Proceedings of the 2000 Hydrogen Program Review

NREL/CP-570-28890

LOW TEMPERATURE AND HIGH PRESSURE EVALUATION OF INSULATED

PRESSURE VESSELS FOR CRYOGENIC HYDROGEN STORAGE

S. M. Aceves,
Lawrence Livermore National Laboratory
7000 East Ave., L-641
Livermore, CA 94551, USA
[email protected]

J. Martinez-Frias
Centro de Ingenieria y Desarrollo Industrial
Queretaro, Qro., Mexico
[email protected]

O. Garcia-Villazana
FIMEE, Universidad de Guanajuato
Salamanca, Gto. Mexico
[email protected]

Abstract

Insulated pressure vessels are cryogenic-capable pressure vessels that can be fueled with liquid
hydrogen (LH2) or ambient-temperature compressed hydrogen (CH2). Insulated pressure vessels
offer the advantages of liquid hydrogen tanks (low weight and volume), with reduced
disadvantages (fuel flexibility, lower energy requirement for hydrogen liquefaction and reduced
evaporative losses). The work described here is directed at verifying that commercially available
pressure vessels can be safely used to store liquid hydrogen. The use of commercially available
pressure vessels significantly reduces the cost and complexity of the insulated pressure vessel
development effort. This paper describes a series of tests that have been done with aluminumlined, fiber-wrapped vessels to evaluate the damage caused by low temperature operation. All
analysis and experiments to date indicate that no significant damage has resulted. Required
future tests are described that will prove that no technical barriers exist to the safe use of
aluminum-fiber vessels at cryogenic temperatures.

Introduction
Hydrogen-fueled vehicles present features that make them serious candidates as alternatives to
todays petroleum-powered vehicles. Hydrogen vehicles can use the advanced technology of
electric vehicles to improve environmental quality and energy security, while providing the
range, performance, and utility of todays gasoline vehicles.
Probably the most significant hurdle for hydrogen vehicles is storing sufficient hydrogen on
board. Hydrogen storage choices can determine the refueling time, cost, and infrastructure
requirements, as well as indirectly influence energy efficiency, vehicle fuel economy,
performance, and utility. There are at least three viable technologies for storing hydrogen fuel on
cars. These are compressed hydrogen gas (CH2), metal hydride adsorption, and cryogenic liquid
hydrogen (LH2). Each of these has significant disadvantages.
Storage of 5 kg of hydrogen (equivalent in terms of energy to 19 liters; 5 gallons of gasoline) is
considered necessary for a general-purpose vehicle, since it provides a 640 km (400 mile) range
in a 34 km/liter (80 mpg) hybrid vehicle or fuel cell vehicle. Storing this hydrogen as CH2
requires a volume so big that it is difficult to package in light-duty cars (Pentastar Electronics,
1997). The external volume for a pressure vessel storing 5 kg of hydrogen at 24.8 MPa (3600
psi) is 320 liters (85 gal). Hydrides are heavy (300 kg for 5 kg of hydrogen, Michel et al., 1996),
resulting in a substantial reduction in vehicle fuel economy and performance.
Low-pressure LH2 storage is light and compact, and has received significant attention due to its
advantages for packaging (Braess and Strobl, 1996). Significant recent developments have
resulted in improved safety (Pehr, 1996a, 1996b), and fueling infrastructure (Hettinger et al,
1996). Disadvantages of low-pressure LH2 storage are the substantial amount of electricity
required for liquefying the hydrogen (Peschka, 1992); the evaporation losses that may occur
during fueling low-pressure LH2 tanks (Wetzel, 1996); and the evaporative losses that occur
during periods of inactivity, due to heat transfer from the environment.
An alternative is to store hydrogen in an insulated pressure vessel that has the capacity to operate
at LH2 temperature (20 K), and at high pressure (24.8 MPa; 3600 psi). This vessel has the
flexibility of accepting LH2 or CH2 as a fuel. Filling the vessel with ambient-temperature CH2
reduces the amount of hydrogen stored (and therefore the vehicle range) to about a third of its
value with LH2.
The fueling flexibility of the insulated pressure vessels results in significant advantages.
Insulated pressure vessels have similar packaging characteristics as liquid hydrogen tanks (low
weight and volume), with reduced energy consumption for liquefaction. Energy requirements for
hydrogen liquefaction are lower than for liquid hydrogen tanks because a car with an insulated
pressure vessel can use, but does not require, cryogenic hydrogen fuel. A hybrid or fuel cell
vehicle with 34 km/l (80 mpg) gasoline-equivalent fuel economy could be refueled with
ambient-temperature CH2 at 24.8 MPa (3600 psi) and still achieve a 200 km range, suitable for
the majority of trips. The additional energy, cost, and technological effort for cryogenic refueling
need only be undertaken (and paid for) when the additional range is required for longer trips.
With an insulated pressure vessel, vehicles can refuel most of the time with ambient-temperature

hydrogen, using less energy, and most likely at lower ultimate cost than LH2, but with the
capability of having 3 times the range of room-temperature storage systems. Use of compressed
hydrogen in all trips under 200 km (which represent 85% of all the distance traveled in the USA,
(Klinger and Kuzmyak, 1984) reduces the total energy consumption by 16% over the energy
consumed by a vehicle that is always filled with LH2.
Insulated pressure vessels also have much reduced evaporative losses compared to LH2 tanks.
This has been demonstrated in a previous work (Aceves and Berry, 1998), which presents a
thorough analysis of evaporative losses in cryogenic pressure vessels based on the first law of
thermodynamics. Figure 1 illustrates some of the main results. This figure shows hydrogen
losses during vehicle operation. The figure assumes that two vehicles are fitted with cryogenic
hydrogen storage tanks with the same capacity (5 kg). One vehicle has a low-pressure (0.5 MPa;
70 psia maximum) conventional liquid hydrogen tank, and the other has an insulated pressure
vessel. The vehicles are identical in every respect, except for the tanks. The vessels are filled to
full capacity with liquid hydrogen, and then the vehicles are driven a fixed distance every day.
When the fuel runs out, the amount of fuel burned by the engine and the amount of fuel lost to
evaporation are calculated, and the results are shown in Figure 1. The figure shows total
cumulative evaporative hydrogen losses out of a full tank as a function of the daily driving
distance, for a high-efficiency vehicle (34 km/l or 80 mpg gasoline equivalent fuel economy). As
expected, evaporative losses increase as the daily driving distance is reduced, because less
driving results in a longer time for hydrogen evaporation. The figure shows that a low-pressure
LH2 tank loses hydrogen even when driven 100 km per day. Losses from a LH2 tank grow
rapidly as the daily driving distance drops. A vehicle driven 50 km per day (the average for the
USA, Aceves and Berry, 1998) loses almost 1 kg (20%) of the fuel to evaporation. On the other
hand, insulated pressure vessels lose hydrogen only for very short daily driving distances (less
than 5 km/day). Most vehicles are driven considerably more than this distance, so that most
vehicles equipped with an insulated pressure vessel would never lose any hydrogen to
evaporation.
The low losses in insulated pressure vessels are the result of the flow work (work required to
extract the hydrogen from the vessel, VanWylen and Sonntag, 1978). The hydrogen stored in the
vessel does work as the hydrogen is being extracted, cooling down in the process. This effect is
very significant for hydrogen, due to its low molecular weight.
From an engineering and economic perspective, insulated pressure vessels strike a versatile
balance between the cost and bulk of ambient-temperature CH2 storage, and the energy
efficiency, thermal insulation and evaporative losses of LH2 storage.
Considering all the potential benefits of insulated pressure vessels, it is important to determine
what type of pressure vessel could be operated at both high pressure and cryogenic temperature.
Of the available pressure vessel technologies commonly used for vehicular storage of natural gas
(Institute of Gas Technology, 1996), it appears that aluminum-lined, composite-wrapped vessels
have the most desirable combination of properties for this application (low weight and affordable
price). However, commercially available aluminum-composite pressure vessels are not designed
for low temperature applications.

5
low-pressure LH2 tank
insulated pressure vessel

hydrogen losses, kg

0
0

10

20

30

40

50

60

70

80

90

100

daily driving distance for a 34 km/l (80 mpg) car, km

Figure 1. Cumulative hydrogen losses in kg as a function of daily
driving distance, for vehicles with 17 km/liter (40 mpg); or 34 km/l
(80 mpg) fuel economy, for three cryogenic hydrogen storage
vessels.

This paper describes work in progress directed at evaluating the possibility of using
commercially available aluminum-fiber pressure vessels at cryogenic temperatures and high
pressures, as would be required for vehicular hydrogen storage in insulated pressure vessels. The
paper gives a description of previous and ongoing tests, followed by a list of future tests. The
purpose of these tests is to demonstrate that no technical barriers exist that prevent the use of
aluminum-fiber pressure vessels at cryogenic temperatures. However, it is recognized that
further tests may be necessary for certification, which is required for commercialization of
insulated pressure vessels. Certification also requires subjecting several pressure vessels to the
required tests, to determine the statistical significance of the results (Bauer, 1996). Obtaining
vessel certification is beyond the scope of this work. It is the authors belief that the current work
will provide vessel manufacturers with an additional option for vehicular hydrogen storage. It
would then be left to manufacturers to conduct the required tests to achieve certification and
guarantee safety under all possible conditions. It is worth pointing out that the aluminum-fiber
pressure vessels used in all these tests have been certified according to the DOT standards (CFR-

DOT, Title 49, 1996a) for storage of compressed natural gas. The current tests are being
performed because these pressure vessels are being used outside their typical range of operation,
and also because they are being modified to attach a vacuum insulation to them.
Previous and Ongoing Tests
Pressure and Temperature Cycling
Pressure vessels have been cycled through 900 high-pressure cycles and 100 low-temperature
cycles. The cycles are alternated, running 9 pressure cycles followed by a temperature cycle, and
repeating this sequence 100 times. This test is expected to replicate what would happen if these
vessels were used in a hydrogen-fueled car. Liquid nitrogen is used for low-temperature cycling
and gaseous helium for high-pressure cycling. To accomplish the required testing, an
experimental setup has been built inside a high-pressure cell. A schematic is shown in Figure 2.
The valves shown in the schematic are controlled by computer, which allows the system to run
with no supervision, resulting in fast cycling. An aramid-aluminum and a carbon fiber-aluminum
pressure vessel have been cycled. The characteristics of these are listed in Table 1.
pressure
transducer
V-1

RV-1

V-2
Pump

NC

SV-3

34 MPa Vessels

NO

SV-4

RV-2

Helium16.5 MPa
transducer
SV-5
NC

NO

LN Dewar

RV-3 SV-6

Test
Vessel

Figure 2. Schematic of the experimental setup for temperature and pressure

cycling of pressure vessels.

Table 1. Characteristics of the Tested Hydrogen Vessels

and Their Planned Insulation

Mass of hydrogen stored, kg

Vessel weight, kg
Internal volume, liters
Internal diameter, m
Internal surface area, m2
Design pressure, MPa (psi)
Performance factor1, m (106in)
Safety factor
__________________________
1 defined as burst pressure*volume/weight.

AramidAluminum
1.13
10
17.6
0.2
0.48
24.1 (3500)
13000 (0.5)
3.0

Carbon FiberAluminum
0.44
4.1
6.8
0.17
0.25
31 (4500)
13115 (0.51)
2.5

Two cyclic tests have been completed, one on an aramid-aluminum pressure vessel and other on
a carbon fiber-aluminum pressure vessel. The vessels have not failed during the test, and they
have not shown superficial evidence of damage under observation. The carbon fiber-aluminum
vessel was instrumented with strain gages in addition to the thermocouples and pressure sensor.
Results from the strain gages will be used for validating the finite element analysis.
Burst Test
The aramid-aluminum and the carbon fiber-aluminum pressure vessels were burst-tested after
being cycled and ultrasound-tested. The burst test was conducted according to the Code of
Federal Regulations-Department of Transportation standards for pressure vessel certification
(CFR-DOT, 1996a). Figure 3 shows the variation of pressure as a function of time for the
aramid-aluminum vessel. Failure occurred by hoop mid cylinder separation, which is the
preferred mode of failure. The burst pressure was 94.2 MPa (13.7 ksi), which is substantially
higher than the minimum burst pressure of 72.4 MPa (10.5 ksi). The very high value of the burst
pressure compared to the minimum burst pressure may be due in part to work hardening that
took place during the cold cycling of the vessel. The carbon fiber-aluminum also failed at a
pressure higher than the minimum required.
Finite Element Analysis
Cyclic, ultrasound and burst testing of the pressure vessels is being complemented with a finite
element analysis, which will help to determine the causes of any potential damage to the vessel
during low temperature operation. Finite element analysis is currently in progress. A mesh has
been built and preliminary runs have been made. Physical properties of fiber-epoxi laminae were
obtained from available literature at ambient and cryogenic temperatures (Reed and Golda, 1994,
Morgan and Allred, 1989). Lamina properties are then converted into properties of the composite
matrix. This is done by using a computer program (Hull and Clyne, 1996). This program

assumes that the matrix is a homogeneous, orthotropic material. The properties of the matrix will
be used in the finite element thermal and stress analysis. Results of the finite element analysis
will be validated by comparison with the strain gage measurements.

burst pressure: 13657 psig

14

pressure, ksi

12
minimum burst pressure

10
8
6

maximum operating pressure

4
2
0

20

40

60

80

100

120

140

160

180

time, seconds
Figure 3. Pressure as a function of time during the burst test of the aluminumlined, aramid-wrapped vessel. The burst pressure was 94.17 MPa (13657 psig).

Insulation Design and Insulated Pressure Vessel Construction

Insulated pressure vessels have been designed to operate with multilayer vacuum superinsulation
(MLVSI). MLVSI has a good thermal performance only under a high vacuum, at a pressure
lower than 0.01 Pa (7.5x10-5 mm Hg; Kaganer, 1969). Therefore, the use of MLVSI requires that
an outer jacket be built around the vessel. Two designs for the insulation have been built: a firstgeneration design and a second-generation design. The first-generation vessel is a 1/5-scale
vessel that stores about 1 kg of liquid hydrogen, and it is shown in Figure 4. This design has been
built for cyclic testing and for DOT certification tests. The insulation design includes access for
instrumentation for pressure, temperature and level, as well as safety devices to avoid a
catastrophic failure in case the hydrogen leaks into the vacuum space. Five pressure vessels have
been built according to the first-generation pressure vessel design. One of these has been
extensively cycled at a high-pressure cell, and has also been tested with liquid hydrogen. The
remaining 4 insulated pressure vessels have been delivered to the pressure vessel manufacturer to
verify compliance with DOT standards.

Figure 4. Insulation design for first-generation pressure vessel. The figure shows
a vacuum space, for obtaining high thermal performance from the multilayer
insulation, and instrumentation for pressure, temperature and level. Dimensions
are given in cm.

The second-generation pressure vessel design is shown in Figure 5. This vessel can store about 6
kg of liquid hydrogen. This design includes a vapor shield to reduce evaporative losses in
addition to the instrumentation and safety devices that exist in the first generation vessel. These
vessels are currently being built. The second generation of pressure vessels will be used for DOT
and SAE tests, and for incorporation into a demonstration vehicle.
Cyclic Testing of Insulated Pressure Vessels
The insulated pressure vessels of the first generation (Figure 4) have been cycle tested. This is
done to verify that the pressure vessel or the outer jacket does not develop leaks during repeated
stresses that occur during cycling. One of these first-generation pressure vessels has been
subjected to 1000 cycles, following the same procedure as previously used for the pressure
vessels with no insulation (see Pressure and Temperature Cycling above). The remaining four
first-generation pressure vessels have been subjected to a cold shock and pressure test before
being delivered to the vessel manufacturer for DOT testing. The experimental setup for this test
is the same as previously used for cyclic testing (Figure 2). The test procedure is as follows: The
vessel is pressurized with compressed helium to 1.2 times the Maximum Allowable Working
Pressure (MAWP). The pressure is held for a minimum of 30 minutes. Then, the pressure vessel
is shock conditioned by cycling it 3 times to low temperature with liquid nitrogen. Finally, The
vessel is leak tested with helium to 0.25 times the MAWP. Any leakage detected with a mass
spectrometer leak detector is unacceptable.

Figure 5. Insulation design for second-generation pressure vessel. The figure

shows a vacuum space, for obtaining high thermal performance from the
multilayer insulation, instrumentation for pressure, temperature and level, and a
vapor shield for reducing hydrogen evaporative losses.

The same shock conditioning test procedure will be used for the second-generation, full-size
pressure vessel before being tested according to the DOT and the SAE standards.
Liquid and Gaseous Hydrogen Testing
A first-generation insulated pressure vessel has been tested with liquid and gaseous hydrogen.
The vessel was first shock-tested and leak-tested. The insulated pressure vessel was then
transported to a remote facility for testing with liquid hydrogen. Testing involved filling the
vessel with LH2 to study the insulation performance, the performance of the sensors, and the
problems involved with pumping the LH2 into the vessel. This test is expected to replicate what
would happen to the vessel during fueling and operation in an LH2-fueled car. The test was
conducted successfully. There was no damage to the vessel due to the low temperature operation,
all the instrumentation operated properly at the low temperature, and there was no hydrogen
ignition or explosions.

Future Work
Future work includes performing additional tests on insulated pressure vessels. A list of tests has
been obtained from standards issued by institutions dedicated to pressure vessel and vehicular
safety, including the Department of Transportation (DOT), the Society of Automotive Engineers
(SAE) and the National Fire Protection Association (NFPA). These institutions have established
test procedures for pressure vessels that guarantee safety under regular operating conditions.
Successful completion of these tests by insulated pressure vessels will demonstrate that no
significant safety problems exist with pressure vessel operation at low temperature. However,
additional tests (and multiple execution of any individual test) will be still required for insulated
pressure vessel certification. The list of planned tests is:

Cycling, ambient temperature. 10000 cycles from less than 10% of the service pressure to
the service pressure, 10 cycles per minute maximum (CFR-DOT, Title 49, 1996a). Each
test cylinder must withstand the cycling pressurization test without any evidence of
visually observable damage, distortion, or leakage.

Cycling, environmental. 10 cycles per minute maximum. 1) 5000 cycles from zero to
service pressure with tank at 60oC (140oF) and air at ambient temperature and 95%
humidity, 2) 5000 cycles from zero to service pressure with tank at 51.1oC (-60oF) and
air at ambient temperature, 3) 30 cycles from zero to service pressure, ambient
conditions 4) burst test the cycled vessel (CFR-DOT, Title 49, 1996a). Each test cylinder
must withstand the cycling pressurization test without any evidence of visually
observable damage, distortion, or leakage.

Cycling, Thermal. 10 cycles per minute maximum. 1) 10 000 cycles from zero to service
pressure at ambient temperature, 2) 20 thermal cycles with tank temperature varying from
93.3oC (200oF) to 51.1oC (-60oF) at service pressure, 3) burst test the cycled vessel
(CFR-DOT, Title 49, 1996a). Each test cylinder must withstand the cycling
pressurization test without any evidence of visually observable damage, distortion, or
leakage.

Gunfire. Pressurize vessel with air or nitrogen to service pressure, and impact the vessel
with a 0.30 caliber armor-piercing projectile with a speed of 853 m/s (2800 ft/s). The
cylinder is positioned in such a way that the impact point is in the cylinder side wall at a
45o angle with respect to the longitudinal axis of the cylinder. The distance from the
firing location to the cylinder may not exceed 45.7 meters (150 feet) (CFR-DOT, Title
49, 1996a). The cylinder shall not fail by fragmentation.

Bonfire. Pressurize cylinder with air or nitrogen to service pressure. Set pressure relief
devices to discharge at 83% of the cylinder test pressure. The cylinder shall be exposed to
fire until the gas is fully vented. The temperature measured on the surface tank exposed
to the fire has to be between 850 and 900oC (CFR-DOT, Title 49, 1996a). The venting of
the gas must be predominantly through the pressure relief device.

Drop Test from 3 m (10 ft). 1) The cylinder is dropped vertically onto the end, 2) the
cylinder is dropped horizontally onto the side wall, 3) the cylinder is dropped onto a 3.8 x
0.48 cm (1 x 3/16 inch) piece of angle iron, 4) after the drops, the vessel is cycled over
1000 pressure cycles from 10% of service pressure to the service pressure, at 10 cycles
per minute (CFR-DOT, Title 49, 1996). The cylinder then has to be burst tested; the burst
pressure of this vessel has to be at least 90 % of the minimum burst pressure.

Drop tests from 10 m and 3 m. 1) Drop from 10 m. The drop test subjects a full-size
vehicle fuel tank to a free-fall impact onto an unyielding surface from a height of 10 m.
The fuel tank is released by firing one or more explosive cable cutters simultaneously.
The fuel tank impacts the outer shell on the critical area as determined by the
manufacturer. The fuel tank is filled with an equivalent full weight of liquid nitrogen
saturated to at least 50% of the maximum allowable working pressure of the fuel tank. 2)
Drop from 3 m. The drop test subjects a full-size vehicle fuel tank to a free-fall impact
onto an unyielding surface from a height of 3 m. The fuel tank is released by firing one or
more explosive cable cutters simultaneously. The fuel tank impacts the outer shell on the
critical area as determined by the manufacturer. The fuel tank is filled with an equivalent
full weight of liquid nitrogen saturated to at least 50% of the maximum allowable
working pressure of the fuel tank (SAE J2343, 1997). There shall be no loss of product
for a period of 1 hour after the drop other than relief valve operation and loss of vapor
between the filler neck and the secondary relief valve in the case of a test involving the
filler neck. Loss of vacuum, denting of the vessel, piping and piping protection, and
damage to the support system are acceptable.

Flame test. The tank should contain an equivalent full level of liquid nitrogen saturated at
one half the maximum allowable working pressure (MAWP). The tank should be
inverted and subjected to an external temperature of 538oC (1000oF) for 20 minutes
without the vessel reaching relief pressure (SAE J2343, 1997).

Insulated pressure vessels for the first four tests have already been delivered to the vessel
manufacturer. These will be tested shortly. Additional plans include the installation of insulated
pressure vessels into demonstration hydrogen-powered vehicles. For this application, the NFPA
(NFPA 57, 1996; NFPA 52, 1998), and CFR-DOT (Title 49, 1996) standards will be reviewed to
prepare the required tests to guarantee the safety of the operation. Future work will also focus on
developing a testing procedure for achieving certification of insulated pressure vessels.
Conclusions
Insulated pressure vessels are being developed as an alternative technology for storage of
hydrogen in light-duty vehicles. Insulated pressure vessels can be fueled with either liquid
hydrogen or compressed hydrogen. This flexibility results in advantages compared to
conventional hydrogen storage technologies. Insulated pressure vessels are lighter than hydrides,
more compact than ambient-temperature pressure vessels, and require less energy for
liquefaction and have less evaporative losses than liquid hydrogen tanks.

For reduced cost and complexity it is desirable to use commercially available aluminum-fiber
pressure vessels for insulated pressure vessels. However, commercially available pressure
vessels are not designed for operation at cryogenic temperature. A series of tests has been carried
out to verify that commercially available pressure vessels can be operated at cryogenic
temperature with no performance losses. All analysis and experiments to date indicate that no
significant damage has resulted. Required future tests are described that will establish that no
significant safety issues exist with pressure vessel operation at cryogenic temperature.
Acknowledgments
This project is funded by the DOE Hydrogen Program, Sig Gronich and Neil Rossmeisl,
Program Managers. The authors also express their appreciation for the significant contributions
of Structural Composites Industries (SCI) to this project. Work performed under the auspices of
the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract W7405-ENG-48.
References
Aceves, S. M., Berry, G.D., 1998, Thermodynamics of Insulated Pressure Vessels for Vehicular
Hydrogen Storage, ASME Journal of Energy Resources Technology, June, Vol. 120, pp. 137
142.
Baur L., 1995, Composite Pressure Vessel with Metal Liner for Compressed Hydrogen
Storage, Proceedings of the 1st IEA Workshop on Fuel Processing for Polymer Electrolyte Fuel
Cells, Paul Scherrer Institut, Villigen, Switzerland, International Energy Agency, Swiss Federal
Office of Energy, September 25-27, 1995, p. 45-69.
Braess, H.H., and Strobl, W., 1996, Hydrogen as a Fuel for Road Transport of the Future:
Possibilities and Prerequisites, Proceedings of the 11th World Hydrogen Energy Conference,
Stuttgart, Germany.
Code of Federal Regulations, Department of Transportation (CFR-DOT), 1996a, Basic
Requirements for Fully Wrapped Fiber Reinforced Aluminum Lined Cylinders, Title 49, CFR
107.105 Standard.
Hettinger, W. Michel, F., Ott, P., and Theissen, F., 1996, Refueling Equipment for Liquid
Hydrogen Vehicles, Proceedings of the 11th World Hydrogen Energy Conference, Stuttgart,
Germany, pp. 1135-1143.
Hull, D., and Clyne, T.W., 1996, An Introduction to Composite Materials, Cambrige
University Press, Cambridge, Great Britain.
Institute of Gas Technology, 1996, Compressed Natural Gas Storage Optimization for Natural
Gas Vehicles, Gas Research Institute Report GRI-96/0364, Des Plaines, IL.
Kaganer, M.G., 1969, Thermal Insulation in Cryogenic Engineering, Israel Program for
Scientific Translation Ltd., Jerusalem, Israel.
Klinger, D., Kuzmyak JR., 1984, Personal Travel in the United States, Vol. 1, 1983-1984,
Nationwide Personal Transportation Study, Report PB89-235378, prepared for the US
Department of Transportation, Office of Highway Information Management, Washington, DC.

Michel, F., Fieseler, H., Meyer, G., and Theissen, F., 1996, Onboard Equipment for Liquid
Hydrogen Vehicles, Proceedings of the 11th World Hydrogen Energy Conference, Stuttgart,
Germany, pp. 1063-1077.
Morgan, R.J. and Allred, R.E., 1989, Aramid fiber reinforcements, in Reference Book for
Composite Technology, Edited by Lee, S.M., pp. 143-166, Technomic Publishing, Lancaster,
PA.
National Fire Protection Association (NFPA), 1998, NFPA 52: Compressed Natural Gas (CNG)
Vehicular Fuel System Code, Quincy, MA.
National Fire Protection Association (NFPA), 1996, NFPA 57: Liquefied Natural Gas (LNG)
Vehicular Fuel System Code, Quincy, MA.
Pehr, K., 1996a, Experimental Examinations on the Worst Case Behavior of LH2/LNG Tanks
for Passenger Cars, Proceedings of the 11th World Hydrogen Energy Conference, Stuttgart,
Germany.
Pehr, K., 1996b, Aspects of Safety and Acceptance of LH2 Tank Systems in Passenger Cars,
International Journal of Hydrogen Energy, Vol. 21, pp. 387-395.
Pentastar Electronics, 1997, Direct-Hydrogen-Fueled Proton-Exchange-Membrane Fuel Cell
System for Transportation Applications, Conceptual Design Report, Report DOE/CE/50390-9,
prepared for U.S. Department of Energy, Office of Transportation Technologies, under contract
DE-AC02-94CE50390.
Peschka, W., 1992, Liquid Hydrogen, Fuel of the Future, Springer-Verlag, Vienna, Austria.
Reed, R.P., and Golda, M., 1994, Cryogenic Properties of Unidirectional Composites,
Cryogenics, Vol. 34, No. 11, pp. 909-928.
Society of Automotive Engineers (SAE), 1997, Recommended Practices for LNG Powered
Heavy-Duty Trucks, SAE.J2343.
Wetzel, F.J., 1996, Handling of Liquid Hydrogen at Filling Stations, Proceedings of the 11th
World Hydrogen Energy Conference, Stuttgart, Germany, pp. 1123-1134.

Figure Captions
Figure 1. Cumulative hydrogen losses in kg as a function of daily driving distance, for
vehicles with 17 km/liter (40 mpg); or 34 km/l (80 mpg) fuel economy, for three
cryogenic hydrogen storage vessels.
Figure 2. Schematic of the experimental setup for temperature and pressure cycling of pressure
vessels.
Figure 3. Pressure as a function of time during the burst test of the aluminum-lined, aramidwrapped vessel. The burst pressure was 94.17 MPa (13657 psig).
Figure 4. Insulation design for pressure vessel. The figure shows a vacuum space, for obtaining
high thermal performance from the multilayer insulation, and instrumentation for pressure,
temperature and level. Dimensions are given in cm.
Figure 5. Insulation design for second-generation pressure vessel. The figure shows a vacuum
space, for obtaining high thermal performance from the multilayer insulation, instrumentation for
pressure, temperature and level, and a vapor shield for reducing hydrogen evaporative losses.