AFP Tank

Composites Part B 253 (2023) 110573
Contents lists available at ScienceDirect
Composites Part B
journal homepage: www.elsevier.com/locate/compositesb
A review of Type V composite pressure vessels and automated fibre

placement based manufacturing
Alexander Air, Md Shamsuddoha 1, B. Gangadhara Prusty *
ARC Training Centre for Automated Manufacture of Advanced Composites (AMAC), School of Mechanical & Manufacturing Engineering, UNSW, Sydney, NSW, 2052,
Australia
A R T I C L E I N F O A B S T R A C T
Keywords: Hydrogen is emerging as a promising future energy medium in a wide range of industries. For mobile applica
A: Carbon fibre tions, it is commonly stored in a gaseous state within high-pressure composite overwrapped pressure vessels
A: Tow (COPVs). The current state of the art pressure vessel technology, known as Type V, eliminates the internal
E: Automation
polymer gas barrier used in Type IV vessels and instead relies on carbon fibre laminate to provide structural
E: Filament winding
Composite pressure vessel
properties and prevent gas leakage. Achieving this functionality at high pressure poses several engineering
challenges that have thus far prohibited commercial application. Additionally, the traditional manufacturing
process for COPVs, filament winding, has several constraints that limit the design space. Automated fibre
placement (AFP), a highly flexible, robotic composites manufacturing technique, has the potential to replace
filament winding for composite pressure vessel manufacturing and provide pathways for further vessel optimi
sation. A combination of both AFP and Type V technology could provide an avenue for a new generation of high-
performance composite pressure vessels. This critical review presents key work on industry-standard Type IV
vessels alongside the current state of Type V CPV technology including manufacturing developments, challenges,
cost, relevance to commercial standards and future fabrication solutions using AFP. Additionally, a novel Type V
CPV design concept for a two-piece AFP produced vessel is presented.
1. Introduction pressure vessels. The high strength of carbon fibre allows for the con
struction of lightweight tanks with gravimetric storage densities of more
To achieve levels of decarbonisation required to meet the Intergov than four times that of a steel vessel [4].
ernmental Panel on Climate Change target of preventing global warming Pressure vessels commonly used within industry today can be clas
greater than 1.5 ◦ C above pre-industrial levels, strong global action to sified into five types [5], as shown in Fig. 1. Type I vessels are the
reduce carbon emissions is being undertaken. Hydrogen as an energy simplest to design and manufacture and make up most vessels in use
carrier has been identified as suitable for meeting these climate objec today. Type II, III and IV vessels are known as composite overwrapped
tives [1]. A key challenge to enabling the use of hydrogen, particularly pressure vessels (COPVs) and commonly utilise carbon fibre to with
in weight-critical aerospace and automotive environments, is storing it stand part or all the structural load and a metal or plastic liner for gas
with high volumetric and gravimetric density. containment. Many ground transport applications use Type IV vessels, as
Various methods for storing hydrogen in solid, liquid and gas form they offer the lightest weight and highest storage densities. Two
are in use around the world, with each method having unique benefits to well-known hydrogen fuel cell electric vehicles, the Toyota Mirai and
relevant industry. However, the storage of hydrogen as a gas has Honda Clarity, both use Type IV vessels with an operating pressure of
emerged as the leading form of hydrogen storage globally [2]. Despite its 700 bar [6,7].
popularity, the storage of hydrogen as a gas is challenging as it is the The structure of a cylindrical Type IV vessel is shown in Fig. 1. The
lightest element and must be held at very high pressure (i.e., 350–700 plastic liner, commonly HDPE or PA [9], is used to prevent gas leakage.
bar) to achieve practical densities [3]. A common method for storing The ends of the vessel feature polar openings, that are fitted with polar
hydrogen to meet these requirements is within carbon fibre composite bosses. The boss is used for filling and emptying the vessel and
* Corresponding author.
E-mail address: g.prusty@unsw.edu.au (B. Gangadhara Prusty).
1
Current Affiliation: BAE Systems Australia.
https://doi.org/10.1016/j.compositesb.2023.110573
Received 23 August 2022; Received in revised form 19 December 2022; Accepted 25 January 2023
Available online 3 February 2023
1359-8368/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
A. Air et al. Composites Part B 253 (2023) 110573
commonly includes access for sensors [10]. The boss on a Type I – III structural health monitoring.
vessel is typically integrated with the liner, however in Type IV and V
vessels it is a separate metal or composite component.
1.2. Existing work on Type V pressure vessels
Type IV vessels are the subject of intense research due to their cur
rent relevance to many industries. However, the state of the art in
Work on Type V vessels began in the aerospace industry as early as
composite pressure vessel technology is known as Type V. A Type V, or
the 1980s [47]. A number of groups have developed Type V tanks for
linerless, vessel does not have an internal liner so the composite acts as
commercial applications or industrial research, however the existing
both the gas barrier and load bearing structure. As the composite is not
peer-reviewed literature available for critical review is highly limited.
overwrapped on a liner in a Type V vessel, they can be referred to as
Several works are presented below on the design and manufacturing of
CPVs. No liner eliminates the need for strain compatibility between the
Type V CPVs.
liner and composite, leading to increased fatigue performance and the
Mallick et al. [48] produced numerous Type V test vessels in various
potential to achieve a 10–20% weight reduction [11,12]. The possible
sizes and evaluated their helium permeability and mechanical perfor
operating pressures of Type V tanks are still not yet fully explored [4].
mance. One design achieved a 15% increase in storage volume and a
Traditionally, COPVs/CPVs are manufactured using filament wind
25% reduction in mass compared with an equivalent Type I tank.
ing (FW), a technique that winds tensioned bands of fibres around a
Another design achieved a pressure of 558.5 bar before helium leakage
rotating mandrel. Whilst this technique is capable and mature, it has
was detected and a burst pressure of 620.5 bar, which is close to the
several inherent limitations. The design of COPVs/CPVs can be
automotive industry standard operating pressure of 700 bar. Jones et al.
improved with the use of more advanced manufacturing methods, pri
[11] also designed and manufactured full scale demonstrator tanks that
marily Automated Fibre Placement (AFP). AFP is an evolution of the
were proof tested to 41.4 bar and pressure cycled 2000 times at 27.6 bar.
traditional filament winding technique and provides high accuracy and
Morimoto et al. [49] prepared two linerless pressure vessels manufac
quality as well as low wastage [13].
tured with quasi-isotropic layup patterns by Fuji Heavy Industry Co. Ltd.
Type V design and AFP technology both have high potential for
The vessels were pressurised to 11 bar at liquid nitrogen temperature
increasing the gravimetric and volumetric density of hydrogen gas
and helium leak tested. Strain gauges and acoustic emission testing was
storage. However, several engineering challenges must be overcome
used to evaluate damage to the vessels. No damage or leakage was
before they can be adopted commercially and offer a reliable alternative
detected at the cryogenic temperature or afterwards.
to Type IV tanks. Despite the research need, there are no existing review
Rafiee et al. [29] investigated burst pressure prediction in lined vs
articles in open literature covering either Type V pressure vessels or
linerless tanks, accounting for stochastic effects present in the
related applications of AFP. This article provides a critical review of the
manufacturing process. The authors found that both the first ply failure
existing work on both Type IV and Type V pressure vessels published
and burst occurred at lower pressures for the linerless tank, with both
from 2000 until present as well as the challenges facing Type V vessels.
Hashin and Hoffman failure criteria able to accurately predict first ply
The manufacturing limitations of filament winding are also presented
failure in the linerless tank.
and the challenges and benefits of replacing the process with AFP are
discussed. Finally, a novel design for an AFP produced linerless tank is
presented, which will form the basis for future work. 2. Traditional manufacturing using filament winding
Except for a small number of fuel tanks for space applications, most
1.1. Existing work on Type IV pressure vessels pressure vessels are manufactured using filament winding. This process
has been in use since the 1940s and while it is recognised for its preci
Type III and IV pressure vessels feature full composite overwraps and sion, automation, and low cost [12], it has several drawbacks (Fig. 2)
have been studied extensively as well as used commercially. Key liter that may limit its flexibility and optimisation potential.
ature related to Type IV pressure vessels is summarised in Table 1.
Notably, much of the published work on Type III vessels is also relevant 1. Filament winding holds the fibre under tension during placement,
to the design of Type IV vessels, however only work specifically relating and as a result can only wrap around convex surfaces, as concave
to Type IV vessels is covered here to control the scope of this review. surfaces would result in fibre bridging. Thus, a polar boss, or end
Each work has been categorised based on its content. For example, if a fitting, is required at each end of the vessel for the fibre band to wrap
publication is listed in the ‘Materials’ category, then it contains specific around and return when it reaches the dome ends. These metal polar
material related outcomes, not simply a mention of the materials used. bosses complicate the design and add mass.
From reviewing the literature, the mathematical theory (referring to 2. The fibre band is required to be continuous throughout the winding
papers that present governing equations and mathematical theory) and process, which causes excess thickness build-up around the polar
analysis of pressure vessels is well covered, and many works provide boss. This can violate basic shell theory assumptions used in stress
experimental data on the coupon and/or full vessel scale. However, few calculations [50]. It also results in excess fibre, that is required for
works address the design of the polar boss and liner in detail or dome reinforcement, being placed on the cylinder region [51]. The
Fig. 1. Five common types of pressure vessel (reproduced with permission from Ref. [8]).
2
Table 1
Summary of previous work on Type IV pressure vessels, categorised by content.
Author Country Year Mathematical Theory Simulation Experimental Data Materials Boss Design Liner Design Health Monitoring
Alves et al. [14] Brazil 2022 X X X

Wang et al. [15] China 2022 X X
Rahul et al. [16] India 2022 X X
Sharma et al. [17] India 2022
Rafiee et al. [18] Iran 2022 X X X
Nebe et al. [19] Germany 2022 X X X
Su et al. [20] China 2021 X X
Schakel et al. [21] Germany 2021 X
Park et al. [5] Korea 2021 X X X
Munzke et al. [22] Germany 2021 X X
Jois et al. [23] Germany 2021 X X X
Zhang et al. [24] China 2020 X X X
Souza et al. [25] Brazil 2020 X X
Sapre et al. [26] India 2020 X X X
Alam et al. [27] USA 2020 X X
Nebe et al. [28] Germany 2019 X X
Rafiee et al. [29] Iran 2018 X X X
Cho et al. [30] Korea 2018 X X X
Johnson et al. [31] USA 2017 X X X X X
Hua et al. [10] USA 2017 X X X X
O Brádaigh et al. [32] Scotland 2016 X X X X
Shao et al. [33] Japan 2016 X X X
Yamashita et al. [34] Japan 2015 X X
Magneville et al. [35] France 2015 X X
Leh et al. [36] France 2015 X X X
Leh et al. [37] France 2015 X X
Berro Ramirez et al. [38] France 2015 X X X X
Leavitt et al. [39] USA 2014 X X X X X
Gentilleau et al. [40] France 2014 X X X
Roh et al. [41] USA 2013 X X
Liu et al. [42] China 2012 X X
Bertin et al. [43] France 2012 X X
Villalonga et al. [44] France 2011 X
Villalonga et al. [45] France 2009 X X X X X
Funck and Fuchs [46] Germany 2001 X
Fig. 2. Limitations of using filament winding for pressure vessel manufacturing (reproduced with permission from Ref. [52]).
designer does not have the freedom to tailor the thickness profile of 75–300 mm [54]) at high speeds on relatively simple surfaces, making it
the domes, instead, it is coupled to the thickness in the cylinder re useful for large open moulds such as aircraft wing skins. However, it
gion and vessel geometry. performs poorly on complex contours or internal geometries due to the
3. As they are placed under tension, the fibres self-align to geodesic build-up of wrinkles in the wide tapes [53].
paths. To achieve non-geodesic paths, friction is required to prevent To overcome these limitations, AFP was introduced, with develop
the fibres from slipping [12]. Thus, non-geodesic winding capabil ment beginning in the 1980s [55]. AFP uses a narrower tow than ATL
ities are limited to the available friction, restricting the design space. (typically 3.175–12.7 mm [54]) and incorporates elements of both
filament winding and ATL [13,53] to allow for placement on curved
3. Automated fibre placement surfaces and orientation of fibres in varying directions. However, this
flexibility is often at the sacrifice of reduced layup speed. A distinct
Automated fibre placement (AFP) and automated tape laying (ATL) benefit of the AFP process over filament winding and ATL is its ability to
are robotic composites manufacturing techniques that are considered perform tow steering. Tow steering involves placing fibres on
the most advanced and widely used automated composites non-geodesic trajectories to achieve full coverage of a surface and enable
manufacturing methods available [53]. ATL lays wide tapes (typically optimised and variable stiffness [56]. During the AFP process (schematic
3
shown in Fig. 3) the incoming tow is delivered to a compaction roller capability of tape placement techniques that can be achieved using
from a set of spools. The tow is heated using a torch or other heat source dedicated AFP programming software. Most other existing applications
and compacted by the roller to promote adhesion to existing layers. At of AFP to pressure vessels have been on tanks larger than those for
the end of a length of tows (known as a course), a knife cuts the tape, and automotive applications, where the scale of filament winding equipment
the process can restart. Most AFP robots can place multiple tows required may be impractical or unable to meet the design requirements.
simultaneously. Several authors have performed recent and compre In particular, large-scale tanks have been studied for the transport of
hensive reviews on the AFP process, its characteristics and defects, cryogenic space fuels such as LOx and LH2 in rockets or similar vehicles.
prospects, and limitations [13,53,55,57,58]. Doyle et al. [64] proposed the design of a cryogenic fuel tank made
using thermoplastic CF/PEEK. The cylinder region of the tank was
produced using ATL, and the domes via hand layup. Limited technical
3.1. Applications to pressure vessels detail was provided, but sample flat panels were manufactured to
evaluate void content and basic strength properties. An amorphous
Existing applications of AFP to pressure vessels have been limited, interlayer bonding technique was proposed to join the domes to the
with details in publications often omitted to protect intellectual prop cylinder. This used a polyetherimide film at the bond interface that melts
erty. The US DoE (Department of Energy) Hydrogen Program performed when the parts are forced together under heat, creating a structural
an investigation on Type IV pressure vessels made partially using AFP bond.
[39,51,60,61]. The approach used AFP to manufacture the vessel end
domes, which were subsequently placed on a mould and overwrapped 3.2. Potential benefits of AFP to pressure vessels
using filament winding to produce a complete vessel. The project aimed
to use the cut-and-add capability of AFP to provide selective reinforce 3.2.1. Variable angle tow and fibre steering
ment to the dome areas, without adding excess material to the cylinder, AFP allows for both non-geodesic and steered paths to be produced,
which is unavoidable during filament winding as the fibre band must providing the opportunity for greater optimisation of the fibre layup
traverse the cylinder to move from one dome to the other. Numerous than in FW. This capability is referred to as variable angle tow (VAT).
prototype vessels were made over the 6-year project and studies were Work by Rouhi et al. [65] demonstrates the benefits of VAT well. A
conducted to evaluate the benefits of the hybrid AFP/FW tank at a mass cylinder was manufactured with 50% of the plies VAT (and 50% using a
production scale (500,000 units per year). One design showed a 32% constant angle), a 28% increase in bending-induced buckling load was
weight saving over the entirely filament wound baseline tank. calculated compared with a quasi-isotropic layup, which translated to an
NASA and Boeing collaborated on the Composite Cryotank Tech experimental improvement of 18.5%.
nologies and Demonstration (CCTD) Project. Two sample vessels were VAT has also been applied to pressure vessel design to eliminate
constructed with diameters of 2.4 m and 5.5 m. Both vessels were pro bending stress in the vessel wall, which typically occurs at the dome/
duced from carbon fibre/epoxy material using AFP, manufacturing of cylinder interface. Daghighi et al. [66] demonstrated the possibility of
the larger 5.5 m tank is shown in Fig. 4. A combination of 3.125 mm (1/ producing a bend-free composite pressure vessel using a super ellipsoid
8′′ ) and 6.35 mm (1/4”) tows were utilised, with the thinner tow used on of revolution. The design achieved uniform through-thickness stress
the curved dome regions. Weight and cost reductions of 25% and 30%, distribution in the composite and decreased external dimensions by
respectively were achieved compared with a traditional aluminium- using a low-profile elliptical dome. Several geometric and material
lithium tank. The tanks were successfully pressure cycled using cryo induced limitations were identified that constrain the possible super
genic LH2 at 9.3 bar and passed a variety of tests including inspection via ellipsoid shapes. The same authors have performed further work on
acoustic emission and helium leak testing. The project is detailed in the thermo-mechanical loading [67] and calculation of allowable internal
book chapter by McCarville et al. [62]. pressure using composite failure criterion [68] for the bend-free design.
Laser assisted tape placement (LATP) using thermoplastic compos Fan et al. [69] also derived the governing equation for an elliptical
ites has also been applied to COPVs/CPVs in a small number of studies head free from bending stress. Numerical simulation was performed on
[44,63]. Schakel et al. [21] present the most detailed work on this topic, vessels with elliptical heads with five different aspect ratios and a cy
which is discussed in Sections 3.3 (Challenges in Applying AFP to lindrical centre section, unlike Daghighi et al. who considered an
Pressure Vessels) and 5.1 (Materials used in Pressure Vessel Construc ellipsoid. The suppression of bending stress was verified in the simula
tion). Schakel et al. designed their tank using Composicad, a filament tion and a reduction in stress on the dome side of the dome-cylinder
winding software so the design did not fully utilise the cut and restart interface of 13.7% and 34.9% was found when compared with an
isotropic material dome and constant fibre angle dome respectively. The
authors also noted that the ideal fibre angle on the cylinder and dome
was different, creating an angle compatibility issue, as shown in Fig. 5.
This was not corrected for in simulation and would pose a
manufacturing challenge to smoothly transition between these regions.
While the work on bend-free pressure vessels may influence future
designs, it has not yet been demonstrated experimentally. The bend-free
design also does not account for the need to install a polar boss on at
least one end of the vessel, which will disrupt the required fibre pattern.
Additionally, the steering radii required to achieve the necessary fibre
angles may exceed the steering capability of current AFP equipment.
The tight steering radii required can be seen on the dome centres in
Fig. 5. One potential solution to this is ‘Continuous tow shearing’ (CTS).
Kim et al. [70] used a bespoke fibre placement head to steer fibres at
radii as tight as 30 mm, a considerable improvement over the traditional
minimum of around 500 mm, by taking advantage of the intra-fibre
shear properties. However, this technique is still in its infancy, with
low layup speeds up to 3 mm/s achieved during the study.
Fig. 3. Automated fibre placement process (reproduced from Ref. [59] under
CC BY 4.0 license).
4
Fig. 4. AFP manufacturing of the 5.5 m diameter cryotank as part of the CCTD project (reproduced from public domain figure [62]).
designed and constructed a Type III pressure vessel using doily layers for
dome reinforcement. The addition of doily layers was found to increase
the burst pressure by 29% and move the failure location from the dome
to the cylinder, which is considered safer.
3.3. Challenges in applying AFP to pressure vessels
3.3.1. Defect control

When compared with filament winding, AFP is less mature and used
on fewer commercial products. The quality of an AFP part is highly
dependent on a large number of variables [13]. For example, the size of
the roller must be matched to the part curvature to ensure good
Fig. 5. Bend-free fibre angle mappings for various aspect ratio domes by Fan
compaction of the tows. If not matched correctly, the precision of the
et al. [69] (reproduced with permission).
end cutting operation can be reduced and defects such as tow slipping,
or bridging may occur [72–74]. Oromiehie et al. [13] and Belnoue et al.
3.2.2. Selective reinforcement [75] have extensively reviewed AFP defects, their creation and impact
Selective reinforcement is a second key benefit of using AFP to on performance. Fig. 7 depicts four of the most common defect types.
construct pressure vessels, as it can directly address the second shortfall Gaps and overlaps are one of the most frequently encountered defects
listed in Section 2. Apart from the US DoE Hydrogen Program work between tows and lead to non-uniform thickness in the region of the
discussed previously, there is no literature quantifying the possible im defect [13]. Croft et al. [76] found that gaps led to a reduction in
provements that can be achieved with AFP specifically. However, local strength in tensile, compressive and open hole tensile testing.
reinforcement using a doily provides a good indication of potential Conversely, overlaps were found to increase strength in each of these
improvements. tests. Woigk et al. [77] performed similar work with AFP defects simu
For example, Roh et al. [41] performed FEA optimisation of a Type lated using hand layup. The authors found that gaps or overlaps on their
IV vessel with selective reinforcement via doilies in the dome. The doily, own did not have a statistically significant impact on tensile or
illustrated in Fig. 6, consisted of carbon fibre ‘strips’ that were placed compressive strength, but when combined they had a negative impact.
between helical layers on the dome with a fibre direction close to 90◦ , Another important defect to examine, not pictured above, is the void
providing hoop reinforcement and allowing the thickness of the helical content of AFP laminates. Oromiehie et al. [79] investigated the porosity
layer to be reduced. The authors reported a theoretical reduction in the of AFP-made laminates using microscopy and neutron tomography.
amount of composite material required of nearly 10%. Kartav et al. [71] Increased porosity was shown to decrease the interlaminar shear
strength. The AFP processing parameters were also shown to affect the
void content, in particular the compaction force and temperature were
found to have a significant and non-linear effect.
3.3.2. Physical limitations

Physical limitations in the AFP process also pose a challenge for
pressure vessel manufacturing. The large size of current AFP heads
means that collision with the part or tooling can occur if improperly
considered. On rocket-scale tanks used in space applications this is less
of an issue since as the tank diameter is increased, the size difference
between the AFP head and the part and tool also increases, providing
relatively more space for the head to move in. Smaller tanks also
necessitate more rapid head movements due to the lower radius end
domes, which can be problematic for the robot’s controller. On a small
scale tanks, i.e., for automotive applications, Schakel et al. [21] (pre
viously discussed in Section 3.1) present the only work that overcomes
this challenge. The authors produced several 304 mm diameter GF/PA6
Fig. 6. Schematic of doily reinforcement in the dome (reproduced with sample tanks with a PA6 liner using the Fraunhofer “Multi-Material-
permission from Ref. [41]). Head” with a laser assisted tape placement (LATP) setup. As shown in
5
Fig. 7. Common AFP defects (reproduced from public domain figure [78]).
Fig. 8. LATP manufacturing of a small diameter CPV (left) (reproduced with permission from Ref. [21]) and typical thermoset AFP head (right).
Fig. 8, the compaction roller had a large diameter and was cantilevered,
allowing tape to be placed very close to the headstock, as done in fila
ment winding, without collision. These features are distinctly different
from a regular, off-the-shelf thermoset placement head, pictured to the
right in Fig. 8, that supports the roller on either side. It is to be noted that
the LATP head is placing a single 12 mm wide tape and the thermoset
head is placing 4 × 6.35 mm wide tapes. From visual inspection of Fig. 8,
the thermoset head would be incapable of placing material tangent to
the headstock without collision, as achieved by the Multi Material Head
– creating a challenge for manufacturing with most commercially
available AFP hardware. Without modifying the placement head, this
challenge may be addressable by innovating the tooling and support
structure used to hold the vessel.
Qu et al. [80] presented a simulation method to evaluate the
accessibility of the AFP head to a placement surface. Their method
considered the ability of the roller to both access the part surface (po
sition space) and place the roller normal to the surface (altitude space).
The possible position and altitude space was evaluated for a specific AFP
setup and used to verify the accessibility of a sample tool surface.
In some cases, complex geometry may be impossible to manufacture
Fig. 9. Y-shaped tubular geometry used by Hély et al. [81] (reproduced
using current AFP equipment. Hély et al. [81] performed a simulation of with permission).
the AFP layup of a complex y-shaped tube, shown in Fig. 9. It was found
that the geometry would be very difficult to cover with conventional
6
AFP equipment as the highly curved fibre paths required would lead to
tow buckling. The authors noted that CTS could be capable of over
coming this issue. Further, there were numerous areas where no
coverage could be achieved due to limitations in the range of motion of
the robot joints. The authors also assumed that the placement head was
small enough to avoid collision, which would not hold true in a
real-world manufacturing scenario. Even if collision difficulties are
overcome, complex dome curvature and three-dimensional head
movements will limit the placement speed. While AFP layup rates of up
to 150 kg/h are possible on simple geometry, as the complexity increases
it may be necessary to slow layup speeds, or use shorter courses or
narrower tows [82]. Filament winding is less affected by this phenom
enon, so differences in manufacturing time between AFP and FW should
be considered alongside other benefits and drawbacks.
These manufacturing challenges can be addressed with a deep un
derstanding of the AFP process using a digital twin. Digital twins merge
the physical process with computer-based simulation to enable smart
manufacturing and complex process control [83,84]. Zambal et al. [85]
built a digital twin of a dry fibre AFP process. Real time data was
collected on defects and other relevant information. The model was able Fig. 10. CCTD Project 2.4 m tank layup tool (reproduced from public domain
to calculate the effect of defects in using finite element analysis as they figure [62]).
occurred. The same authors have also proposed a probabilistic model for
image-based defect detection in the AFP process using artificial intelli
4.2. Permeation
gence [86]. Sun et al. [87] have reviewed a wide range of online defect
detection techniques used in AFP manufacturing.
FRP laminates are known to exhibit high permeability to numerous
gases and in particular hydrogen, due to its small molecular diameter
4. Development areas for Type V pressure vessels
and high kinematic viscosity [15,88,89]. This is a property that has thus
far confined linerless pressure vessels to low pressure operation [9],
Type IV pressure vessel technology is mature and actively used in
where leak rates are lower, and cryogenic storage, where the hydrogen is
commercial applications. However, Type V pressure vessels are pres
in a liquid state. Additional challenges are encountered when using AFP,
ently at a lower stage of technology readiness, demonstrated by the lack
as AFP/ATL made laminates have shown increased permeability
of available literature. Two challenge areas have been identified that
compared with hand layup [64]. The US Department of Energy
must be addressed before mass adoption can be achieved:
Hydrogen Program has specified an ultimate leakage target of 0.05
manufacturing and permeation control. Each of these areas is discussed
(g/hr)/kg of stored hydrogen gas to help guide researchers in their work
below.
towards minimising permeability [90]. Similarly, Adams et al. [91]
proposed a maximum allowable permeation rate of 8.0 (mL/hr)/L of
4.1. Manufacturing water capacity at 20 ◦ C. This permeation is dominated by two phe
nomena, shown in Fig. 11 (a).
In either the filament winding process or AFP, a mandrel must be
used to wind the fibre around. In the case of a Type II-IV tank, the in • Diffusion causes the pressurised gas to adsorb into the vessel wall,
ternal liner is used as the mandrel and is left inside the vessel once diffuse through the material and desorb into the atmosphere.
winding is complete. For a Type V tank, the mandrel must be removed to Humpenöder [92] has published details on the mathematical and
achieve the linerless design. A traditional approach is to use a soluble conceptual background of gas diffusion in composite materials.
mandrel that can be removed from the tank using water. Sand, plaster • Interconnected microcracking in the matrix opens ‘channels’ for the
and PVA have all been used for this purpose [12]. Jones et al. [11] noted gas to escape through, as depicted in Fig. 11 (b).
that these cores are not always feasible due to limitations in material
compatibility and difficulty in removing the dissolved core depending While both mechanisms contribute a measurable amount to the total
on the polar opening size. Instead, the authors used a 2-piece composite permeation, leakage due to microcracking has been shown to be several
shell for the mandrel. This was constructed by filament winding over orders of magnitude higher than that due to diffusion [94–96]. In some
solid tooling and cutting the part in half to remove the tool, before mathematical models diffusion is even ignored [97]. The permeability is
rebonding the halves together. Producing a composite mandrel in two affected by multiple material and environmental factors including
halves has benefits including easier installation of components inside the temperature, pressure, matrix properties, fibre distribution, fibre type
tank as well as ease of inspection. and ply stacking sequence [20,98,99]. However, material based prop
The CCTD project [62], previously discussed in Section 3.1 (Appli erties have been shown to change permeability by less than an order of
cations to Pressure Vessels), also used solid tooling. A segmented tool magnitude [100].
(Fig. 10) was constructed in place of a dissolvable mandrel. The 2.4 m There is also little understanding of the permeation in polymers at
diameter demonstrator tank used 24 solid composite segments, with a high (i.e., 700 bar) pressure environments [20]. Permeation measure
split at the equator, that were fastened together to produce a complete ment is often performed at low pressures due to a lack of high-pressure
mandrel. Joint faces and critical geometry were match machined and testing equipment. Fujiwara et al. [101] constructed and validated a
indexed off coordination features to ensure accurate assembly. To 900 bar permeability cell to solve this problem, however, only polymers
remove the tool after AFP layup, segments were unfastened from inside were tested and not composites. Typical permeation cells also measure
the tank and extracted through the polar opening. The opening on the the permeability with the sample unloaded. This means that most cracks
2.4 m tank had a diameter of 457 mm, allowing a human worker inside are closed during the measurement, which is not representative of the
to remove the segments. This approach is well suited to large tanks, operating conditions of a pressurised vessel. Laeuffer et al. [102,103]
however, has not yet been demonstrated on a smaller scale tank such as have addressed this using a novel test device to measure the
those used in automotive applications.
7
Fig. 11. (a) Leakage mechanisms and (b) detailed mechanical leakage mechanism due to intersecting matrix microcracks (reproduced with permission
from Ref. [93]).
permeability of a pressurised cylindrical sample. modelled. Typical fracture mechanics and strength-based methods fail
Matrix microcracking is caused by mechanical and thermal loading to satisfactorily model microcracking since the cracks tend to form
of the composite laminate and occurs through the thickness of a ply and instantly instead of growing like a traditional crack. Work on modelling
parallel to the fibres [88,104]. The build-up of such cracks can open up this phenomenon can largely be divided into two areas: experimental/
‘channels’ through the wall of the vessel that allow the gas to escape, this empirical methods and numerical methods.
is often considered the first mode of failure for a composite tank [47,89]. Nairn [104] has presented an empirical finite fracture mechanics
While individually these cracks will not cause the vessel to catastroph method. The energy-based model determined that a new microcrack is
ically fail, they result in an overall deterioration of the mechanical formed when the total amount of energy released in the formation of the
properties of the laminate and may grow to cause complete failure of the crack reaches a critical level denoted the microcracking fracture
tank [104]. Thus, it is greatly important for designers to understand the toughness (MFT), or Gmc , which is a material property. Composite MFT
formation of these cracks and how they affect the tank’s performance. ranges greatly, from 200 to 3000 J/m2 for different reinforcement and
matrix combinations, so it is necessary to perform comparative studies
4.2.1. Experimental evaluation of gas permeability in fibrous composites when selecting a material for a particular application. Several authors
Several authors have studied the gas permeability of composite have experimentally determined the MFT [47,110–112], in particular
laminates. Much of the existing work has been performed for permeation Mallick et al. [47] presents the experimental process well. Additionally,
at cryogenic temperatures, however many publications also include re Prasad et al. [113] have recently reviewed semi-empirical methods for
sults at room temperature. The gas leak rate has been found to increase permeability modelling in composites that combine experimental data
with mechanical strain, crack density and decreasing temperature as with computational methods.
well as vary with crack intersection angle [105–108]. Layup sequence A large proportion of gas permeability modelling in composites has
also effects permeability, laminates where plies of the same angle are been numerical, utilising finite element methods. Bois et al. [114]
grouped rather than dispersed through the thickness have shown developed a multiscale model for predicting the gas flow path through
increased leakage [106]. Additionally, the composite fabric type im damaged composites subject to in plane loading. The model was appli
pacts permeability, with stitched materials exhibiting higher perme cable to CPV’s and, if combined with a flow model, could predict global
ability than unstitched materials [109] and laminated composites leakage rate based on damage density, cracking opening, and fluid
exhibiting higher permeability than textile composites [106]. properties. Ren et al. [115] also developed a multiscale approach for
Flanagan et al. [95] investigated the permeability of thermoset and predicting matrix cracking in composite pressure vessels under both
thermoplastic composites manufactured using both regular hand layup uniaxial and biaxial loading using in situ matrix properties at cryogenic
and autoclave techniques as well as thermoplastic ATL. Permeability temperatures. The predictions of the model agreed with experimental
values were compared with pure PEEK and PVC. All samples made using results.
autoclave curing exhibited near-Fickian behaviour. This is an interesting Yuan et al. [116] produced an FEA model for permeability predic
observation as Fick’s Laws describe diffusion through homogenous tion. The model successfully predicted permeability for three different
materials, and composites are not homogenous. However, samples made specimens at room and cryogenic temperature. Xu et al. [117] also built
using ATL did not follow Fick’s Laws and exhibited a significant coef a finite element based method for permeability prediction in cross-ply
ficient of variance of 70% in the leak rate. The authors attributed this to laminates. Permeability was found in terms of crack density, opening
the presence of microcracking, debonding and other discontinuities in displacement and an experimentally determined constant. Crack density
the ATL laminate that were identified using micrographs. Additionally, and opening displacement were found to have a significant effect on
the permeability of thermoset and thermoplastic composites was found permeability as they determine the intersection area. Crack density was
to be on the same order of magnitude. found to have a greater influence than opening displacement. O
Brádaigh et al. [32] used x-ray scanning and microscopy to map crack
4.2.2. Modelling gas permeability in laminated composites growth from cryogenic cycling. This data was fed into a combined XFEM
As with diffusivity, microcrack susceptibility can be measured and cohesive zone model for future microcrack prediction. More damage
8
was detected in thick sections of the laminate compared to thin, with vinyl-ester COPVs and found a the vinyl-ester vessel had a 20% higher
residual stress suggested as a reason for this. burst pressure. The authors also found the vinyl-ester matrix to be less
Most recently, Ebermann et al. [8] compared analytical and nu prone to intralaminar cracking than epoxy.
merical approaches to determine the effective permeability of a com The matrix material has a significant effect on the thermomechanical
posite material. The laminate properties were homogenised using two performance of the tank. Particularly for hydrogen applications, large
analytical methods (Weiner bounds, Hashin-Shtrikman bounds) and temperature changes can occur during the fast-filling process [124]. This
numerical FEA calculation of a representative volume element and the leads to thermomechanical stresses in the vessel wall. Failure in com
results compared. Numerical and analytical results agreed well, with posites has been shown to occur more easily under thermal loading than
diffusion shown as the major driver of permeation in an undamaged mechanical loading [125], with only a small number of cycles required
laminate. However, as soon as a small threshold volume (<1%) of cracks to initiate thermomechanical fatigue damage [126]. This damage can
was introduced, permeation increased significantly. Above this degrade strength and stiffness even at small variations from room tem
threshold, crack volume had little effect on permeability. perature [127]. Thermomechanical effects can be minimised by mini
mising the volume-to-surface-area ratio of the tank [128], this favours
5. Materials and material enhancement smaller scale tanks. So, the tank size and duty cycle should be considered
during the material selection process.
5.1. Materials used in pressure vessel construction Degradation due to thermomechanical stresses are particularly
relevant for epoxy matrices [129], so thermoplastic materials have also
5.1.1. Fibres been investigated. Benefits such as increased fracture toughness [130],
Composite pressure vessels are usually reinforced with carbon, glass repeated formability, weldability, repairability, ease of recycling and
or aramid fibre, or a hybrid [12]. However, most composite pressure lower manufacturing times [21] are attractive for use in Type V tanks.
vessels use carbon fibre reinforcement for its high strength and stiffness Both AFP and filament winding can be used with either thermoset or
characteristics. Unidirectional fibre is used almost exclusively in thermoplastic materials. The application of in-situ consolidated (AFP
COPV/CPV construction as it is compatible with FW and AFP and FW) thermoplastic materials has been recently reviewed by Boon
manufacturing and provides the highest strength along the fibre direc et al. [59].
tion. In some cases woven fabric materials have been used, however this Schakel et al. [21] designed and tested a thermoplastic pressure
is typically in vessels where weight is not critical, due to their inferior vessel using laser assisted tape winding (LATP). The LATP process was
strength when compared with unidirectional tow [50]. characterised during manufacturing. The domes were found to pose
A major consideration when selecting a fibre is the size effect. Hwang manufacturing challenges due to high fluctuations in tape feed rate and
et al. [118] investigated the size effect of fibre strength on vessel per variation in the laser substrate area. Each dome (headstock and tailstock
formance. As mechanical properties are closely linked to defects in the in Fig. 8) was affected differently due to the different position of the
material, a strength reduction can be expected with increasing compo robot arm affecting the feed rate. The feed rate was also found to vary
nent size. A 16–32% reduction in delivered fibre strength was predicted significantly from the simulated values.
and experimentally validated when scaling from a fibre strand test to a Villalonga et al. [44] have constructed a filament wound 700 bar
filament wound pressure vessel, highlighting the importance of under thermoplastic vessel using a PA6 liner and resin system. The vessel was
standing the size effect when performing coupon level tests. Hwang et al. tested for commercial use, however minimal details were provided in
[119] also developed the ring burst test. This test involves loading a FW the published article. Höck et al. [63] designed and manufactured two
ring that is representative of the hoop layer on a pressure vessel until pressure vessels using LATP for aerospace applications. The prototype
failure with a mechanically simulated hoop stress. Compared with tanks had a diameter of 1.3 m and a polar opening radius of 175 mm.
traditional tensile testing, this method captures the FW process param The authors did not provide any technical detail on the working pressure
eters and yielded failure strain results identical to a full-scale vessel, of the tanks or the substance that they were designed to hold. O Brádaigh
within the 5% significance level. Additionally, Cohen et al. [120,121] et al. [32] presented the design of a thermoplastic fuel tank for space
presented the effect of fibre volume fraction (FVF) on the strength of a applications. A moulded PEEK liner was produced, and segments were
FW vessel. The ultimate strain-to-failure was found to increase with extracted for permeation testing. Some liner segments were also over
increasing FVF in the hoop plies, which directly relates to the failure wrapped using LATP and cryogenically cycled for measurement of crack
pressure. growth.
Whilst traditional fibres remain the most popular option, sustainable An important consideration when applying thermoplastic materials
and natural fibres have also been investigated for composite pressure to AFP is temperature control and distribution. Zaami et al. [131] con
vessels. They have been previously applied to low pressure applications structed a numerical model for simulating the temperature of the ther
for CNG storage [122] and have recently been studied for high pressure moplastic during the LATP process. The vessel scale and dimensions
usage. Bouvier et al. [123] investigated alternative fibre choices for were found to affect the temperature. With an increasing tank radius,
Type IV COPVs. The authors presented different fibre options based on liner temperature (for a Type IV tank) decreased and cooling time
several optimisation parameters (i.e., mechanical performance, cost, increased, which should be considered in a mass production scenario.
recyclability). An E-glass/T700S carbon hybrid yielded the lowest cost, Larger diameter tanks also had a higher nip point temperature in the
whilst basalt/recycled T700S and flax/recycled T700S hybrids are tape and substrate. Clearly, process parameters should be adjusted based
promising for reducing greenhouse gas emissions for 700 bar and 350 on the vessel scale.
bar vessels respectively. Notably, an entirely T700S vessel remained the
obvious choice for maximum mechanical performance. 5.1.3. Curing
Composite cure cycles affect the final properties of a composite
5.1.2. Matrix materials pressure vessel. Of particular importance for Type V CPVs, the cure cycle
The majority of existing COPVs/CPVs designs are manufactured impacts microcrack formation. For cures at elevated temperature, a
using a thermoset matrix, as they are easier and more reliable to produce stress-free temperature exists, typically close to the cure temperature.
[21]. Specifically, for works listed in Table 1 that utilise thermoset Deviations in the part temperature (i.e., as it cools during cure, or during
materials and provide material information, all but two used an epoxy operation at a given temperature) from this stress-free temperature
matrix. Johnson et al. [31], used a nanomodified vinyl-ester resin system induce thermal stresses that can lead to microcrack formation and hence
that showed a 5–7% weight reduction compared with epoxy, due to its an increase in gas permeability. These stress residuals exist due to a
lower density. Additionally, Shao et al. [33] compared epoxy and mismatch in the coefficient of thermal expansion between the fibres and
9
matrix [132]. The effect of cure temperature on crack density has been failure was observed.
investigated by Timmerman et al. [133]. The authors varied the cure
temperature of CF/epoxy laminates cured in an autoclave. Increasing 5.2.2. Improvement using nanomodifications
the temperature from 70 ◦ C to 180 ◦ C caused the crack density to in A second important method for reducing the microcracking in a
crease from 10 to 35 cracks/cm2, as the higher cure temperature composite is through nano modifications and toughening. Beaumont
resulted in a higher stress-free temperature. et al. [144] have published a detailed discussion on the mechanisms,
A wide range of curing techniques are available for composite challenges and opportunities surrounding nanoengineered composites.
pressure vessels. For thermoset materials made using both FW and AFP, The use of carbon nanotubes for fracture toughness modification of
autoclave and out-of-autoclave (OoA) processing remains popular. In CFRP has also been extensively covered and recently reviewed by Zhang
the context of reducing the stress-free temperature, OoA curing can et al. [145]. Relevant work using other forms of nanomodification is
often be performed at lower temperatures than autoclave curing. OoA presented below in Table 2. Key results alongside technical details are
processes involve curing in the absence of external pressure from an presented for various works. Overall, resin nanomodifications are highly
autoclave, often using a regular oven and vacuum bag only [134]. OoA effective at improving material properties that reduce microcracking
curing is commonly used for large tanks that cannot fit inside an auto and permeation such as fracture toughness. Existing work is heavily
clave, this is of keen interest to aerospace applications working with focused on the improvement of epoxy resin systems, which remain
tanks on the scale of a rocket or plane. For this reason, an OoA process highly relevant to the field. Many works also mix the nanomodification
was utilised in the CCTD project to cure both demonstrator tanks. into the resin and introduce it to the fibres using a wet layup or infusion
To further reduce the cure temperature, resin infusion or resin technique, this is convenient in a laboratory setting but is not entirely
transfer moulding processes can be used. In filament winding, wet applicable to AFP, which typically uses prepreg materials.
winding remains the most popular and cost-effective technique [12]. The reduction in microcracking achievable with nanomodifications
Wet winding involves passing dry fibre through a resin bath immedi is well depicted by Islam et al. in Fig. 12, which compares micrographs
ately before winding onto the mandrel. This requires a standard curing of composites modified with nCuO (as investigated by Change et al.) and
process that typically involves elevated temperatures. However, dry polydopamine coated nCuO. Other notable results include those of
winding is in the early stages of development [135], which eliminates Haight et al. Their resin system has been developed into a commercially
the resin bath and instead adds a secondary infusion or moulding pro available product and demonstrated on full scale composite pressure
cess. These processes can be performed at room temperature, providing vessels. Additionally, the highly effective hydrogen permeability
an interesting opportunity to lower residual stress induced micro reduction achieved by Wei resulted in permeability values lower than
cracking. Dry fibre placement is also available for AFP, where it is also in those of common Type IV COPV liner materials including HDPE, LDPE,
early stages of development. Veldenz et al. [136–138] have performed PA, and PTFE.
detailed work on material selection for, and manufacturing of dry fibre
AFP components. 5.2.3. Next steps for permeation reduction
Clearly, several methods exist for effectively improving the fracture
5.2. Material enhancement to reduce permeation toughness of composite laminates and reducing hydrogen permeability.
However, existing work is largely on a sample scale, with system level
Several methods exist for improving the permeability of composites. evaluation required. At the tank level, permeation should be measured
Increased FVF has been shown to reduce permeability by increasing the with the US DoE units of (g/hr)/kg of stored hydrogen, allowing for
tortuosity of the laminate [8,139]. Tortuosity refers to the effect that the direct comparison with target leakage rates. Additionally, the scalability
fibres have on physically blocking the flow of the gas, essentially of manufacturing processes used to achieve these improvements must be
creating a maze. Tortuosity is also increased by using randomly oriented evaluated on a tank level. For thin ply materials, the additional cost
or chopped strand fibre [8,98], however unaligned fibres are infeasible associated with the extra manufacturing time and areal fibre require
for the high loads in a pressure vessel. Popular material modifications ment needed to produce a desired thickness should be investigated. For
are presented below. example, using a thin ply tow that is half the thickness of a regular tow
could double placement/winding time as twice the number of layers
5.2.1. Improvement using thin ply materials would be required to achieve equivalent thicknesses. For nano
Thin ply laminates have a ply thickness that is smaller than typical modifications, the method of introducing the modification may work on
material options, their manufacturing, properties and usage has been a sample scale, but not be suitable for filament winding or AFP. For
reviewed by Galos [140]. Hamori et al. [141] have evaluated thin ply example, spray coatings are harder to control over large, curved surfaces
laminates as a method to reduce permeability. In this study a ply than on flat samples, and methods where the nanoparticle is mixed with
thickness of 0.06 mm was used. The permeability of a [04/904]2S the neat resin may not be applicable to AFP, which primarily uses pre
cross-ply laminate (total thickness 1.8 mm) using regular thickness preg materials, although dry fibre AFP is possible [137,155,156]. Lastly,
layers was compared to a [04/90404/904/0/90]S laminate (total thick existing work has largely been applied to samples made using hand
ness 2 mm), where the centre four plies were thin, under biaxial loading. layup techniques. Results may vary when applied to AFP produced
The sample with the thin plies demonstrated a several order of magni laminates due to common defects and the generally lower technology
tude reduction in permeability against helium gas, at equivalent strain readiness level of AFP compared to traditional hand layup.
levels. Kumazawa et al. [99] attributed this result to decreased crack
formation and an increased number of matrix intersections. 6. Cost
As part of the European Commission CHATT project [142,143] a
linerless cylinder with an external diameter of 165 mm was constructed The cost of composite pressure vessels is a major factor in their
and permeability tested. The design utilised a combination of regular adoption within industry. Type IV pressure vessels manufactured using
and thin ply material to balance both low manufacturing time and low FW are currently used in many applications worldwide and filament
permeability. Using thin, spread tow material from TeXtreme, a void winding is recognised as an ‘affordable’ composite manufacturing
content of 0% was achieved. The authors observed that the ply thickness technology [12]. Both Type V technology and AFP based manufacturing
had a significant effect on the microcracking onset stress, with it being will introduce cost savings and losses in different areas when compared
possible to double this value given correct control of the ply thickness. with existing technology that should be understood.
The cylinder was axially loaded within a helium gas atmosphere at 3 bar Azeem et al. [12] presented the cost per litre of Type I – IV pressure
and − 150 ◦ C, an axial strain of 1.6% was recorded before any leakage or vessels. Type IV COPVs were found to be 3.5 times per expensive than
10
Table 2
Summary of resin modifications for permeation reduction.
Author Country Year Modifier Matrix Application/Mixing Result
Method
Concentration Key Effects
Churruca Argentina 2022 Halloysite nanotubes Epoxy Ultrasonic mixing 5 wt% 50% increase in cracking onset stress.
et al. [147] 10 wt% 2x increase in fracture toughness.
Ning et al. China 2022 4 morphologies of polybutyl Epoxy Mixed into resin 10 wt% Core/shell (soft/hard) morphology yielded
[148] acrylate (soft phase) and fracture toughness increases of 851% and
polymethyl methacrylate (hard 185% in the neat matrix and composite
phase) respectively.
Wei [149] USA 2022 Epoxy-reacted fluorographene Epoxy Centrifugal mixing 10 wt% 19.6% and 17.7% increases in elastic modulus
and tensile strength respectively and 81-fold
reduction in hydrogen permeability.
Chang et al. Australia 2021 Cupric oxide nanorods (nCuO) Epoxy Hand mixing/probe 4 wt% 133% increase in fracture energy and 18%
[150] sonification increase in tensile strength at room
temperature and 261% and 21% increases at
− 196 ◦ C.
Haight et al. USA 2021 ‘high viscosity modifier’ Epoxy Unclear 5% Helium permeability exceeding CCTD
[151] requirements at 24 ◦ C and − 253 ◦ C for strains
between 0 and 7500με.
Hübner et al. Germany 2021 ‘toughened epoxy-amine’ Epoxy Mixed into resin Unclear 125% and 52% increases in fracture energy and
[152] toughness at room temperature and 31% and
14% increases respectively at − 50 ◦ C.
Islam et al. Australia 2021 Polydopamine coated nCuO Epoxy Probe sonification 1 wt% 2x increase in fracture toughness at room
[146] temperature and 3x increase at − 196 ◦ C.
Nano-silica 8 wt% 1.9x increase in fracture toughness at room
temperature and 2.4x increase at − 196 ◦ C.
Yao et al. UK 2021 Graphene nanoplatelets Epoxy Spray coating 1.2 vol% Reduction of CO2 permeability of 61% at 1.32
[153] bar.
Yonemoto Japan 2009 Highly oriented nano clay Epoxy Thin film placed 1 layer 2-3 order of magnitude reduction in hydrogen
et al. [154] between prepreg permeability
sheets
Fig. 12. Optical micrographs of laminates quenched at LN2 temperature (a) produced with nCuO and (b) PDA coated nCuO (reproduced with permission
from Ref. [146]).
Type I, 2.2–2.5 times more than Type II and approximately 1.3 times carbon fibre material formed most of the cost, this conclusion has also
more than Type III. These values were not for specific operating pres been made by several other authors [12,158,159]. The higher pressure
sures. The authors acknowledged Type V but did not present any cost 700 bar tank does not cost double the 350 bar tank as much of the
information. The storage pressure also has an effect on cost, Hua et al. manufacturing equipment and balance of plant does not change signif
[157] built a cost model of a Type IV tank and compared different icantly when increasing the storage pressure. A similar result was found
pressures. The cost of 350 bar and 700 bar tanks were found to be by Sun et al. [159], who analysed the cost of Type III hydrogen tanks for
$15.4/kWh and $18.7/kWh (USD) respectively. At both pressures, the automotive applications. The authors found that material cost scales
11
nonlinearly with storage pressure, i.e., doubling pressure does not Table 3
double cost. The authors found an operating pressure that optimised cost Composite pressure vessel commercial standards.
and storage density of 500–550 bar. This is notably less than the 700 bar Standard Year Name Type V AFP Association
Type IV tanks currently used in the automotive industry. Covered?
The cost effectiveness of AFP based manufacturing for pressure ISO FDIS 2022 Gas cylinders — No No, fibres must
vessels has also not been investigated in detail. The cost benefit of AFP 11515 Refillable composite be continuous
compared with other methods varies with part size and geometry, the [169] reinforced tubes of
process is best suited to large components where high layup rates can be water capacity between
450L and 3000 L —
maintained [160]. The US DoE Hydrogen Program performed a Design, construction
comprehensive cost assessment of Type IV COPVs made using FW vs a and testing
hybrid AFP/FW approach [39]. The lowest cost was achieved by pro BS EN 2022 Transportable gas Yes Unclear, refers
ducing AFP reinforcements for the dome on a separate production line (i. 12245 cylinders - Fully to ‘wound
[170] wrapped composite composite’
e., in parallel) to the main vessel and integrating them in secondary step.
cylinders Tank must be a
This yielded an 11% cost saving and 15% material reduction. These single piece
savings were achieved despite the significantly lower material deposi CSA/ 2021 Compressed Hydrogen Unknown Unknown
tion rate of AFP vs FW (0.9 kg/h vs 13.2 kg/h), as the authors calculated ANSI Gas Vehicle Fuel
the material cost to be 84–90% of the total vessel cost at a scale of 500, HGV 2 Containers
[171]
000 units per year, meaning that machine utilisation is of secondary ASME 2021 BPVC Section X-Fiber- Yes, No, applicable
interest behind optimising composite layup. BPVC- Reinforced Plastic specifies manufacturing
An additional cost consideration when using AFP is the need for X Pressure Vessels that ‘liners processes are
inspection. Björnsson [82] found that 32% of total AFP machine time is [172] may be dictated
used’
spent inspecting the part. Similarly, layup time has been estimated as
ISO 2020 Gas cylinders - Design, Yes, for test No, fibres must
low as 24–26% [55,161]. To improve these statistics, numerous authors 11119- construction and testing pressures up be continuous
have investigated automated inspection techniques. Sacco et al. [162, 3 [173] of refillable composite to 60 bar or Tank must be a
163] developed the ACSIS system for automated ply inspection using gas cylinders and tubes - operating single piece
laser profilometry and machine learning for defect tracking. Krombholz Part 3: Fully wrapped pressures up
fibre reinforced to 40 bar
et al. [164] used laser edge detection to detect deviations from the
composite gas cylinders
specified layup path and correct them on-the-fly. Additionally, Oro and tubes up to 450l
miehie et al. [13] and Brasington et al. [55] have recently reviewed with non-load-sharing
inspection techniques for AFP. metallic or non-metallic
liners or without liners –
Third Edition
7. Commercial standards BS EN 2020 Gaseous hydrogen. Unknown Not specified
17533 Cylinders and tubes for
A final important consideration in the adoption of both AFP and Type [174] stationary storage
V technology in pressure vessels is its ability to conform to standards. BS EN 2020 Transportable gas No Unclear, refers
17339 cylinders. Fully to ‘winding
Numerous standards for the design, manufacturing, and certification of [175] wrapped carbon process’
COPVs/CPVs exist. Many countries have their own local standard, composite cylinders and
however the ISO (Switzerland), ASME (USA) and EN (Europe) standards tubes for hydrogen
are popularly used worldwide. Mair et al. [165–168] have extensively ISO/TS 2019 Gas cylinders — No Unclear, refers
17519 Refillable permanently to ‘winding
researched composite pressure vessel regulations from a safety and
[176] mounted composite process’
effectiveness perspective. Table 3 lists major composite pressure vessel tubes for transportation
standards and their coverage of both Type V tanks and AFP. For stan ISO FDIS 2019 Gaseous hydrogen — No No, filament
dards outside the public domain that were inaccessible, information that 19884 Cylinders and tubes for wound only
could not be found within publicly available abstracts or descriptions of [177] stationary storage
SAE 2018 Standard for Fuel Unknown Unknown
the standard has been marked as ‘unknown’. Some non-current stan J2579 Systems in Fuel Cell and
dards have also been included where relevant. In general, these stan [178] Other Hydrogen
dards fail to sufficiently cover both Type V pressure vessels and Vehicles
manufacturing using AFP. ANSI/ 2018 Standard: Space Unknown Unknown
AIAA Systems—Composite
Several of the standards cover linerless (Type V) tanks, making it
S–081B Overwrapped Pressure
possible to certify a filament would Type V tank. However, these stan [179] Vessels
dards are primarily written to cover Type IV vessels and only feature a ISO 2013 Gas cylinders — Yes, for test No, filament
small number of extra clauses to encapsulate Type V, rather than specific 11119- Refillable composite gas pressures up wound only
and tailored design, testing, and certification information. In the case of 3 [180] cylinders and tubes Part to 60 bar or Tanks made
3: Fully wrapped fibre operating from two pieces
ISO 11119-3 2013 and 2020, the operating pressure of linerless tanks is reinforced composite pressures up allowed
capped far below the industry-standard pressure of 700 bar. In this gas cylinders and tubes to 40 bar
standard, allowable permeation for Type IV and V tanks is the same, up to 450L with non-
meaning that a linerless tank would have to achieve the same low load-sharing metallic or
non-metallic Liners –
permeability as the dedicated liner in a Type IV tank to achieve
Second Edition
certification. ISO DIS 2011 Gaseous hydrogen and No No, fibres must
No standard explicitly admits AFP as an acceptable manufacturing 15869 hydrogen blends — be continuous
process. Several standards state filament winding and/or fully contin [181] Land vehicle fuel tanks
uous fibre as a requirement. This practically eliminates the ability to use AS 1210 2010 Pressure vessels Not Not specified
[182] specified
AFP, since a key advantage of the AFP process is the ability to incor 2005 Filament-wound FRP Unknown No, filament
porate discontinuous fibre bands to achieve local reinforcement. Other pressure vessels. wound only
standards are more ambiguous and use terminology such as ‘winding’ or (continued on next page)
‘wrapping’ that implies filament winding but does not explicitly require
12
Table 3 (continued ) is only achievable using AFP as it requires tow drop offs and fibre paths
Standard Year Name Type V AFP Association traversing the dome centre that would be difficult to achieve with fila
Covered? ment winding. Variable angle tow is incorporated along the cylinder
BS EN Materials, design,
region to tailor the stiffness. To analyse and optimise the tank design a
13923 manufacturing and novel FEA process has been developed to link CGTech VERICUT VCP, an
[183] testing AFP programming software [186], to ANSYS for semi-automated anal
ISO 2003 Space systems — Not Not specified ysis of AFP layups.
14623 Pressure vessels and specified
The next steps to translate this concept to reality include design and
[184] pressurised structures —
Design and operation manufacturing of the custom collapsible tool and prototype tank
ISO 2002 Gas cylinders of Yes No, filament manufacturing tests. Firstly, fibre paths for both vessel halves will be
11119- composite construction wound only programmed using VERICUT VCP. A prototype vessel will then be
3 [185] — Tanks made manufactured by making both halves on the collapsible tool and
Specification and test from two pieces
methods — Part 3: Fully allowed
bonding them as depicted in Fig. 13 with optical fibre sensors embedded
wrapped fibre for structural health monitoring. This step aims to validate the tooling
reinforced composite design and AFP fibre path programming. 3D scanning will be used to
gas cylinders with non- compare the actual geometry and thickness of the parts with computer
load-sharing metallic or
models. After polar boss installation and bonding of the vessel, hydro
non-metallic liners –
First Edition static pressure testing will be performed, and strains recorded using the
embedded sensors for comparison to finite element predictions.
Following this initial work, the vessel will be overwrapped with tows
it. Some standards also require the tank to be a single piece. This limits that traverse both halves of the tank, for added reinforcement and
the scope of compliant manufacturing processes as fewer tank archi increased pressure capacity. The overwrap fibre path will utilise steered
tectures are possible. fibres to create a variable stiffness laminate, this is not only required for
fibre continuity but fully demonstrates the advantages of AFP over
8. Novel design for an AFP produced Type V pressure vessel filament winding. Finally, research on permeation reduction, more
efficient and accurate simulation as well as improved tooling design will
Considering the findings of the presented literature review, a design be undertaken to progress the project.
for an AFP manufactured Type V CPV has been produced that will form
the basis for future experimental work and address the gaps identified in 9. Conclusions
this critical review. The ARC Training Centre for Automated Manufac
ture of Advanced Composites at UNSW Sydney is equipped with an This critical review has summarised the current progress in Type IV
Automated Dynamics AFP robot with thermoset and thermoplastic and Type V composite pressure vessels and discussed key challenges
placement capability as well as a coordinated spindle axis, positioning associated with the progression of their development. Additionally, the
the Centre well to develop COPVs and CPVs. use of AFP as a manufacturing process to replace traditional filament
Fig. 13 depicts the design concept, this initial design aims to address winding has been discussed. Filament winding suffers from several
the manufacturing challenges associated with Type V tanks identified in limitations that reduce its flexibility, including the need for the fibre to
this review. The thermoset tank is constructed from two halves which be continuous, inability to wind around concave surfaces without fibre
are separately manufactured on a collapsible tool and cured before being bridging and minimal non-geodesic path capability due to frictional
bonded together. This forms a linerless, composite mandrel over which limitations. AFP can address each of these limitations with its selective
additional continuous layers are wrapped. Producing the tank in two reinforcement and VAT capability, potentially offering the ability to
halves avoids the use of a dissolvable core and allows each half to be further optimise composite pressure vessel designs. However, existing
manufactured simultaneously in a mass production scenario. Uniquely, applications of AFP to pressure vessels have generally been limited to
the tank only uses a polar boss on a single end, with the other end being rocket-scale tanks for space applications. Additional work is required to
fully overwrapped. This feature provides a weight reduction benefit and
Fig. 13. Novel AFP produced Type V pressure vessel concept.
13
address interference caused by the large physical size of the AFP head on [5] Park G, Jang H, Kim C. Design of composite layer and liner for structure safety of
hydrogen pressure vessel (type 4). J Mech Sci Technol Aug. 2021;35(8). https://
smaller automotive and aerospace scale vessels.
doi.org/10.1007/s12206-021-0723-9.
Type V pressure vessels are currently facing challenges in both the [6] Toyota Motor Corporation. Toyota Mirai FCV poster. https://www.toyota-euro
manufacturing process and gas permeability reduction. To produce a pe.com/download/cms/euen/Toyota%20Mirai%20FCV_Posters_LR_tcm-11-5
linerless vessel, an internal layup tool must be removed. Water soluble 64265.pdf. [Accessed 14 July 2022].
[7] Honda Motor Company. Clarity fuel cell: technology picture book. https://global.
cores are common for this purpose but may not be practical in all designs honda/innovation/FuelCell/Clarity-Fuel-Cell-picturebook.html. [Accessed 14
and are not reusable. Solid, internally collapsible tooling has been July 2022].
demonstrated on larger scale tanks, but not on smaller tanks where the [8] Ebermann M, Bogenfeld R, Kreikemeier J, Glüge R. Analytical and numerical
approach to determine effective diffusion coefficients for composite pressure
polar opening size is reduced and internal access is difficult. Secondly, vessels. Compos Struct Jul. 2022;291. https://doi.org/10.1016/j.
further understanding and reduction in gas permeation through the compstruct.2022.115616.
composite laminate is required. The presence of matrix microcracks in a [9] Balasooriya W, Clute C, Schrittesser B, Pinter G. A review on applicability,
limitations, and improvements of polymeric materials in high-pressure hydrogen
composite laminate contributes significantly to its gas barrier perfor gas atmospheres. Polym Rev Mar. 2021;62(1). https://doi.org/10.1080/
mance. While mechanisms to reduce permeability, primarily through 15583724.2021.1897997.
thin ply materials and nanomodifications, have been demonstrated on a [10] Hua TQ, Roh H-S, Ahluwalia RK. Performance assessment of 700-bar compressed
hydrogen storage for light duty fuel cell vehicles. Int J Hydrogen Energy Oct.
sample level, system level evaluation is still required. In addition, the 2017;42(40):25121–9. https://doi.org/10.1016/j.ijhydene.2017.08.123.
effects of AFP on permeability have not been extensively investigated. [11] Jones B, Li M-C. Liner-less tanks for space application - design and manufacturing
Commercial standards for COPV/CPV manufacturing have been considerations. In: Presented at the 5th conf. Aerospace materials, processes, and
environmental technology, huntsville, AL, USA, Sep. 16-18; 2003.
reviewed, highlighting a gap in coverage for both AFP production and
[12] Azeem M, et al. Application of filament winding technology in composite pressure
Type V vessels. As development progresses on each of these technolo vessels and challenges: a review. J Energy Storage May. 2022;49. https://doi.org/
gies, standards should be updated to maintain relevance with the state of 10.1016/j.est.2021.103468.
the art. The impact on cost of introducing Type V technology and AFP [13] Oromiehie E, Prusty G, Compston P, Rajan G. Automated fibre placement based
composite structures: review on the defects, impacts and inspections techniques.
manufacturing also requires further work, however cost savings due to Compos Struct Sep. 2019;224. https://doi.org/10.1016/j.
reduced material usage in AFP compared with FW has been demon compstruct.2019.110987.
strated in a prototype setting. Finally, a novel design for an AFP pro [14] Alves MP, Gul W, Cimini Junior CA, Ha SK. A review on industrial perspectives
and challenges on material, manufacturing, design and development of
duced Type V pressure vessel has been introduced. Future work will compressed hydrogen storage tanks for the transportation sector. Energies Jul.
include the detailed design, material level testing and prototype 2022;15(14). https://doi.org/10.3390/en15145152.
manufacturing using AFP of this concept to address the existing tech [15] Wang X, et al. Advances on materials design and manufacture technology of
plastic liner of type IV hydrogen storage vessel. Int J Hydrogen Energy Feb. 2022;
nical challenges. 47(13):8382–408. https://doi.org/10.1016/j.ijhydene.2021.12.198.
[16] Rahul B, Dharmahinder Singh C, Dharani J. A comprehensive review on the
performance analysis of composite overwrapped pressure vessels. Eng. Appl. Sci.
CRediT author statement
Res. Apr. 2022;49(2):272–87. https://doi.org/10.14456/easr.2022.29.
[17] Sharma P, et al. Effects of dome shape on burst and weight performance of a type-
Alexander Air: Conceptualisation, Methodology, Investigation, 3 composite pressure vessel for storage of compressed hydrogen. Compos Struct
Writing – Original Draft, Writing – Review & Editing, Visualisation. Md Aug. 2022;293. https://doi.org/10.1016/j.compstruct.2022.115732.
[18] Rafiee R, Salehi A. Estimating the burst pressure of a filament wound composite
Shamsuddoha: Methodology, Writing – Review & Editing, Supervision. pressure vessel using two-scale and multi-scale analyses. Mech Adv Mater Struct
B. Gangadhara Prusty: Conceptualisation, Writing – Review & Editing, 2022;29:1–16. https://doi.org/10.1080/15376494.2022.2062077.
Supervision, Project administration, Funding acquisition. [19] Nebe M, Johman A, Braun C, van Campen JMJF. The effect of stacking sequence
and circumferential ply drop locations on the mechanical response of type IV
composite pressure vessels subjected to internal pressure: a numerical and
experimental study. Compos Struct Aug. 2022;294. https://doi.org/10.1016/j.
Declaration of competing interest compstruct.2022.115585.
[20] Su Y, Lv H, Zhou W, Zhang C. Review of the hydrogen permeability of the liner
material of type IV on-board hydrogen storage tank. World Electr. Veh. J. Sep.
The authors declare that they have no known competing financial
2021;12(3). https://doi.org/10.3390/wevj12030130.
interests or personal relationships that could have appeared to influence [21] Schakel M, Janssen H, Brecher C. Process analysis of manufacturing thermoplastic
the work reported in this paper. type-IV composite pressure vessels with helical winding pattern. In: SAMPE neXus
proc., jun. 29 - jul. 1; 2021.
[22] Munzke D, et al. Monitoring of type IV composite pressure vessels with multilayer
Data availability fully integrated optical fiber based distributed strain sensing. In: 12th int. Conf.
Composite Science and technology (ICCST12), Sorrento, Italy, may. 8-10; 2021.
No data was used for the research described in the article. p. 217–23. https://doi.org/10.1016/j.matpr.2020.02.872.
[23] Jois KC, Welsh M, Gries T, Sackmann J. Numerical analysis of filament wound
cylindrical composite pressure vessels accounting for variable dome contour.
Acknowledgements J. Compos. Sci Feb. 2021;5(2). https://doi.org/10.3390/jcs5020056.
[24] Zhang Q, Xu H, Jia X, Zu L, Cheng S, Wang H. Design of a 70 MPa type IV
hydrogen storage vessel using accurate modeling techniques for dome thickness
The authors would like to acknowledge the support received through prediction. Compos Struct Mar. 2020;236. https://doi.org/10.1016/j.
the following funding schemes of the Australian Government. compstruct.2020.111915.
[25] Souza G, Tarpani JR. Using OBR for pressure monitoring and BVID detection in
ARC LIEF: An Australasian facility for the automated fabrication of
type IV composite overwrapped pressure vessels. J Compos Mater Aug. 2020;55
high-performance bespoke components (LE140100082). (3):423–36. https://doi.org/10.1177/0021998320951616.
ARC ITTC: ARC Training Centre for Automated Manufacture of [26] Sapre S, Pareek K, Vyas M. Investigation of structural stability of type IV
Advanced Composites (IC160100040). compressed hydrogen storage tank during refueling of fuel cell vehicle. Energy
Stor Mar. 2020;2(4). https://doi.org/10.1002/est2.150.
[27] Alam S, Yandek GR, Lee RC, Mabry JM. Design and development of a filament
References wound composite overwrapped pressure vessel. Compos. C: Open Access Oct.
2020;2. https://doi.org/10.1016/j.jcomc.2020.100045.
[28] Nebe M, Maraite D, Braun C, Hülsbusch D, Walther F. Experimental
[1] Hydrogen Council. Path to hydrogen competitiveness: a cost perspective.
characterization of the structural deformation of type IV pressure vessels
Brussels, Belgium: Hydrogen Council; 2020.
subjected to internal pressure. Key Eng Mater Jun. 2019;809:47–52. https://doi.
[2] Rivard E, Trudeau M, Zaghib K. Hydrogen storage for mobility: a review.
org/10.4028/www.scientific.net/KEM.809.47.
Materials Jun. 2019;12(12). https://doi.org/10.3390/ma12121973.
[29] Rafiee R, Torabi MA. Stochastic prediction of burst pressure in composite pressure
[3] Schlapbach L, Züttel A. Hydrogen-storage materials for mobile applications.
vessels. Compos Struct Feb. 2018;185:573–83. https://doi.org/10.1016/j.
Nature Nov. 2001;414(6861):353–8. https://doi.org/10.1038/35104634.
compstruct.2017.11.068.
[4] Hassan IA, Ramadan HS, Saleh MA, Hissel D. Hydrogen storage technologies for
stationary and mobile applications: review, analysis and perspectives. Renew
Sustain Energy Rev Oct. 2021;149. https://doi.org/10.1016/j.rser.2021.111311.
14
[30] Cho S-M, Kim K-S, Lee S-K, Jung G-S, Lee S-K, Lyu S-K. Effect of dome curvature [56] Wehbe R, Sacco C, Baz Radwan A, Albazzan M, Harik R. Influence of process
on failure mode of Type4 composite pressure vessel. Int J Precis Eng Manuf Mar. parameters in AFP fiber steering on cylinders: constant curvature paths. Compos.
2018;19(3):405–10. https://doi.org/10.1007/s12541-018-0048-5. C: Open Access Oct. 2020;2. https://doi.org/10.1016/j.jcomc.2020.100036.
[31] Johnson K, et al. Advancements and opportunities for on-board 700 bar [57] Zhang L, Wang X, Pei J, Zhou Y. Review of automated fibre placement and its
compressed hydrogen tanks in the progression towards the commercialization of prospects for advanced composites. J Mater Sci Feb. 2020;55(17):7121–55.
fuel cell vehicles. SAE Int. J. Alt. Power Jul. 2017;6(2). https://doi.org/10.4271/ https://doi.org/10.1007/s10853-019-04090-7.
2017-01-1183. [58] Smith R, Qureshi Z, Scaife R, El-Dessouky H. Limitations of processing carbon
[32] O Brádaigh C, Grogan DM, Murray B, Leen SB. Lightweight thermoplastic fibre reinforced plastic/polymer material using automated fibre placement
composite fuel tanks for space applications. In: 3rd int. Conf. And Exh. technology. J Reinforc Plast Compos Nov. 2016;35(21):1527–42.
Thermoplastic composites (ITHEC 2016), Bremen, Germany, Oct. 11-12; 2016. [59] Boon YD, Joshi SC, Bhudolia SK. Review: filament winding and automated fiber
[33] Shao Y, et al. High pressure strength of carbon fibre reinforced vinylester and placement with in situ consolidation for fiber reinforced thermoplastic polymer
epoxy vessels. Compos Struct Apr. 2016;140:147–56. https://doi.org/10.1016/j. composites. Polymers Jun. 2021;13(12). https://doi.org/10.3390/
compstruct.2015.12.053. polym13121951.
[34] Yamashita A, Kondo M, Goto S, Ogami N. Development of high-pressure [60] Leavitt M. Development of advanced manufacturing technologies for low cost
hydrogen storage system for the Toyota “Mirai”. In: SAE technical paper 2015-01- hydrogen storage vessels. May. 13 2011.
1169; 2015. https://doi.org/10.4271/2015-01-1169. [61] Leavitt M, Lam P. Development of advanced manufacturing technologies for low
[35] Magneville B, Gentilleau B, Villalonga S, Nony F, Galiano H. Modeling, cost hydrogen storage vessels. May. 15 2013.
parameters identification and experimental validation of composite materials [62] McCarville DA, Guzman JC, Dillon AK, Jackson JR, Birkland JO. Design,
behavior law used in 700 bar type IV hydrogen high pressure storage vessel. Int J manufacture and test of cryotank components. In: Comprehensive composite
Hydrogen Energy Oct. 2015;40(38):13193–205. https://doi.org/10.1016/j. materials II. second ed. Amsterdam, Netherlands: Elsevier; 2018. p. 153–79.
ijhydene.2015.06.121. [63] Höck B, et al. Innovative and efficient manufacturing technologies for highly
[36] Leh D, Saffré P, Francescato P, Arrieux R, Villalonga S. A progressive failure advanced composite pressure vessels. In: Proc. 13th European conf. Spacecraft
analysis of a 700-bar type IV hydrogen composite pressure vessel. Int J Hydrogen structures, materials & environmental testing (ECSSMET 2014), Braunschweig,
Energy Oct. 2015;40(38):13206–14. https://doi.org/10.1016/j. Germany, Apr. 1-4; 2014.
ijhydene.2015.05.061. [64] Doyle K, Doyle A, O Brádaigh CM, Jaredson D. Feasibility of carbon fibre/peek
[37] Leh D, Magneville B, Saffré P, Francescato P, Arrieux R, Villalonga S. composites for cryogenic fuel tank applications. In: Proc. 12th European conf.
Optimisation of 700 bar type IV hydrogen pressure vessel considering composite Spacecraft structures, materials & environmental testing (ECSSMET 2012),
damage and dome multi-sequencing. Int J Hydrogen Energy Oct. 2015;40(38): Noordwijk, Netherlands, mar. 20-23; 2012.
13215–30. https://doi.org/10.1016/j.ijhydene.2015.06.156. [65] Rouhi M, Ghayoor H, Fortin-Simpson J, Zacchia TT, Hoa SV, Hojjati M. Design,
[38] Berro Ramirez JP, Halm D, Grandidier J-C, Villalonga S, Nony F. 700 bar type IV manufacturing, and testing of a variable stiffness composite cylinder. Compos
high pressure hydrogen storage vessel burst – simulation and experimental Struct Jan. 2018;184:146–52. https://doi.org/10.1016/j.
validation. Int J Hydrogen Energy Oct. 2015;40(38):13183–92. https://doi.org/ compstruct.2017.09.090.
10.1016/j.ijhydene.2015.05.126. [66] Daghighi S, Zucco G, Rouhi M, Weaver PM. Bend-free design of super ellipsoids of
[39] Leavitt M, Lam P. Development of advanced manufacturing technologies for low revolution composite pressure vessels. Compos Struct Aug. 2020;245. https://doi.
cost hydrogen storage vessels. 2014. Irvine, CA, USA. org/10.1016/j.compstruct.2020.112283.
[40] Gentilleau B, Touchard F, Grandidier JC. Numerical study of influence of [67] Daghighi S, Zucco G, Weaver P. Design of variable stiffness super ellipsoidal
temperature and matrix cracking on type IV hydrogen high pressure storage pressure vessels under thermo-mechanical loading. In: Presented at the AIAA
vessel behavior. Compos Struct May. 2014;111:98–110. https://doi.org/ SCITECH 2022 Forum, san Diego, CA, USA; 2022.
10.1016/j.compstruct.2013.12.034. [68] Daghighi S, Weaver PM. Three-dimensional effects influencing failure in bend-
[41] Roh HS, Hua TQ, Ahluwalia RK. Optimization of carbon fiber usage in Type 4 free, variable stiffness composite pressure vessels. Compos Struct Apr. 2021;262.
hydrogen storage tanks for fuel cell automobiles. Int J Hydrogen Energy Sep. https://doi.org/10.1016/j.compstruct.2020.113346.
2013;38(29):12795–802. https://doi.org/10.1016/j.ijhydene.2013.07.016. [69] Fan H, Liu P, Yan K, Hu D, Wang X. Momentless design of variable stiffness
[42] Liu PF, Chu JK, Hou SJ, Xu P, Zheng JY. Numerical simulation and optimal design composite cylindrical vessels with elliptical heads. Compos Struct Sep. 2021;271.
for composite high-pressure hydrogen storage vessel: a review. Renew Sustain https://doi.org/10.1016/j.compstruct.2021.114165.
Energy Rev May. 2012;16(4):1817–27. https://doi.org/10.1016/j. [70] Kim BC, Potter K, Weaver PM. Continuous tow shearing for manufacturing
rser.2012.01.006. variable angle tow composites. Compos. A: Appl. Sci. Manuf. 2012;43(8):
[43] Bertin M, Halm D, Magneville B, Renard J, Saffre P, Villalonga S. One year 1347–56. https://doi.org/10.1016/j.compositesa.2012.02.024.
OSIRHYS IV project synthesis: mechanical behaviour of 700 bar type IV high [71] Kartav O, Kangal S, Yücetürk K, Tanoğlu M, Aktaş E, Artem HS. Development and
pressure vessel code qualification. In: 15th European conf. Composite materials analysis of composite overwrapped pressure vessels for hydrogen storage.
(ECCM15), Venice, Italy, jun. 24-28; 2012. J Compos Mater Dec. 2021;55(28):4141–55. https://doi.org/10.1177/
[44] Villalonga S, et al. Applications of full thermoplastic composite for type IV 70 00219983211033568.
MPa high pressure vessels. In: 18th int. Conf. Composite materials (ICCM18), jeju, [72] Land P, Crossley R, Branson D, Ratchev S. Technology review of thermal forming
South Korea, Aug. 21-26; 2011. techniques for use in composite component manufacture. SAE Int. J. Mater.
[45] Villalonga S, et al. Composite 700 bar-vessel for on-board compressed gaseous Manuf. Jan. 2016;9(1):81–9. https://doi.org/10.4271/2015-01-2610.
hydrogen storage. In: 17th int. Conf. Composite materials (ICCM17), Edinburgh, [73] DeVlieg RJ, Kyle, Vogeli Peter. High-speed fiber placement on large complex
UK, jul. 27-31; 2009. structures. In: SAE technical paper 2007-01-3843; 2007. https://doi.org/
[46] Funck R, Fuchs H-P. Development of all-composite compressed natural gas (CNG) 10.4271/2007-01-3843.
pressure vessel for vehicle use. In: 13th int. Conf. Composite materials (ICCM13), [74] Qu W, et al. Placement suitability analysis of automated fiber placement on
Beijing, China, jun. 25-29; 2001. curved surfaces considering the influence of prepreg tow, roller and AFP machine.
[47] Mallick K, et al. Ultralight linerless composite tanks for in-space applications. In: Compos Struct Apr. 2021;262. https://doi.org/10.1016/j.
Presented at the AIAA space 04, san Diego, CA, USA, Sep. 28-30; 2004. compstruct.2021.113608.
[48] Mallick K, Cronin J, Fabian P, Tupper M. Microcrack resistant polymers enabling [75] Belnoue JPH, et al. Understanding and predicting defect formation in automated
lightweight composite hydrogen storage vessels. In: Materials challenges in fibre placement pre-preg laminates. Compos. A: Appl. Sci. Manuf. Nov. 2017;102:
alternative and renewable energy, vol. 224. Westerville, OH, USA: American 196–206. https://doi.org/10.1016/j.compositesa.2017.08.008.
Ceramic Soc.; 2010. [76] Croft K, Lessard L, Pasini D, Hojjati M, Chen J, Yousefpour A. Experimental study
[49] Morimoto T, Ishikawa T, Yokozeki T, Hayashi Y, Shimoda T, Morino Y. of the effect of automated fiber placement induced defects on performance of
Pressurization test on CFRP liner-less tanks at liquefied nitrogen temperature. composite laminates. Compos. A: Appl. Sci. Manuf. May. 2011;42(5):484–91.
Adv Compos Mater Apr. 2004;13(2):81–8. https://doi.org/10.1163/ https://doi.org/10.1016/j.compositesa.2011.01.007.
1568551041718062. [77] Woigk W, Hallett SR, Jones MI, Kuhtz M, Hornig A, Gude M. Experimental
[50] Vasiliev VV. Composite pressure vessels: analysis, design, and manufacturing. investigation of the effect of defects in Automated Fibre Placement produced
Blacksburg, VA, USA: Bull Ridge Pub.; 2009. composite laminates. Compos Struct Oct. 2018;201:1004–17. https://doi.org/
[51] Leavitt M, et al. Development of advanced manufacturing technologies for low 10.1016/j.compstruct.2018.06.078.
cost hydrogen storage vessels. Jun. 8 2010. [78] Harik R, Saidy C, Williams SJ, Gurdal Z, Grimsley B. Automated fiber placement
[52] Engineering Technology Corp. SM+ filament winder. https://etcwinders.com/s defect identity cards: cause, anticipation, existence, significance and progression.
tandard-filament-winders/sm-plus-filament-winder/. [Accessed 22 July 2022]. In: SAMPE 2018, Long Beach, CA, USA; 2018.
[53] Parmar H, Khan T, Tucci F, Umer R, Carlone P. Advanced robotics and additive [79] Oromiehie E, Garbe U, Gangadhara Prusty B. Porosity analysis of carbon fibre-
manufacturing of composites: towards a new era in Industry 4.0. Mater Manuf reinforced polymer laminates manufactured using automated fibre placement.
Process Feb. 2022;37(5). https://doi.org/10.1080/10426914.2020.1866195. J Compos Mater Apr. 2020;54(9):1217–31. https://doi.org/10.1177/
[54] Szcesny M, Heieck F, Carosella S, Middendorf P, Sehrschön H, Schneiderbauer M. 0021998319875491.
The advanced ply placement process – an innovative direct 3D placement [80] Qu W, et al. Algorithms for constructing initial and offset path of automated fiber
technology for plies and tapes. Adv Manuf Polym Compos Sci Feb. 2017;3(1):2–9. placement for complex double-curved surfaces. Appl Compos Mater Jun. 2021;28
https://doi.org/10.1080/20550340.2017.1291398. (3):855–75. https://doi.org/10.1007/s10443-021-09901-2.
[55] Brasington A, Sacco C, Halbritter J, Wehbe R, Harik R. Automated fiber [81] Hély C, Birglen L, Xie W-F. Feasibility study of robotic fibre placement on
placement: a review of history, current technologies, and future paths forward. intersecting multi-axial revolution surfaces. Robot Comput Integrated Manuf Dec.
Compos. C: Open Access Oct. 2021;6. https://doi.org/10.1016/j. 2017;48:73–9. https://doi.org/10.1016/j.rcim.2017.02.005.
jcomc.2021.100182.
15
[82] Björnsson A. Automated layup and forming of prepreg laminates. Ph.D [108] Yokozeki T, Ogasawara T, Ishikawa T. Evaluation of gas leakage through
dissertation. Dept. Management and Engineering, Linköping Univ.; 2017. composite laminates with multilayer matrix cracks: cracking angle effects.
[83] Tao F, Zhang H, Liu A, Nee AYC. Digital twin in industry: state-of-the-art. IEEE Compos Sci Technol Dec. 2006;66(15):2815–24. https://doi.org/10.1016/j.
Trans Ind Inf Apr. 2019;15(4):2405–15. https://doi.org/10.1109/ compscitech.2006.02.024.
TII.2018.2873186. [109] Saha S, Sullivan RW, Baker ML. Gas permeability of three-dimensional stitched
[84] Qi Q, et al. Enabling technologies and tools for digital twin. J Manuf Syst Jan. carbon/epoxy composites for cryogenic applications. Compos B Eng Jul. 2021;
2021;58:3–21. https://doi.org/10.1016/j.jmsy.2019.10.001. 216. https://doi.org/10.1016/j.compositesb.2021.108847.
[85] Zambal S, Eitzinger C, Clarke M, Klintworth J, Mechin PY. A digital twin for [110] Ryan K, et al. Prediction of pressure cycle induced microcrack damage in linerless
composite parts manufacturing: effects of defects analysis based on composite tanks. In: Presented at the 47th AIAA/ASME/ASCE/AHS/ASC
manufacturing data. In: 2018 IEEE 16th int. Conf. Industrial informatics (INDIN), structures, structural Dynamics, and materials conf., Newport, RI, USA, may. 1-4;
porto, Portugal, jul. 18-20; 2018. p. 803–8. https://doi.org/10.1109/ 2006.
INDIN.2018.8472014. [111] Mallick K, et al. An integrated systematic approach to linerless composite tank
[86] Zambal S, Heindl C, Eitzinger C, Scharinger J. End-to-end defect detection in development. In: Proc. 46th AIAA/ASME/ASCE/AHS/ASC structures, structural
automated fiber placement based on artificially generated data. In: Proc. SPIE Dynamics, and materials conf., Austin, TX, USA, Apr. 18-21; 2005.
11172, 14th int. Conf. Quality control by artificial vision, Mulhouse, France, may. [112] Mallick K, Tupper M, Arritt B, Paul C. Thermo-micromechanics of microcracking
15-17; 2019. https://doi.org/10.1117/12.2521739. in a cryogenic pressure vessel. In: Presented at the 44th structures, structural
[87] Sun S, Han Z, Fu H, Jin H, Dhupia JS, Wang Y. Defect characteristics and online Dynamics, and materials conf., Norfolk, VA, USA, Apr. 7-10; 2003.
detection techniques during manufacturing of FRPs using automated fiber [113] Prasad K, Nikzad M, Sbarski I. Modeling permeability in multi-phase polymer
placement: a review. Polymers Jun. 2020;12(6). https://doi.org/10.3390/ composites: a critical review of semi-empirical approaches. Polym Rev Jan. 2021;
polym12061337. 61(1):194–237. https://doi.org/10.1080/15583724.2020.1743306.
[88] Stokes EH. Hydrogen permeability of polymer based composites under Bi-axial [114] Bois C, Malenfant J-C, Wahl J-C, Danis M. A multiscale damage and crack opening
strain and cryogenic temperatures. In: 45th AIAA/ASME/ASCE/AHS/ASC model for the prediction of flow path in laminated composite. Compos Sci
structures, structural Dynamics & materials conf., palm Springs, CA, USA, Apr. Technol Jun. 2014;97:81–9. https://doi.org/10.1016/j.
19-22; 2004. https://doi.org/10.2514/6.2004-1858. compscitech.2014.04.002.
[89] Kane MC. Permeability, solubility, and interaction of hydrogen in polymers- an [115] Ren M-f, Zhang X-w, Huang C, Wang B, Li T. An integrated macro/micro-scale
assessment of materials for hydrogen transport. Aiken, SC, USA: Savannah River approach for in situ evaluation of matrix cracking in the polymer matrix of
National Lab.; 2008. cryogenic composite tanks. Compos Struct May. 2019;216:201–12. https://doi.
[90] Ahluwalia RK, Hua TQ, Peng JK. On-board and Off-board performance of org/10.1016/j.compstruct.2019.02.079.
hydrogen storage options for light-duty vehicles. Int J Hydrogen Energy Feb. [116] Yuan M, Zhao H, Peng Y, Tian L, Yao Y, Chen Ja. Numerical method for predicting
2012;37(3):2891–910. https://doi.org/10.1016/j.ijhydene.2011.05.040. permeability of laminated composites at cryogenic temperature. AIAA J Jul.
[91] Adams P, Bengaouer A, Cariteau B, Molkov V, Venetsanos AG. Allowable 2021;59(7):2784–92. https://doi.org/10.2514/1.J060193.
hydrogen permeation rate from road vehicles. Int J Hydrogen Energy Feb. 2011; [117] Xu J, Sankar B, Bapanapalli S. Finite element based method to predict gas
36(3):2742–9. https://doi.org/10.1016/j.ijhydene.2010.04.161. permeability in cross-ply laminates. J Compos Mater May. 2008;42(9). https://
[92] Humpenöder J. Gas permeation of fibre reinforced plastics. Cryogenics Jan. 1998; doi.org/10.1177/0021998308088610.
38(1):143–7. https://doi.org/10.1016/S0011-2275(97)00125-2. [118] Hwang T-K, Hong C-S, Kim C-G. Size effect on the fiber strength of composite
[93] Hamori H, Kumazawa H, Higuchi R, Yokozeki T. Numerical and experimental pressure vessels. Compos Struct Mar. 2003;59(4):489–98. https://doi.org/
evaluation of the formation of leakage paths through CFRP cross-ply laminates 10.1016/S0263-8223(02)00250-7.
with leak barrier layers. Compos Struct Dec. 2019;230. https://doi.org/10.1016/ [119] Hwang T-K, Park J-B, Kim H-G. Evaluation of fiber material properties in
j.compstruct.2019.111530. filament-wound composite pressure vessels. Compos. A: Appl. Sci. Manuf. 2012;
[94] Kumazawa H, Susuki I, Aoki T. Gas leakage evaluation of CFRP cross-ply 43(9):1467–75. https://doi.org/10.1016/j.compositesa.2012.04.005.
laminates under biaxial loadings. J Compos Mater May. 2006;40(10). https://doi. [120] Cohen D, Mantell SC, Zhao L. The effect of fiber volume fraction on filament
org/10.1177/0021998305056379. wound composite pressure vessel strength. Compos B Eng 2001;32(5):413–29.
[95] Flanagan M, et al. Permeability of carbon fibre PEEK composites for cryogenic https://doi.org/10.1016/S1359-8368(01)00009-9.
storage tanks of future space launchers. Compos. A: Appl. Sci. Manuf. Oct. 2017; [121] Cohen D. Influence of filament winding parameters on composite vessel quality
101:173–84. https://doi.org/10.1016/j.compositesa.2017.06.013. and strength. Compos. A: Appl. Sci. Manuf. Jan. 1997;28(12):1035–47. https://
[96] Yokozeki T, Aoki T, Ishikawa T. The effect of matrix cracks on gas permeability doi.org/10.1016/S1359-835X(97)00073-0.
through CFRP laminates. Adv Compos Mater Jan. 2004;13(3–4):227–36. https:// [122] Pavlovski D, Mislavsky B, Antonov A. CNG cylinder manufacturers test basalt
doi.org/10.1163/1568551042580136. fibre. Reinforc Plast Apr. 2007;51(4):36–9. https://doi.org/10.1016/S0034-3617
[97] Kumazawa H, Ishikawa T, Suzuki I, Aoki T. Modeling of propellant leakage (07)70152-2.
through matrix cracks in composite laminates. In: Presented at the 19th AIAA [123] Bouvier M, Guiheneuf V, Jean-marie A. Modeling and simulation of a composite
applied Aerodynamics conf., jun. 11-14; 2001. high-pressure vessel made of sustainable and renewable alternative fibers. Int J
[98] Khoe C, Sen R, Bhethanabotla Venkat R. Oxygen permeability of fiber-reinforced Hydrogen Energy May. 2019;44(23):11970–8. https://doi.org/10.1016/j.
polymers. J Compos Construct Aug. 2011;15(4):513–21. https://doi.org/ ijhydene.2019.03.088.
10.1061/(ASCE)CC.1943-5614.0000187. [124] Xiao J, Wang X, Bénard P, Chahine R. Determining hydrogen pre-cooling
[99] Kumazawa H, Aoki T, Susuki I. Influence of stacking sequence on leakage temperature from refueling parameters. Int J Hydrogen Energy Sep. 2016;41(36):
characteristics through CFRP composite laminates. Compos Sci Technol Oct. 16316–21. https://doi.org/10.1016/j.ijhydene.2016.06.084.
2006;66(13):2107–15. https://doi.org/10.1016/j.compscitech.2005.12.017. [125] Henaff-Gardin C, Lafarie-Frenot MC. Specificity of matrix cracking development
[100] Schultheiß D. Permeation barrier for lightweight liquid hydrogen tanks. Ph.D in CFRP laminates under mechanical or thermal loadings. Int J Fatig Feb. 2002;24
dissertation. Augsburg, Germany: Fac. Mathematics and Natural Sciences, Univ. (2):171–7. https://doi.org/10.1016/S0142-1123(01)00070-6.
Augsburg; 2007. [126] Alam P, Mamalis D, Robert C, Floreani C, Ó Brádaigh CM. The fatigue of carbon
[101] Fujiwara H, Ono H, Onoue K, Nishimura S. High-pressure gaseous hydrogen fibre reinforced plastics - a review. Compos B Eng Jun. 2019;166:555–79. https://
permeation test method -property of polymeric materials for high-pressure doi.org/10.1016/j.compositesb.2019.02.016.
hydrogen devices (1). Int J Hydrogen Energy Oct. 2020;45(53):29082–94. [127] Ye J, et al. Failure analysis of fiber-reinforced composites subjected to coupled
https://doi.org/10.1016/j.ijhydene.2020.07.215. thermo-mechanical loading. Compos Struct Mar. 2020;235. https://doi.org/
[102] Laeuffer H, Arbaoui J, Bois C, Lavelle F, Perry N, Wahl J-C. A new device to 10.1016/j.compstruct.2019.111756.
measure permeability evolution under pressure loading: application to CFRP [128] Maus S, Hapke J, Ranong CN, Wüchner E, Friedlmeier G, Wenger D. Filling
pipes. Measurement Feb. 2017;98:68–76. https://doi.org/10.1016/j. procedure for vehicles with compressed hydrogen tanks. Int J Hydrogen Energy
measurement.2016.11.023. Sep. 2008;33(17):4612–21. https://doi.org/10.1016/j.ijhydene.2008.06.052.
[103] Laeuffer H, Guiot B, Wahl J-C, Perry N, Lavelle F, Bois C. Relationship between [129] Gabrion X, Placet V, Trivaudey F, Boubakar L. About the thermomechanical
damage and permeability in liner-less composite tanks : experimental study at the behaviour of a carbon fibre reinforced high-temperature thermoplastic composite.
lamina level. In: 14th European conf. Spacecraft structures, materials and Compos B Eng Jun. 2016;95:386–94. https://doi.org/10.1016/j.
environmental testing (ECCSMET 2016), Toulouse, France, Sep. 27-30; 2016. compositesb.2016.03.068.
[104] Nairn JA. Matrix microcracking in composites. In: Polymer matrix composites. [130] El-Dessouky HM, Lawrence CA. Ultra-lightweight carbon fibre/thermoplastic
Amsterdam, Netherlands; 2000. p. 403–32. ch. 13. composite material using spread tow technology. Compos B Eng Jul. 2013;50:
[105] Grenoble R, Gates T. Hydrogen permeability of polymer matrix composites at 91–7. https://doi.org/10.1016/j.compositesb.2013.01.026.
cryogenic temperatures. In: Presented at the 46th AIAA/ASME/ASCE/AHS/ASC [131] Zaami A, Schäkel M, Baran I, Bor TC, Janssen H, Akkerman R. A fully coupled
structures, structural Dynamics & materials conf., Austin, TX, USA, Apr. 18-21; local and global optical-thermal model for continuous adjacent laser-assisted tape
2005. winding process of type-IV pressure vessels. J Compos Mater 2020;55(8):
[106] Choi S, Sankar BV. Gas permeability of various graphite/epoxy composite 1073–1090, Apr. https://doi.org/10.1177/0021998320944598.
laminates for cryogenic storage systems. Compos B Eng Jul. 2008;39(5):782–91. [132] Cox S, et al. Stress free temperature testing and calculations on out-of-autoclave
https://doi.org/10.1016/j.compositesb.2007.10.010. composites. In: SAMPE 13. Long Beach, CA, USA: SAMPE North America; 2013.
[107] Yokozeki T, Ogasawara T, Aoki T, Ishikawa T. Experimental evaluation of gas [133] Timmerman J, Hayes B, Seferis JC. Cure temperature effects on cryogenic
permeability through damaged composite laminates for cryogenic tank. Compos microcracking of polymeric composite materials. Polym Compos Feb. 2003;24(1):
Sci Technol Jul. 2009;69(9):1334–40. https://doi.org/10.1016/j. 132–9. https://doi.org/10.1002/pc.10013.
compscitech.2008.05.019.
16
[134] Ekuase OA, Anjum N, Eze VO, Okoli OI. A review on the out-of-autoclave process [158] Nebe M, Asijee TJ, Braun C, van Campen JMJF, Walther F. Experimental and
for composite manufacturing. J. Compos. Sci Jun. 2022;6(6). https://doi.org/ analytical analysis on the stacking sequence of composite pressure vessels.
10.3390/jcs6060172. Compos Struct Sep. 2020;247. https://doi.org/10.1016/j.
[135] Sofi T, Neunkirchen S, Schledjewski R. Path calculation, technology and compstruct.2020.112429.
opportunities in dry fiber winding: a review. Adv Manuf Polym Compos Sci Jul. [159] Sun B-g, Zhang D-s, Liu F-s. Analysis of the cost-effectiveness of pressure for
2018;4(3):57–72. https://doi.org/10.1080/20550340.2018.1500099. vehicular high-pressure gaseous hydrogen storage vessel. Int J Hydrogen Energy
[136] Veldenz L, Di Francesco M, Giddings P, Kim BC, Potter K. Material selection for Sep. 2012;37(17):13088–91. https://doi.org/10.1016/j.ijhydene.2012.04.126.
automated dry fiber placement using the analytical hierarchy process. Adv Manuf [160] Crosky A, Grant C, Kelly D, Legrand X, Pearce G. Fibre placement processes for
Polym Compos Sci 2018;4(4):83–96. https://doi.org/10.1080/ composites manufacture. In: Boisse P, editor. Advances in composites
20550340.2018.1545377. manufacturing and process design. Cambridge, UK: Woodhead Publishing; 2015.
[137] Veldenz L, Francesco MD, Astwood S, Dell’Anno G, Kim BC, Potter K. p. 79–92. ch. 4.
Characteristics and processability of bindered dry fibre material for automated [161] Maass D. Progress in automated ply inspection of AFP layups. Reinforc Plast Sep.
fibre placement. In: 17th European conf. Compos. Mater. (ECCM17 2016), 2015;59(5):242–5. https://doi.org/10.1016/j.repl.2015.05.002.
Munich, Germany, jun. 26-30; 2016. [162] Sacco C, Baz Radwan A, Anderson A, Harik R, Gregory E. Machine learning in
[138] Veldenz L, Francesco MD, Astwood S, Giddings P, Kim BC, Potter K. Feature-based composites manufacturing: a case study of Automated Fiber Placement
design for manufacturing guidelines for dry fibre AFP. In: Autom. Compos. inspection. Compos Struct Oct. 2020;250. https://doi.org/10.1016/j.
Manuf. Lancaster, PA, USA: Third Intl. Symp. 2018; 2018. compstruct.2020.112514.
[139] Defauchy V, Le Corre H, Colin X. Simulation of the oxygen permeability of a [163] Sacco C, Baz Radwan A, Harik R, Van Tooren M. Automated fiber placement
composite container. J. Compos. Sci Mar. 2018;2(2):21. https://doi.org/10.3390/ defects: automated inspection and characterization. In: SAMPE 18. Long Beach,
jcs2020021. CA, USA: SAMPE North America; 2018.
[140] Galos J. Thin-ply composite laminates: a review. Compos Struct Mar. 2020;236. [164] Krombholz C, Perner M, Bock M, Roestermundt D. Improving the production
https://doi.org/10.1016/j.compstruct.2020.111920. quality of the advanced automated fiber placement process by means of online
[141] Hamori H, Kumazawa H, Higuchi R, Yokozeki T. Gas permeability of CFRP cross- path correction. In: 28th intl. Congr. Aero. Sci. Brisbane: ICAS); 2012.
ply laminates with thin-ply barrier layers under cryogenic and biaxial loading [165] Mair GW, Becker B, John S, Duffner E. Composite storage systems for compressed
conditions. Compos Struct Aug. 2020;245. https://doi.org/10.1016/j. hydrogen - systematic improvement of regulations for more attractive storage
compstruct.2020.112326. units. Int J Hydrogen Energy May. 2020;45(25):13672–9. https://doi.org/
[142] Sippel M, Kopp A, Mattsson D, Freund J, Tapeinos I, Koussios S. Final results of 10.1016/j.ijhydene.2018.04.068.
advanced cryo-tanks research project CHATT. In: Presented at the 6th European [166] Mair GW, Hoffmann M. Regulations and research on RC&S for hydrogen storage
conf. Aeronautics and space Sciences (EUCASS 2015), Krakow, Poland, jun. 29 - relevant to transport and vehicle issues with special focus on composite
jul. 3; 2015. containments. Int J Hydrogen Energy Apr. 2014;39(11):6132–45. https://doi.
[143] Sippel M, Kopp A, Mattsson D, Koussios S. Advanced cryo-tanks structural design org/10.1016/j.ijhydene.2013.08.141.
investigated in CHATT. In: Proc. 13th European conf. Spacecraft structures, [167] Mair GW, Scherer F, Hoffmann M. Type approval of composite gas cylinders –
materials & environmental testing (ECSSMET 2014), Braunschweig, Germany, probabilistic analysis of standards’ requirements concerning minimum burst
Apr. 1-4; 2014. pressure. Int J Hydrogen Energy Apr. 2015;40(15):5359–66. https://doi.org/
[144] Beaumont PWR, Soutis C, Hodzic A. The structural integrity of carbon fiber 10.1016/j.ijhydene.2015.01.161.
composites: fifty years of progress and achievement of the science, development, [168] Mair GW, Thomas S, Schalau B, Wang B. Safety criteria for the transport of
and applications. Cham, Switzerland: Springer; 2017. hydrogen in permanently mounted composite pressure vessels. Int J Hydrogen
[145] Zhang C, Zhang G, Shi X, Wang X. Effects of carbon nanotubes on the interlaminar Energy Mar. 2021;46(23):12577–93. https://doi.org/10.1016/j.
shear strength and fracture toughness of carbon fiber composite laminates: a ijhydene.2020.07.268.
review. J Mater Sci Jan. 2022;57(4):2388–410. https://doi.org/10.1007/s10853- [169] Gas cylinders. Refillable composite reinforced tubes of water capacity between
021-06734-z. 450 L and 3000 L — design, construction and testing. ISO FDIS 11515. Geneva,
[146] Islam MS, Benninger LF, Pearce G, Wang C-H. Toughening carbon fibre Switzerland: ISO; 2022.
composites at cryogenic temperatures using low-thermal expansion nanoparticles. [170] Transportable gas cylinders. Fully wrapped composite cylinders, BS EN 12245.
Compos. A: Appl. Sci. Manuf. Nov. 2021;150. https://doi.org/10.1016/j. Brussels, Belgium: European Standards; 2022.
compositesa.2021.106613. [171] Compressed hydrogen gas vehicle fuel containers, CSA/ANSI HGV 2. Washington,
[147] Churruca MJ, Morán JI, Rodríguez ES. Effect of halloysite nanotubes on matrix DC, USA: American National Standards Institute; 2021.
microcracking in carbon fiber/epoxy composites. Mech Compos Mater May. [172] BPVC section X-fiber-reinforced plastic pressure vessels, BPVC-X. New York, NY,
2022;58(2):293–304. https://doi.org/10.1007/s11029-022-10030-5. USA: ASME; 2021.
[148] Ning N, Wang M, Zhou G, Qiu Y, Wei Y. Effect of polymer nanoparticle [173] Gas cylinders. Design, construction and testing of refillable composite gas
morphology on fracture toughness enhancement of carbon fiber reinforced epoxy cylinders and tubes — Part 3: fully wrapped fibre reinforced composite gas
composites. Compos B Eng Apr. 2022;234. https://doi.org/10.1016/j. cylinders and tubes up to 450 l with non-load-sharing metallic or non-metallic
compositesb.2022.109749. liners or without liners. ISO 11119-3. Geneva, Switzerland: ISO; 2020.
[149] Wei J. Mechanically improved and multifunctional CFRP enabled by resins with [174] Gaseous hydrogen. Cylinders and tubes for stationary storage, BS EN 17533.
high concentrations epoxy-functionalized fluorographene fillers. In: Fiber- Brussels, Belgium: European Standards; 2020.
reinforced plastics. London, UK: IntechOpen; 2022. [175] Transportable gas cylinders. Fully wrapped carbon composite cylinders and tubes for
[150] Chang W, et al. Strengthening and toughening epoxy polymer at cryogenic hydrogen, BS EN 17339. Brussels, Belgium: European Standards; 2020.
temperature using cupric oxide nanorods. Compos Sci Technol May. 2021;208. [176] Gas cylinders. Refillable permanently mounted composite tubes for
https://doi.org/10.1016/j.compscitech.2021.108762. transportation. ISO/TS 17519. Geneva, Switzerland: ISO; 2019.
[151] Haight A, Fabian P, Haynes M. Microcrack resistant matrix materials for out-of- [177] Gaseous hydrogen. Cylinders and tubes for stationary storage. ISO/FDIS 11119-3.
autoclave processing of composite cryogenic tanks. In: SAMPE NeXus 2021, jun. Geneva, Switzerland: ISO; 2019.
30; 2021. [178] Standard for fuel Systems in fuel Cell and other hydrogen vehicles, J2579.
[152] Hübner F, Brückner A, Dickhut T, Altstädt V, Rios de Anda A, Ruckdäschel H. Low Warrendale, PA, USA: Society of Automotive Engineers; 2018.
temperature fatigue crack propagation in toughened epoxy resins aimed for [179] Standard: Space Systems. Composite overwrapped pressure vessels, ANSI/AIAA S-
filament winding of type V composite pressure vessels. Polym Test Oct. 2021;102. 081B,. Reston, VA, USA: American Institute of Aeronautics and Astronautics;
https://doi.org/10.1016/j.polymertesting.2021.107323. 2018.
[153] Yao X, Raine TP, Liu M, Zakaria M, Kinloch IA, Bissett MA. Effect of graphene [180] Gas cylinders. Refillable composite gas cylinders and tubes — design,
nanoplatelets on the mechanical and gas barrier properties of woven carbon construction and testing — Part 3: fully wrapped fibre reinforced composite gas
fibre/epoxy composites. J Mater Sci Aug. 2021;56(35):19538–51. https://doi. cylinders and tubes up to 450L with non-load-sharing metallic or non-metallic
org/10.1007/s10853-021-06467-z. liners. ISO 11119-3. Geneva, Switzerland: ISO; 2013.
[154] Yonemoto K, Yamamoto Y, Okuyama K, Ebina T. Application of CFRP with high [181] ISO DIS 15869 Gaseous hydrogen and hydrogen blends — land vehicle fuel tanks.
hydrogen gas barrier characteristics to fuel tanks of space transportation system. Geneva, Switzerland: ISO; 2011.
Trans. JSASS Space Tech. Japan May. 2009;7(26):13–8. https://doi.org/10.2322/ [182] Pressure vessels, AS 1210. Sydney, Australia: Standards Australia; 2021.
tstj.7.Pc_13. [183] Filament-wound FRP pressure vessels. Materials, design, manufacturing and testing, BS
[155] Klomp-de Boer R. Automated preform fabrication by dry tow placement. EN 13923. Brussels, Belgium: European Standards; 2005.
Amsterdam, Netherlands: Nat. Aerospace Lab. NLR; 2013. [184] Space systems - Pressure vessels and pressurized structures. Design and operation.
[156] Belhaj M, et al. Dry fiber automated placement of carbon fibrous preforms. ISO 14623. Geneva, Switzerland: ISO; 2003.
Compos B Eng Jul. 2013;50:107–11. https://doi.org/10.1016/j. [185] Gas cylinders of composite construction. Specification and test methods — Part 3:
compositesb.2013.01.014. fully wrapped fibre reinforced composite gas cylinders with non-load-sharing
[157] Hua TQ, et al. Technical assessment of compressed hydrogen storage tank systems metallic or non-metallic liners. ISO 11119-3. Geneva, Switzerland: ISO; 2002.
for automotive applications. Int J Hydrogen Energy Feb. 2011;36(4):3037–49. [186] Hasenjaeger B. Programming and simulating automated fiber placement (AFP)
https://doi.org/10.1016/j.ijhydene.2010.11.090. CNC machines. CA, USA: SAMPE, Diamond Bar; 2011.
17

AFP Tank

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

AFP Tank

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

AFP Tank

Uploaded by

Copyright:

Available Formats

Composites Part B 253 (2023) 110573

Contents lists available at ScienceDirect

A review of Type V composite pressure vessels and automated fibre

Alves et al. [14] Brazil 2022 X X X

3.3. Challenges in applying AFP to pressure vessels

3.3.1. Defect control

3.3.2. Physical limitations

Fig. 13. Novel AFP produced Type V pressure vessel concept.

You might also like