78 (2005) 661–669

2.8

2.6

2.4 Time = 2,000 to 14,000 hr

Cell Voltage (Volts)

Decay rate = 3 µV/cell hr

2.2

Time = 400 to 2,000 hr

2.0 Decay rate = 20 µV/cell hr

1.8

1.6
0 2,000 4,000 6,000 8,000 10,000 12,000 14,000
Operating Hours

Fig. 10. Voltage increase with time during continuous electrolysis. (single cell, 929 cm2, 1.1 A/cm2, 10 bar, 51–55 C).

the input from a PV array, this efficiency may be consid- regulator may be used with or without maximum power
erably lower. If there is a good match between the polar- point tracking. PEM electrolyzers are simpler than con-
ization curves of PVs and electrolyzers, experience from ventional alkaline electrolyzers, and can be operated in
a handful of PV/electrolysis pilot plants shows that they variable power input mode. PEM electrolyzers can gen-
can be matched directly (with no power tracking elec- erate hydrogen (and optionally oxygen) at pressures up
tronics) with relatively high efficiency, >93% coupling to 200 bar, with very little additional power consump-
efficiency (Steeb et al., 1990). tion, which may be attractive for the application where
hydrogen needs to be stored or used at elevated pressure.
4.7. Performance degradation with time PEM electrolyzers produce hydrogen with very high
purity (up to 99.999), which may be used in a fuel cell.
When sizing an electrolyzer for operation in conjunc- Economics of PV-hydrogen is determined not only
tion with the PVs, one must take into account voltage in- by the cost of the PV array and the electrolyzer, but also
crease with time. If the efficiency or the hydrogen by the capacity factor (of both PVs and electrolyzer),
generation rate are important, then end of life electro- and the electrolyzer efficiency. Depending on configura-
lyzer performance must be taken into account rather tion of the system and application, the capacity factors
than beginning of life performance. Voltage increase is of the PVs and the electrolyzer do not necessarily have
steeper in the first couple of thousands of hours of oper- to be the same. Economics must also take into account
ation, 20–50 lV/h, but then it levels off (<3 lV/h), as the cost of the competitive technology or commodity
shown in Fig. 10 (Anderson et al., 2002). The reasons (electricity and fuel). Because of high cost of the PVs
for voltage increase are equilibration of water content and the electrolyzer, generation of hydrogen currently
in the membrane, and oxidation of catalyst and other may be justified only in demonstration projects, or in re-
metallic components. With good design and careful mote areas or for special applications.
selection of materials, this performance degradation
may be minimized.
References

Agbossou, K., Chahine, R., Hamelin, J., Laurencelle, F.,

5. Conclusions
Anouar, A., St-Arnaud, J.-M., Bose, T.K., 2001. Renewable
energy systems based on hydrogen for remote applications.
PEM electrolysis is a viable alternative for generation J. Power Sources 96, 168–172.
of hydrogen in conjunction with renewable energy Anderson, E., Molter, T., Mitlitsky, F., 2002. Recent advances
sources. It particularly matches and complements the in hydrogen generation using PEM water electrolysis,
photovoltaics. It may be coupled either directly, if the presented at the 201st Electrochemical Society Meeting,
polarization i–V curves are well matched, or a DC/DC Philadelphia, PA.
 F. Barbir / Solar Energy 78 (2005) 661–669 669

Barbir, F., 1999a. Safety issues of hydrogen in vehicles, hydrogen energy system. Int. J. Hydrogen Energy 22, 465–
presented at Fuel Cell Vehicles Conference, IIR Ltd., 470.
London. Mitlitsky, F., Colella, N.J., Myers, B., Anderson, C.J., 1993.
Barbir, F., 1999b. Integrated renewable hydrogen utility Regenerative fuel cells for solar powered aircraft. In: 28th
system. In: Proc. 1999 DoE Hydrogen Program Annual Intersociety Energy Conversion Engineering Conference
Review Meeting, Lakewood, CO. (IECEC).
Barbir, F., Veziroglu, T.N., 1992. Environmental benefits of the Mitlitsky, F., Weisberg, A.H., Carter, P.H., Dittman, M.D.,
solar hydrogen energy system. In: Singhal, R.K., Mehrotra, Myers, B., Humble, R.W., Kare, J.T., 1999. Water rocket—
K., Fytas, K., Collins, J.-L. (Eds.), Environmental Issues Electrolysis propulsion and fuel cell power, AIAA-99-4609,
and Management of Waste in Energy and Mineral Produc- AIAA Space Technology Conference, Albuquerque, NM.
tion. Balkema, Rotterdam, pp. 1209–1218. Ogden, J.M., Williams, R.H., 1989. Solar Hydrogen: Moving
Barbir, F., Lillis, M., Mitlitsky, F., Molter, T., 2002. Regen- Beyond Fossil Fuels. World Resources Institute, Washing-
erative fuel cell applications and design options. In: Proc. ton, DC.
14th World Hydrogen Energy Conference, Montreal, Oney, F., Veziroglu, T.N., Dülger, Z., 1994. Evaluation of
Canada. pipeline transportation of hydrogen and natural gas mix-
Collier, K., 1994. System performance of a photovoltaic- tures. Int. J. Hydrogen Energy 19, 813–822.
powered water electrolysis system. In: Block, D.L., Veziro- Oud, H.A., Steeb, H., 1996. The German-Saudi HYSOLAR
glu, T.N. (Eds.), Proc. 10th World Hydrogen Energy Program. In: Veziroglu, T.N., Winter, C.-J., Baselt, J.P.,
Conference, Cocoa Beach, FL, Hydrogen Energy Progress Kreysa, G. (Eds.), Proc. 11th World Hydrogen Energy
X, vol. 1, pp. 449–457. Conference, Stuttgart, Germany, Hydrogen Energy Pro-
Coluccia, M., Gaggio, G., Guarna, S., Spazzafumo, G., 1994. gress XI, vol. 1, pp. 273–282.
Pre-feasibility analysis of an energy supply system for Rosa, F., Lopez, E., Garcia-Konde, A.G., Luque, R., del Pozo,
Southern Europe: Technical. Int. J. Hydrogen Energy 19, F., 1994. Intermittent operation of a solar hydrogen
957–963. production facility: yearly evaluation. In: Block, D.L.,
Dienhart, H., Siegel, A., 1992. Hydrogen storage in isolated Veziroglu, T.N. (Eds.), Proc. 10th World Hydrogen Energy
electrical energy systems with photovoltaic and wind energy. Conference, Cocoa Beach, FL, Hydrogen Energy Progress
In: Veziroglu, T.N., Derive, C., Pottier, J. (Eds.), Proc. 9th X, vol. 1, pp. 421–430.
World Hydrogen Energy Conference, Paris, France, Hydro- Sakai, T., Takenaka, H., Torikai, E., 1986. Gas diffusion in
gen Energy Progress IX, vol. 2, pp. 713–731. dried and hydrated nafions. J. Electrochem. Soc. 133, 88–92.
Drolet, B., Gretz, J., Kluyskens, D., Sandmann, F., Wurster, Spinadel, E., Gil, J., Gamallo, F., 2002. Wind-powered
R., 1996. The euro-québec hydro-hydrogen pilot project electrolysis for stand alone operation: an approach to a
[EQHHPP]: demonstration phase. Int. J. Hydrogen Energy specific wind converter design. In: Proc. 14th World
21, 305–316. Hydrogen Energy Conference, Montreal, Canada.
Galli, S., De Paoli, G., Ciancia, A., 1994. An experimental solar Steeb, H.A., Brinner, A., Bubmann, H., Seeger, W., 1990.
PV-hydrogen fuel cell energy system: preliminary results. In: Operation experience of a 10 kW PV-electrolysis system in
Block, D.L., Veziroglu, T.N. (Eds.), Proc. 10th World different power matching modes. In: Veziroglu, T.N.,
Hydrogen Energy Conference, Cocoa Beach, FL, Hydrogen Takahashi, P.K. (Eds.), Hydrogen Energy Progress VIII,
Energy Progress X, vol. 1, pp. 439–447. vol. 2. Pergamon Press, New York, pp. 691–700.
Gretz, G., Baselt, J.P., Ullmann, O., Wendt, H., 1990. The 100 Swain, M.R., Swain, M.N., 1992. A comparison of H2, CH4,
MW Euro-Quebec hydro-hydrogen pilot project. Int. J. and C3H8 fuel leakage in residential settings. Int. J.
Hydrogen Energy 15, 419–424. Hydrogen Energy 17, 807–815.
Hollmuller, P., Joubert, J.-M., Lachal, B., Yvon, K., 1998. Szyszka, A., 1996. Nine years solar hydrogen demonstration
Evaluation of a 5 kW photovoltaic hydrogen production project Neunburg vorm Wald, Germany. In: Veziroglu,
and storage installation for a residential home in Switzer- T.N., Winter, C.-J., Baselt, J.P., Kreysa, G. (Eds.), Proc.
land. In: Bolcich, J.C., Veziroglu, T.N. (Eds.), Proc. XII 11th World Hydrogen Energy Conference, Stuttgart, Ger-
World Hydrogen Energy Conference, Buenos Aires, Argen- many, Hydrogen Energy Progress XI, vol. 1, pp. 259–272.
tina, Hydrogen Energy Progress XII, pp. 423–443. Thomas, C.E., 1996. Preliminary hydrogen vehicle safety
Justi, E.W., 1987. A Solar-Hydrogen Energy System. Plenum report, prepared by Directed Technologies, Inc. for Ford
Press, New York. Motor Company, contract No. DE-AC02-94CE50389, US
Keith, G., Leighty, W., 2002. Transmitting 4,000 MW of new Department of Energy, Washington, DC.
windpower from N. Dakota to Chicago: new HVDC electric Vanhanen, J.P., Lund, P.D., Tolonen, J.S., 1998. Electrolyser-
lines or hydrogen pipeline. In: Proc. 14th World Hydrogen metal hydride-fuel cell system for seasonal energy storage.
Energy Conference, Montreal, Canada. Int. J. Hydrogen Energy 23, 267–271.
Kocha, S., Plasse, P., Onishi, L., Wheeler, D., Bett, J., 2002. In Veziroglu, T.N., Barbir, F., 1991. Solar-hydrogen energy
situ electrochemical measurement of hydrogen crossover system: the choice of the future. Environ. Conser. 18, 304–
currents in PEMFC. In: Proc. Fuel Cell Technology: 312.
Opportunities and Challenges, AIChE Spring National Veziroglu, T.N., Barbir, F., 1998. Hydrogen Energy Technol-
Meeting, New Orleans, LA, 2002, pp. 307–312. ogies, UNIDO-Emerging Technologies Series, United
Lehman, P.A., Chamberlin, C.E., Pauletto, G., Rocheleau, Nations Industrial Development Organisation, Vienna,
M.A., 1997. Operating experience with a photovoltaic- Austria.