Towards separator-free structural composite supercapacitors

Abstract

Structural supercapacitors can both carry load and store electrical energy. An approach to build such devices is to modify carbon fibre surfaces to increase their specific surface area and to embed them into a structural electrolyte. We present a way to coat carbon fibres with graphene nanoplatelets by electrophoretic deposition in water. The effect of time and voltage on the mechanical properties of the carbon fibres, the structure of the coating and the specific surface area of the coated carbon fibres are discussed. A specific capacity of 1.44 F/g was reached, which is 130% higher than state-of-the-art structural electrodes. We demonstrate the scalability of the deposition process to continuous production of coated carbon fibres. These carbon fibre electrodes were used to realise large (21 cm long) structural supercapacitor demonstrators without the need for a separator, having a specific capacity of 623 mF/g.

Full text

Composites Science and Technology 217 (2022) 109126
Available online 2 November 2021
0266-3538/© 2021 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Towards separator-free structural composite supercapacitors
Olivier Hubert
a
, Nikola Todorovic
a
, Alexander Bismarck
a
,
b
,
*
a
Polymer & Composite Engineering (PaCE) Group, Institute of Materials Chemistry and Research, Faculty of Chemistry, University of Vienna, W¨
ahringer Str. 42, 1090
Vienna, Austria
b
Department of Chemical Engineering, Imperial College London, South Kensington Campus, London, SW7 2AZ, UK
ARTICLE INFO
Keywords:
A
Carbon bres
Multifunctional composites
B
Electro-chemical behaviour
Multifunctional properties
Surface treatments
ABSTRACT
Structural supercapacitors can both carry load and store electrical energy. An approach to build such devices is to
modify carbon bre surfaces to increase their specic surface area and to embed them into a structural elec-
trolyte. We present a way to coat carbon bres with graphene nanoplatelets by electrophoretic deposition in
water. The effect of time and voltage on the mechanical properties of the carbon bres, the structure of the
coating and the specic surface area of the coated carbon bres are discussed. A specic capacity of 1.44 F/g was
reached, which is 130% higher than state-of-the-art structural electrodes. We demonstrate the scalability of the
deposition process to continuous production of coated carbon bres. These carbon bre electrodes were used to
realise large (21 cm long) structural supercapacitor demonstrators without the need for a separator, having a
specic capacity of 623 mF/g.
1. Introduction
An increasing interest in electric vehicles has driven researchers to
develop new solutions for electrochemical energy storage devices. Be-
tween regular capacitors with a high power but low energy density and
lithium-ion batteries with high energy but limited power density and
consequently long charging time, electrostatic double-layer capacitors
(supercapacitors) offer a solution for fast charging and decent energy
density [1]. The energy storage mechanism in supercapacitors only in-
volves electrostatic interactions. Compared to batteries, the absence of
chemical reactions not only allows higher power densities but also
higher reversibility; up to 10000 cycles without capacity loss have been
reported [2]. Electrochemical energy storage systems remain nonethe-
less much heavier than petrol for the same amount of stored energy.
Therefore, new solutions to reduce the total weight of electric vehicles
are investigated. Structural energy storage systems are among them. By
using a multifunctional material that can simultaneously bear loads and
store electrical energy, substantial weight savings can be achieved [3].
Luo et al. [4] presented the rst electrochemical structural energy
storage device in 2001; a capacitor using carbon bres acting simulta-
neously as electrodes, current collectors and reinforcement. Carbon -
bres are strong, stiff [5] and electron conductors [6] and thus are used
both as reinforcement and current collector in structural energy storage
devices [7]. However, the potential drop induced by the higher resis-
tance of carbon bres (1.3⋅10
-3
Ω cm for T800S carbon bres [8]) when
compared to common metallic current collectors (2.65⋅10
−6
Ω cm for
aluminium [9]) prevent to use them for very large applications without
risking an energy loss by Joule heating [10]. Several other approaches
were investigated and devices such as structural batteries [11], fuel cells
[12] and supercapacitors [13] were reported. Structural energy storage
would allow to increase range and/or reduce the overall weight of
(hybrid) electric vehicles. In weight critical applications such as aircrafts
[14], structural energy storage allows for more design opportunities and
potential energy savings [15].
Supercapacitors are composed of two electrodes, a separator and an
electrolyte [16]. In the case of symmetric devices, the positive and
negative electrodes are similar. The device capacity is proportional to
the specic surface area of the electrodes [17]. Therefore, the electrode
material should be chosen carefully. Typically supercapacitor electrodes
are carbon materials [6,18,19]. Activated carbon is the oldest and still
most commonly used electrode material for supercapacitors as it is
cheap and widely available. More recently, carbon nanotubes [20] and
graphene [21] have been investigated as potential electrode materials.
Graphene is a two-dimensional hexagonal lattice of carbon. Individual
graphene sheets have a specic surface area of 2600 m
2
/g. Graphene
networks produced from graphene oxide can reach specic surface areas
* Corresponding author. Polymer & Composite Engineering (PaCE) Group, Institute of Materials Chemistry and Research, Faculty of Chemistry, University of
Vienna, W¨
ahringer Str. 42, 1090, Vienna, Austria.
E-mail address: alexander.bismarck@univie.ac.at (A. Bismarck).
Contents lists available at ScienceDirect
Composites Science and Technology
journal homepage: www.elsevier.com/locate/compscitech
https://doi.org/10.1016/j.compscitech.2021.109126
Received 3 June 2021; Received in revised form 17 September 2021; Accepted 31 October 2021

Composites Science and Technology 217 (2022) 109126
2
above 550 m
2
/g before treatment and 3500 m
2
/g after chemical acti-
vation [22]. Graphene has already been used in supercapacitor appli-
cations [23,24], including structural energy storage devices [25].
Several routes to produce structural electrodes have already been
investigated; Activated carbon bres and carbon papers made from
carbon nanotubes or nanofoams were also investigated but have poor
mechanical properties and hence are not suitable for structural appli-
cations [26]. Chemical and physical activation of structural carbon -
bres can increase the specic surface area without degrading the
mechanical properties. Activation of carbon bres with KOH increased
the specic surface area from 0.33 m
2
/g to 32.8 m
2
/g [27]. The specic
capacity, measured in aqueous KCl (3 M), increased for such activated
carbon bres from 0.06 F/g to 2.63 F/g [28]. Similar results were ob-
tained when grafting or sizing carbon bres with carbon nanotubes [29].
Coating carbon bres with carbon aerogel resulted in the highest spe-
cic capacity reported so far for structural electrodes with 14.3 F/g in
3M KCl in water [30]. Nevertheless, these techniques are work and
energy intensive and most materials lose much of their capacity when
combined with a structural polymer electrolyte, such as poly(ethylene
glycol)-based electrolytes. The authors [30] attributed the capacity drop
between liquid electrolyte and structural electrolyte to the lower ionic
conductivity in the latter. The highest specic capacity of a structural
supercapacitor reported thus far was 603 mF/g [30].
Electrophoretic deposition (EPD) is a cheap, easily scalable and
adjustable technique to deposit materials on a conductive substrate. It
has been used for various materials such as ceramics, porous materials,
biomaterials and nanoparticles [31]. The particles to be deposited are
suspended in a liquid medium and a voltage is applied between the
substrate and a counter electrode. Charged particles consequently move
towards the oppositely charged electrode and coat its surface. The
amount of deposited material is proportional to the deposition time, the
surface of the electrodes, the applied electric eld, the particle con-
centration and their electrophoretic mobility [32]. Time, electrode
surface and applied electric eld can be directly controlled and adjusted
before or during the coating process. Electrophoretic mobility in a given
medium can be modied by changing the chemical environment of the
suspended particles, e.g. by changing the ionic strength of the solution.
In an aqueous suspension, the electrophoretic mobility of particles de-
pends strongly on the pH. EPD of electrode material was used for the
production of structural lithium-ion battery cathodes [33] but no in-
formation was found for structural supercapacitor applications. A
continuous process for the EPD of graphene oxide on carbon bres was
already studied but only with the aim to improve the mechanical
properties of composites [34].
We describe a simple method to produce carbon material coated
structural carbon bres by electrophoretic deposition. We optimised the
EPD process parameters and outcome of graphene coated carbon bres
in a batch process. The coating quality and the capacity of the electrodes
in both liquid and structural electrolyte was analysed. Furthermore, we
developed a continuous EPD process to produce graphene coated carbon
bres. Separator-free structural supercapacitor demonstrators were
prepared using these coated bres. The impact of the coating process on
the tensile properties of the carbon bres is also presented.
2. Experimental
2.1. Materials
Unsized, untreated polyacrylonitrile (PAN) based carbon bres (12k
AS4D) were kindly provided by Hexcel (Duxford, UK). The suspended
carbon materials included high specic surface area (1800 m
2
/g) carbon
black (YP50F), multi-walled carbon nanotubes kindly provided by
Kuraray Co. Ltd. (Tokyo, Japan) and Arkema (Lacq, France), respec-
tively, and graphene nanoplatelets (XGnP C-750, XGSciences) with a
surface area of 750 m
2
/g purchased from SigmaAldrich. As binder we
used styrene butadiene rubber (BM400B) kindly provided by Zeon corp.
(Düsseldorf, Germany). NaOH was supplied by Sigma Aldrich. Sigraex
F01513TH graphite paper (SGL Carbon, Germany) was used as counter
electrode. For the electrolytes, tetraethyl ammonium tetrauoroborate
(TEABF
4
), propylene carbonate (PC), poly(ethylene glycol) diglycidyl
ether (PEGDGE) and triethylenetetramine (TETA) were all purchased
from Sigma Aldrich and 1-ethyl-3-methylimidazolium tetrauoroborate
(EMIBF
4
) from Iolitec (Heilbronn, Germany). The separators used were
cellulose-based TF40-30 kindly provided by NKK Nippon Kodoshi Corp.
(Koshi, Japan). All chemicals were used as received.
2.2. Spreading of carbon bres
The carbon bres were spread using an air-assisted bre tow
spreading unit (Izumi International Inc., USA) following the method
described by Diao et al. [35] In brief, this device sucks air through the
bre tow at low bre tension resulting in the bres in the roving being
separated from each other. The tow width increased from 20 mm to 50
mm reducing the average tow thickness from ~120
μ
m to ~50
μ
m.
2.3. EPD of carbon materials on carbon bres
To increase the specic surface area of the carbon bres (geometric
surface area ≈0.3 m
2
/g), we coated them with three different high
specic surface area carbon materials: carbon black, carbon nanotubes
and graphene nanoplatelets. We also investigated a mixture of carbon
black with graphene nanoplatelets or carbon nanotubes. The carbon
black had a particle size over 5
μ
m on average, too close to the diameter
of the carbon bres (6.7
μ
m) to produce a good coating quality at the
bre level. The original carbon nanotubes investigated could not be
suspended in water due to their hydrophobic nature. After plasma
treating (Pico, Diener electronic, Ebhausen, Germany) them for 2 min in
air, the nanotubes were more hydrophilic but unfortunately aggregated
after less than 5 min. Therefore, we chose to investigate further only
graphene suspensions. Preliminary results of other carbons can be found
in the supplementary information (ESI Figs. S1–4). The graphene sus-
pension used for EPD contained 90 wt% of graphene nanoplatelets and
10 wt% of styrene butadiene rubber with a total concentration of 2 mg/
mL in water. The pH was adjusted to 10 using NaOH. After mixing all
components, the suspension was stirred for 10 min, and then sonicated
for 15 min. The suspension was again stirred 1 min and sonicated 3 min
between subsequent depositions. Spread carbon bres were immersed
into the suspension using a purpose-built sample holder (Fig. 1); The
working electrode was 12 cm long spread 12k tow carbon bres. To
avoid any metal oxide in the suspension, we used graphite paper taped
to PVC plates as counter electrode consisting of four interconnected
graphite paper sheets of 4 ×5 cm
2
arranged around and in-between the
working electrode. Voltage ranging from 10 to 30 V were applied for a
duration of 30–180 s using a laboratory power generator (EA-PS 3065-
05 B, Elektro-Automatik, Germany). After EPD, the carbon bres were
pre-dried on the sample holder using a heat gun (air around 40 ◦C).
Finally, the samples were dried to constant weight in an oven at 110 ◦C
for 1 h to remove any residual water.
2.4. Development of a continuous EPD coating process
The main interest in using EPD is the ability to scale up the process to
be run in a continuous fashion. For the continuous process we adapted
our laboratory composite production line [36]; the bres were spread
inline, EPD coated and dried resulting in ready-to-assemble spread tow
electrodes. After passing the spreading unit, the bres passed 20 cm
through a suspension bath in which EPD was performed. We adjusted
the line speed to 0.15 m/min resulting in a residence time of the bres
between the electrodes of 80 s. The voltage was set to 30 V or 40 V
because the distance between the carbon bres acting as working elec-
trode and the counter electrodes was larger as compared to the batch
process. The counter electrodes were two 20 ×5 cm
2
graphite paper
O. Hubert et al.

Composites Science and Technology 217 (2022) 109126
3
sheets taped to PVC plates. The carbon bres were connected to the
electrical circuit by passing over two graphite sleeves, one at the
beginning and one at the end of the EPD bath. After EPD, the bres went
through two 80 cm long infrared heated ovens, the rst operated at
120 ◦C and the second at 110 ◦C. The bres were pulled by a winding
unit at a speed adjusted to match the speed of the spreading process.
Fig. 2 shows a schematic of the bre coating line.
2.5. Characterisation of graphene and binder suspensions
To assess the ability of the particles to move in an applied electrical
eld and, therefore, to coat the carbon bres, the electrophoretic
mobility
μ
of the suspended materials was measured. Suspensions of 2
mg/mL of graphene nanoplatelets, binder and a 9:1 (graphene:binder)
mixture were characterised by electrophoresis (Zetasizer Nano ZS,
Malvern Panalytical, UK).
μ
was measured from pH 3 to 11 in a 1 mM
solution of KCl in water. For each point, three samples were measured
three times each. The sedimentation of the particles in suspension was
also evaluated qualitatively. Zeta (ζ) potential is proportional to the
electrophoretic mobility
μ
as given by Henry’s equation [37,38]:
ζ=3%
η
%
μ
2%
ε
%f(κa)(1)
where
ε
(=
ε
r%
ε
0) is the dielectric constant,
η
the viscosity of water and f
(κa) Henry’s function, which typically takes values from 1 to 1.5.
2.6. Characterisation of graphene coated carbon bres
The graphene coating on the carbon bres was inspected by scanning
electron microscopy (Zeiss Supra 55 VP SEM) to characterize the
thickness and morphology of the coating. To ensure a good conductivity
of the surface and consequently a good quality of the micrographs, the
carbon bres were rst coated with a thin layer of gold. The specic
surface area of the graphene coated carbon bres was determined by
nitrogen adsorption using the Brunauer–Emmett–Teller (BET) method
(Tristar II Plus, Micromeritics). The graphene loading on the carbon -
bres was determined by measuring the length and weighing the samples.
The linear mass of the carbon bres (m
l
) is 0.765 g/m. The following
equation gives the graphene loading G%:
G%=mt−lt%ml
mt−lnc%ml
%100 (2)
where m
t
is the total mass of the sample {coating +coated carbon bres
+uncoated carbon bres}, l
t
the total length and l
nc
the uncoated length
of bres (see ESI Fig. S5 for illustration).
The surface of the bres prior and after coating was analysed using X-
ray photoelectron spectroscopy (XPS). 5 samples were analysed: gra-
phene nanoplatelets, pristine carbon bres, two samples of coated car-
bon bres after having removed the coating and graphene coated carbon
bres. To remove the coating, the coated carbon bres were sonicated in
deionized water for 30 min, rinsed with ultrapure water and dried. The
data were acquired using an X-ray photoelectron spectrometer (Nexsa,
Thermo Scientic, UK) using Al-K
α
X-rays and a spot size of 400
μ
m. First
a survey spectrum was recorded and then element specic high-
resolution spectra with an energy step size of 0.1 eV were taken. We
Fig. 1. Scheme of the electrophoretic deposition setup.
Fig. 2. Scheme of the bre treatment line for continuous deposition on carbon bres.
Fig. 3. (a) A schematic of the assembly of a supercapacitor with liquid electrolyte, (b) a photograph of a structural supercapacitor pouch cell.
O. Hubert et al.

Composites Science and Technology 217 (2022) 109126
4
monitored specically carbon (279–298 eV), oxygen (525–545 eV), and
nitrogen (392–410 eV).
2.7. Assembly of supercapacitors
To assemble supercapacitors coated carbon bre samples were cut in
the middle and both sides served as a square 5 ×5 cm
2
electrode. 5 cm of
uncoated bres were left to allow for electrical connection and the rest
was cut off on each side. The supercapacitors consisted of one layer of
aluminium foil, one layer of coated carbon bres, the separator, then
again coated carbon bres and aluminium. Fig. 3a shows a schematic of
this layup. The layup was performed in a 3D-printed PLA holder that
allowed pushing all layers together to ensure good contact between
them. The electrolyte used was PC containing 1M TEABF
4
.
2.8. Assembly of structural supercapacitors
Structural supercapacitors were assembled in a glovebox (MBraun,
Germany). The electrodes and separator layup was placed between two
sheets of release lm (Upilex-25S, UBE, Osaka, Japan) and impregnated
with 0.8 mL of structural electrolyte. The electrolyte used was described
before [30], and consisted of 82.6 wt% PEGDGE, 7.4 wt% TETA and 10
wt% EMIBF
4
. The samples were cured 24 h at 80 ◦C in the oven
compartment of the glovebox. During curing, this layup was sandwiched
between two metallic plates using a spring clamp to apply pressure. The
uncoated part of the carbon bre electrodes was not impregnated with
electrolyte to be able to connect them to the potentiostat. When the
samples were fully cured, they were sealed in a plastic pouch before
being taken out of the glovebox for testing. The plastic pouch included
two aluminium connectors in contact with the uncoated bre section
(Fig. 3b).
2.9. Assembly of structural separator-free supercapacitor demonstrators
Continuously coated bres were cut into 23 cm long strips to prepare
21 cm electrodes with extra 2 cm for electrical connections. Copper tape
was attached to one side of the electrode to allow for electrical contact to
the potentiostat (Fig. 11). A vacuum bag was prepared on an aluminium
plate and the prepared bre electrodes were placed into the bag. The
layup consisted of a layer of polyimide release lm on the Al plate, a
layer of PTFE coated glass bre peel-ply (FF03PM, Cytec Engineered
Materials Ltd., UK), two coated carbon bre electrodes side-by-side,
another layer of peel-ply followed by release lm and nally the vac-
uum bag (see Fig. 4). 0.8 mL of structural electrolyte was drop casted on
each carbon bre electrode. The vacuum bag was then sealed with
thermal resistant tape (Airdam 1, Airtech, Luxembourg) and a 21 ×21
cm
2
metal plate was placed on top. The vacuum bag was then press-
claved. Vacuum was applied and the temperature of the hot press raised
to 80 ◦C. A pressure of 1.2 MPa was applied. After 10 min, the vacuum
pump was turned off. After 4 h, the pressure was released, the plate
removed from the hot press and allowed to cool down for 30 min. Then,
the individual coated carbon bre electrodes were superimposed over
the active 21 cm, with the copper connections on opposite sides. 0.3 mL
of electrolyte was coated over the interface for bonding the two layers.
This layup was then placed between two Al plates covered with release
lm and placed in an oven with a 5 kg weight on top. The oven was
operated at 80 ◦C and the assembled supercapacitors were left to cure for
20 h. Three supercapacitor demonstrators were manufactured.
2.10. Electrochemical characterisation of the supercapacitors
The assembled supercapacitors were tested using a potentiostat
(Reference 600, Gamry). Cyclic voltammograms were recorded between
1 V and −1 V at a charging rate of 5 mV/s. The capacity C was
calculated:
I=C%(dV
dt )V=0(3)
where I is the current and V the voltage. C was calculated at zero voltage
to limit the impact of resistance and pseudo-capacitance on the results.
Three cycles were recorded. The reported capacity is the average value
of the 2nd and the 3rd cycles. A minimum of ve supercapacitors was
tested for each condition.
2.11. Characterisation of separator-free supercapacitor demonstrators
The composites were weighed to evaluate the bre weight fraction
and their thickness was measured using a micrometre screw gauge. The
separator-free supercapacitors were electrochemically tested using cy-
clic voltammetry at a rate of 5 mV/s between −1 V and 1 V for 1500
cycles to assess their performance. We calculated the capacity for each
cycle using Eq. (3). The energy density of the separator-free structural
supercapacitor demonstrators was calculated using a galvanostatic
charge-discharge measurement. A voltage step of 1 V was applied for 60
s and then the specimen was allowed to discharge for another 60 s. The
tting and calculation were performed as described by Qian et al. [30].
2.12. Characterisation of single bre tensile properties
To quantify the impact of EPD on the mechanical properties of the
carbon bres we measured the breaking stress of pristine and coated
bres (Favimat +single-bre tester, Textechno). For both pristine and
coated carbon bres, a minimum of 80 single bres was characterized at
4 different gauge lengths (18, 25, 35 and 50 mm). The bres to be
characterised were extracted from continuously EPD (U =40 V, t =80 s)
coated bre tows. The measurements were performed following the
standard ASTM C1557 with a test speed of 0.5 mm/min. To determine
the bre diameter, the linear density was measured using the frequency
of resonance of the bres and then divided by the density of carbon -
bres (1.79 g/cm
3
). All data were then processed using unimodal Weibull
analysis [39,40] following the standard ASTM C1239 to report the
characteristic strength, Weibull modulus as well as the 90% condence
interval for each tested condition.
2.13. Characterisation of tensile properties of composite supercapacitors
The ultimate tensile strength as well as the Young modulus of the
separator-free structural supercapacitors were measured with a method
Fig. 4. Scheme of the vacuum bag. 1: Metal plate, 2: vacuum bag, 3: carbon bres, 4: peel-ply, 5: release lm, 6: vacuum valve, 7: breathing cloth, 8: sealing tape.
O. Hubert et al.

Composites Science and Technology 217 (2022) 109126
5
adapted from ASTM D3039. First the sample were cut to a width of 25
mm. Then, glass bre epoxy composite end tabs of 40 ×25 mm
2
were
attached on each side of the specimens using Araldite® glue. The
remaining gauge length was 130 mm and the thickness, varying for each
sample, from 260
μ
m to 190
μ
m. The prepared specimen were loaded in
tension using a universal test frame (Instron 5969) equipped with a 50
kN load cell at a 1 mm/min rate. 5 samples were measured.
3. Results and discussion
3.1. Zeta potential and electrophoretic mobility of graphene and binder
The graphene used was easily suspended in water without the need
for any surfactant. Graphene suspensions with a pH between 7 and 11
appeared stable (clear black suspensions) over more than 3h (ESI
Fig. S6). Fig. 5a shows the ζ-potential as a function of pH of graphene
and binder. ζ =f(pH) of graphene conrmed the presence of Brønsted
acid surface oxides with a ζ
plateau
of −35 mV and an i.e.p., where ζ =0,
of 4.2. Similar results for graphene were reported in the literature before
[41]. A suspension is usually considered stable when |ζ| >25 mV [42].
All graphene suspensions with pH >8 were, according to this criterion,
stable. However, at pH <8 the graphene suspensions became unstable,
and most particles settled to the bottom of the vial. When emptied, some
sediment remained at the bottom of every vial, conrming that sedi-
mentation did occur. The ζ-potential of the binder remained virtually
constant over the whole pH range with an average ζ = − 48 mV, which
was likely caused by the presence of an anionic surfactant used for its
synthesis.
Fig. 5. (a) Zeta potential of graphene nanoplatelets and binder as a function of pH and (b) electrophoretic mobility of graphene and binder as a function of pH both
measured at supporting electrolyte concentration of [KCl] =1 mM.
Fig. 6. Characteristic micrographs of carbon bres EPD coated with graphene nanoplatelets using the batch process using the following process parameters (a) 10V
1min (b) 20V 1min (c) 30V 1min (d) 10V 3min and (e) 20V 3min.
O. Hubert et al.

Composites Science and Technology 217 (2022) 109126
6
The measured electrophoretic mobility, from which ζ was calculated,
gives the ability of a suspended particle to move in an applied electrical
eld. During EPD the higher the electrophoretic mobility the faster the
coating will form. Fig. 5b shows
μ
=f(pH) for graphene, binder and a 9:1
(graphene:binder weight ratio) mixture. The addition of binder to the
graphene suspension increased its electrophoretic mobility for pH <8
but had very little effect on suspensions at higher pH. The electropho-
retic mobility of a suspension that had been used for deposition was
analysed and an electrophoretic mobility of
μ
= − 2.52 ±0.09
μ
m cm/Vs
was measured, showing that there was no signicant variation of this
value after deposition.
3.2. Morphology, graphene loading and surface properties of coated
carbon bres
Micrographs of carbon bres coated with graphene at different
voltage and duration are shown Fig. 6. We can see from these micro-
graphs that the coating consists of aggregates of graphene nanoplatelets
attached to the carbon bres. When the voltage and/or time was
increased a continuous layer of graphene formed on top of the bre tow;
not encasing individual bres. During handling, some of the coating
detached and many cracks can be seen on the surface of the bre elec-
trode (Fig. 6e). A higher graphene loading will inevitably result in a
lower bre volume fraction in the prepared structural supercapacitor
electrodes and hence lower mechanical properties of the nal compos-
ite. The graphene aggregates stacking between the bres were not
affected by handling, graphene not only adhered to the carbon bres but
also stuck between the bres occupying the volume between them.
The graphene loading of bres continuously EPD coated at low
voltage was much lower as compared to the batch process because of the
larger gap between counter electrodes and carbon bres used as working
electrode. When using a voltage of 30 V (Fig. 7a), the carbon bres
started to be coated with graphene aggregates, but the amount was still
very low compared to the batch process. However, when using 40 V
(Fig. 7b), the morphology was very similar to the one observed for
carbon bres EPD coated at 10 V for 1 min in the batch process (Fig. 6a).
The specic surface area A
s
of the graphene coated carbon bres
were consistent with the amount of coating deposited as observed in the
SEM (Fig. 6); A
s
increased with increasing applied voltage and EPD time
(Fig. 8a) indicating an increased graphene loading on the low A
s
(0.49
m
2
/g) carbon bres (Fig. 8b). The gravimetric method used to determine
the graphene loading (G%) did not allow for very high precision, thus
the large standard error. After 30 s of EPD, the voltage had a large
impact on the resulting specic surface area: 33 m
2
/g at 10 V, 72 m
2
/g at
20 V and 153 m
2
/g at 30 V. However, after 60 s, the gap between the
surface areas for different applied voltage decreased. After 180 s, EPD at
Fig. 7. Characteristic micrographs of graphene nanoplatelet coated carbon bres produced using the continuous EPD process at the following conditions (a) 30V
1min and (b) 40V 1min.
Fig. 8. Specic surface area of graphene nanoplatelets coated carbon bres calculated using BET theory (a) as a function of time and (b) as a function of gra-
phene loading.
O. Hubert et al.

Composites Science and Technology 217 (2022) 109126
7
20 V and 30 V produced coated bres with identical A
s
(191 m
2
/g and
193 m
2
/g, respectively). Previous work reported specic surface areas
up to 163 m
2
/g for carbon aerogel coated carbon bres [30]. The spe-
cic surface area of the coated bres normalized to carbon aerogel was
calculated to be 741 m
2
/g. When normalising our results of coated
carbon bres to graphene and binder content resulted in an average
specic surface area of the coating material of 455 m
2
/g, conrming the
formation of graphene aggregates. Much lower specic surface areas (A
s
=36.6 ±1.5 m
2
/g) were achieved when using the continuous EPD
process as compared to the batch process, which was likely due to the
coating being removed from the bre tows when pulled over sleeves
after exiting the coating bath and during winding (see Fig. 2).
The current owing in the circuit during EPD can cause anodic
oxidation of the carbon bre electrodes. Oxidation of the surface of the
carbon bres can affect both the mechanical properties of the bres and
the adhesion of the resin to the bres, which will have an impact on the
mechanical properties of the nal composite [43]. Therefore, we char-
acterized the surface composition of the carbon bres (and graphene) by
XPS. The atomic percentages of C, N and O for each sample are reported
in Table 1. Coated CF 1 and 2 refer to the analysis of carbon bres after
removal of the coating. All the associated XP spectra can be found in the
ESI Figs. S8–9. The XPS analysis of XGnP C-750 graphene nanoplatelets
was discussed before [44]. Compared to their results we have a slightly
higher carbon content but overall the results are similar. The XPS
analysis of the graphene coating shows a higher oxygen content as
compared to virgin graphene nanoplatelets. This difference can be
explained by i) the graphene was oxidised during deposition and ii) the
presence of the binder in the coating. The XPS spectra of the industrially
oxidised AS4D carbon bres and those after EPD process were not
signicantly different. A previous study reported a variation of the
oxygen content during anodic oxidation of more than 7% [45]. We
observed less than 1% variation in oxygen content between the carbon
bres before and after EPD process. The inuence of this small variation
in chemical composition on the tensile strength of the carbon bres is
discussed below.
3.3. Electrochemical properties of small-scale supercapacitors with liquid
and solid electrolyte
Supercapacitors assembled using graphene coated carbon bres were
rst tested via cyclic voltammetry in a liquid TEABF
4
electrolyte. The
specic capacities are summarised in Fig. 9a. The capacity was divided
by the total mass of the electrodes, including the coated carbon bres
and the coating. The capacity and the specic surface area follow a
linear correlation. An areal capacitance of 21 mF/m
2
with a good cor-
relation (R
2
=0.989) was determined by tting. Supercapacitors made
using carbon aerogel coated carbon bres possessed a higher areal
capacitance in aqueous 3M KCl with values up to 132 mF/m
2
[29].
Nevertheless, aqueous electrolytes typically result in higher capacitance,
due to the pore size distribution, but limit the voltage window for
electrochemical cycling [46]. Some pores in porous carbons can be
accessed by K
+
/Cl
−
ions but not by the solvated ions dissolved in an
organic electrolyte [47,48]. Furthermore, our measured values were
acquired in a fully functional supercapacitor and not in a three electrode
setup. Therefore, as presented later, it is closer to the capacity that will
be obtained in a nal structural supercapacitor.
Fig. 9b shows a characteristic cyclic voltammogram of graphene
coated carbon bres recorded in liquid electrolyte. The voltammogram
is very close to the rectangular theoretical shape of a supercapacitor.
Only a small oxidation peak can be seen for voltages approaching 1 V
and −1 V. This peak could be caused by residual moisture still present in
the supercapacitor layup, as the assembly was not performed in a
controlled atmosphere. The charging and discharging curves are also
slightly tilted, which is due to resistive effects in the supercapacitor.
These effects should be cancelled at 0 V, from which the capacity was
calculated.
Structural supercapacitors were assembled with a polyethylene
glycol-based solid electrolyte. The specic capacity was again calculated
accounting for the weight of the coated carbon bres including the
coating (Fig. 10). All our results exceed the highest reported specic
capacity of a structural supercapacitor made using carbon aerogel
Table 1
Surface elemental composition of pristine carbon bres (AS4D), EPD coated
carbon bres and after removal of the graphene coating as well as graphene;
Atomic percentage of carbon, nitrogen and oxygen determined by XPS.
C O N
Pristine CF 88.35 9.31 2.33
Coated CF 1 87.82 10.24 1.94
Coated CF 2 87.84 10.3 1.86
Coating 91.84 7.21 0.94
Graphene 95.35 4.26 0.39
Fig. 9. (a) Specic capacity of supercapacitors made using pristine and graphene coated carbon bres in 1M TEABF
4
in PC as function of specic surface area of the
bre electrodes and (b) characteristic cyclic voltammogram of a supercapacitor prepared using graphene coated carbon bre electrodes in 1M TEABF
4
in PC recorded
at a rate of 5 mV/s.
O. Hubert et al.

Composites Science and Technology 217 (2022) 109126
8
coated carbon bres and a very similar solid polymer electrolyte [29],
indicating the potential of our electrode preparation method. The
highest specic capacity of our small-scale structural supercapacitors
was 1.44 F/g, an increase of more than 130% compared to similar de-
vices previously reported.
3.4. Separator-free structural supercapacitor demonstrators
Previous studies investigated different separator materials, such as
polypropylene (PP) separators [30] or glass bres [29,49], in structural
supercapacitors. PP separators (such as Celgard) are widely used in
supercapacitors. However, PP does not provide any additional rein-
forcement to structural energy storage devices but can cause early
delamination of multifunctional composites [30]. As an alternative,
glass bres are used in most structural energy storage composites
because they are insulators and act as reinforcement, unfortunately with
a weight penalty. The thinnest glass bre separator reported for struc-
tural energy storage application was a 50
μ
m thick glass bre fabric
(with an areal density of 53 g/m
2
) [50]. The volume occupied by the
separator is inactive in the nal assembly. Coated carbon bres act as
electrode and current collector and the matrix as electrolyte but the
separator prevents short circuits. Being able to use the matrix as a
separator would allow an increased energy density of structural super-
capacitors. We will show proof-of-concept of such a system.
Fully cured supercapacitor composites assembled using two EPD
graphene coated carbon bre electrodes had a thickness of 267 ±40
μ
m
(Fig. 11). The bre weight fraction was around 20% (around 14% bre
volume fraction), which is very low compared to standard structural
composites, which typically have bre weight fractions exceeding 40%.
For demonstration purpose, we added excess electrolyte to ensure no
electrical contact between the electrode layers. A picture of the cross-
section of such a composite can be found in ESI Fig. S7.
The measured specic capacity for these supercapacitor demon-
strators was 623 ±52 mF/g. Qian et al. [30] reported a specic capacity
of 71 mF/g for a structural supercapacitor containing two carbon aer-
ogel coated carbon bre electrodes and a glass bre separator impreg-
nated with a similar polymer electrolyte. A supercapacitor consisting of
activated carbon bre fabric electrodes and a glass bre separator but
with a slightly different polymer electrolyte system in which the ionic
liquid was replaced by a standard electrolyte (1M LiTFSI in an EC:PC
mixture), Reece et al. [51] reported a specic capacity of 102 mF/g. Our
system without separator had a 6-fold higher specic capacity compared
with their results, highlighting the benet of removing the separator.
The measured energy density and power density were 16.9 mWh/kg and
5.2 W/kg, respectively. Higher power densities have been reported
before as it only depends on the internal resistance and applied voltage
[52]. The limited power density is due to the higher resistance of carbon
bres as compared to metallic current collectors caused by the use of
carbon bres as current collectors. However, the energy density of these
demonstrators exceed the ones reported before for structural super-
capacitors [25,30]. Moreover, the use of ionic liquid could allow for
Fig. 10. Specic capacity of structural supercapacitors made using graphene
coated carbon bre electrodes and a PEG-based solid electrolyte as a function of
the specic surface area of the electrodes. The green line indicates the highest
reported value of similar devices [29]. (For interpretation of the references to
colour in this gure legend, the reader is referred to the Web version of
this article.)
Fig. 11. Photograph of a separator-free structural supercapacitor demonstrator.
Fig. 12. Normalized capacity of separator-free structural supercapacitor dem-
onstrators over 1500 cycles measured by cyclovoltammetry at a rate of 5 mV/s.
O. Hubert et al.

Composites Science and Technology 217 (2022) 109126
9
operating voltages up to 6 V if the device was protected from moisture
thus allowing to increase the energy density even further. Further im-
provements will have to address the low carbon bre volume fraction
enabling a higher energy density but also better mechanical properties.
Fig. 12 shows the evolution of the capacity of three separator-free
structural supercapacitor demonstrators over 1500 cycles. The capac-
ity was normalized to the average capacity of the 10 rst cycles. The
supercapacitors 1–3 retained 95%, 77% and 103% of their initial ca-
pacity. This cycling performance was in the range expected for super-
capacitors containing carbon based electrodes [48]. The observed
uctuations are artefacts of the calculation method which only takes
into account two points per cycle.
3.5. Single bre tensile properties of carbon bres used in structural
supercapacitors
EPD in water can damage the surfaces of carbon bres by anodic
oxidation. No clear difference in the breaking stress was measured be-
tween coated and pristine carbon bres (Fig. 13a). The values for the 18
mm gauge length seem to be deviating from the trend, probably due to
device bias when measuring at small gauge length. The same deviation is
also observed in the Weibull moduli. The Weibull moduli ranged be-
tween 4.1 and 6.4, which is typical for carbon bres [53,54]. The un-
biased Weibull moduli (m
UF
) are summarised in Fig. 13b. The graphs
containing all measured data can be found in the supplementary infor-
mation (ESI Figs. S10–11).
We conclude from these results that the composite supercapacitors
assembled with EPD graphene coated carbon bres should retain their
tensile properties compared to the composites made using pristine car-
bon bres. Moreover, surface oxidation of the carbon bres has been
shown to improve the bre-matrix interface [43]. EPD is a cheap and
easy method to coat carbon bres with active electrode materials
without drastically affecting the mechanical properties of the bre
substrates.
3.6. Tensile properties of separator-free supercapacitor demonstrators
The stress – strain curves are shown ESI Fig. S12. The calculated
breaking stress was 350 ±100 MPa and the Young’s modulus 26 ±3
GPa. This places our material in the range of previously reported
structural supercapacitors using a similar electrolyte [23]. It is
important to note that 4 out of 5 samples failed catastrophically,
showing the ability of the electrolyte to distribute the load between the
bres (ESI Fig. S13). The early failure of the fth specimen is assumed to
be linked to bre misalignment in the unidirectional laminate.
4. Conclusion
Structural supercapacitor electrodes were successfully manufactured
by electrophoretic deposition of graphene nanoplatelets onto carbon
bres. Increasing time and voltage increased the graphene loading on
the bres, which consequently resulted in higher specic surface areas
of the structural electrodes approaching 190 m
2
/g. The highest specic
capacity measured for small-scale structural composite supercapacitors
was 1.44 F/g. The use of a structural polymer electrolyte allowed for the
removal of the separator from the structural supercapacitor assembly,
thus removing what is usually the parasitic material in composite
supercapacitors. Our composite supercapacitor demonstrators had an
average specic capacity of 623 mF/g and an energy density of 16.9
mWh/kg. Finally, we demonstrated that the EPD process does not
signicantly affect the tensile properties of the carbon bres used as
substrate for deposition of active electrode materials. With the possi-
bility to coat material continuously, the process presented in this paper
will allow for production of large-scale structural composite super-
capacitors after optimisation of the process.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgements
The research leading to these results has been performed in the
framework of the HyFiSyn project, which was funded by the European
Union’s Horizon 2020 research and innovation programme under the
Marie Skłodowska-Curie grant agreement No 765881. We would also
thank Dr. Andreas Mautner for his help and support in the project, Prof.
Bo Madsen for his help with the analysis of the single bre test data and
the exchange student Maximilien Epeh Eyengue from University of
Toulon, France for his help with the experimental work.
Fig. 13. (a) Characteristic breaking stress for pristine and EPD graphene coated carbon bres. The error bars are the 90% condence interval. (b) Unbiased estimate
of Weibull moduli and associated 90% condence interval for pristine and graphene coated carbon bres determined using the unimodal Weibull analysis.
O. Hubert et al.

Composites Science and Technology 217 (2022) 109126
10
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.compscitech.2021.109126.
Author statement
OH: experimental design, experiments, data collection and analysis,
manuscript drafting and revision, NT: experiments, data collection and
testing AB: conceptualisation, data analysis manuscript writing and
revision, supervision and funding acquisition.
References
[1] M. Winter, R.J. Brodd, What are batteries, fuel cells, and supercapacitors? Chem.
Rev. 104 (2004) 4245–4270, https://doi.org/10.1021/cr020730k.
[2] O. Mashtalir, M.R. Lukatskaya, A.I. Kolesnikov, E. Raymundo-Pi˜
nero, M. Naguib,
M.W. Barsoum, Y. Gogotsi, The effect of hydrazine intercalation on the structure
and capacitance of 2D titanium carbide (MXene), Nanoscale 8 (2016) 9128–9133,
https://doi.org/10.1039/C6NR01462C.
[3] W. Johannisson, D. Zenkert, G. Lindbergh, Model of a structural battery and its
potential for system level mass savings, Multifunct. Mater. 2 (2019), 035002,
https://doi.org/10.1088/2399-7532/ab3bdd.
[4] X. Luo, D.D.L. Chung, Carbon-ber/polymer-matrix composites as capacitors,
Compos. Sci. Technol. 61 (2001) 885–888, https://doi.org/10.1016/S0266-3538
(00)00166-4.
[5] D.D. Edie, The effect of processing on the structure and properties of carbon bers,
Carbon 36 (1998) 345–362, https://doi.org/10.1016/S0008-6223(97)00185-1.
[6] A.G. Pandolfo, A.F. Hollenkamp, Carbon properties and their role in
supercapacitors, J. Power Sources 157 (2006) 11–27, https://doi.org/10.1016/j.
jpowsour.2006.02.065.
[7] E.D. Wetzel, Multifunctional composites intergrate power, communications and
structure, AMPTAC Q 8 (2004) 91–95.
[8] T800S technical data sheet, (n.d.). https://www.toraycma.com/wp-content/uploa
ds/T800S-Technical-Data-Sheet-1.pdf.pdf. (Accessed 16 September 2021)
accessed.
[9] R.A. Serway, Principles of Physics, Saunders College Pub., Fort Worth, 1998. http
://archive.org/details/principlesofphys00serw. (Accessed 16 September 2021).
accessed.
[10] W. Johannisson, D. Carlstedt, A. Nasiri, C. Buggisch, P. Linde, D. Zenkert, L.E. Asp,
G. Lindbergh, B. Fiedler, A screen-printing method for manufacturing of current
collectors for structural batteries, Multifunct. Mater. 4 (2021), 035002, https://doi.
org/10.1088/2399-7532/ac2046.
[11] L. Christodoulou, J.D. Venables, Multifunctional material systems: the rst
generation, JOM 55 (2003) 39–45, https://doi.org/10.1007/s11837-003-0008-z.
[12] J.T. South, R.H. Carter, J.F. Snyder, C.D. Hilton, D.J. O’Brien, E.D. Wetzel,
Multifunctional power-generating and energy-storing structural composites for U.
S. Army applications, MRS Proc 851 (2004), https://doi.org/10.1557/PROC-851-
NN4.6. NN4.6.
[13] D.J. O’Brien, D.M. Baechle, E.D. Wetzel, Multifunctional structural composite
capacitors for U.S. Army applications, in: SAMPE Fall Tech. Conf. Proc., Society for
the Advancement of Material and Process Engineering, Dallas, TX, 2006, p. 11.
[14] A. Bernasconi, Aircraft with electric batteries, in particular a hybrid aircraft, US
2019/0263498 A1. https://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Se
ct2=HITOFF&d=PG01&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.
html&r=1&f=G&l=50&s1=%2220190263498%22.PGNR.&OS=DN/20190
263498&RS=DN/20190263498, 2019. (Accessed 17 May 2021) accessed.
[15] S.N. Nguyen, A. Millereux, A. Pouyat, E.S. Greenhalgh, M.S.P. Shaffer, A.R.J.
Kucernak, P. Linde, Conceptual multifunctional design, feasibility and
requirements for structural power in aircraft cabins, J. Aircr. 0 (n.d.) 1–11. https://
doi.org/10.2514/1.C036205.
[16] H.I. Becker, Low Voltage Electrolytic Capacitor, US2800616A, 1957.
[17] B.E. Conway, Transition from “supercapacitor” to “battery” behavior in
electrochemical energy storage, J. Electrochem. Soc. 138 (1991) 1539, https://doi.
org/10.1149/1.2085829.
[18] Y. Zhang, H. Feng, X. Wu, L. Wang, A. Zhang, T. Xia, H. Dong, X. Li, L. Zhang,
Progress of electrochemical capacitor electrode materials: a review, Int. J. Hydrog.
Energy. 34 (2009) 4889–4899, https://doi.org/10.1016/j.ijhydene.2009.04.005.
[19] G. Wang, L. Zhang, J. Zhang, A review of electrode materials for electrochemical
supercapacitors, Chem. Soc. Rev. 41 (2012) 797–828, https://doi.org/10.1039/
C1CS15060J.
[20] C. Du, N. Pan, Supercapacitors using carbon nanotubes lms by electrophoretic
deposition, J. Power Sources 160 (2006) 1487–1494, https://doi.org/10.1016/j.
jpowsour.2006.02.092.
[21] K. Sheng, Y. Sun, C. Li, W. Yuan, G. Shi, Ultrahigh-rate supercapacitors based on
electrochemically reduced graphene oxide for ac line-ltering, Sci. Rep. 2 (2012)
1–5, https://doi.org/10.1038/srep00247.
[22] Z. Li, B. Song, Z. Wu, Z. Lin, Y. Yao, K.-S. Moon, C.P. Wong, 3D porous graphene
with ultrahigh surface area for microscale capacitive deionization, Nano Energy 11
(2015) 711–718, https://doi.org/10.1016/j.nanoen.2014.11.018.
[23] S. Wang, B. Pei, X. Zhao, R.A.W. Dryfe, Highly porous graphene on carbon cloth as
advanced electrodes for exible all-solid-state supercapacitors, Nano Energy 2
(2013) 530–536, https://doi.org/10.1016/j.nanoen.2012.12.005.
[24] M. Wang, L.D. Duong, N.T. Mai, S. Kim, Y. Kim, H. Seo, Y.C. Kim, W. Jang, Y. Lee,
J. Suhr, J.-D. Nam, All-solid-state reduced graphene oxide supercapacitor with
large volumetric capacitance and ultralong stability prepared by electrophoretic
deposition method, ACS Appl. Mater. Interfaces 7 (2015) 1348–1354, https://doi.
org/10.1021/am507656q.
[25] X.F. S´
anchez-Romate, A.D. Bosque, J. Artigas-Arnaudas, B.K. Mu˜
noz, M. S´
anchez,
A. Ure˜
na, A proof of concept of a structural supercapacitor made of graphene
coated woven carbon bers: EIS study and mechanical performance,
Electrochimica Acta 370 (2021) 137746, https://doi.org/10.1016/j.
electacta.2021.137746.
[26] J.F. Snyder, E.L. Wong, C.W. Hubbard, Evaluation of commercially available
carbon bers, fabrics, and papers for potential use in multifunctional energy
storage applications, J. Electrochem. Soc. 156 (2009) A215, https://doi.org/
10.1149/1.3065070.
[27] A. Javaid, K.K.C. Ho, A. Bismarck, M.S.P. Shaffer, J.H.G. Steinke, E.S. Greenhalgh,
Multifunctional structural supercapacitors for electrical energy storage
applications, J. Compos. Mater. 48 (12) (2013) 1409–1416, https://doi.org/
10.1177/0021998313487239.
[28] H. Qian, H. Diao, N. Shirshova, E.S. Greenhalgh, J.G.H. Steinke, M.S.P. Shaffer,
A. Bismarck, Activation of structural carbon bres for potential applications in
multifunctional structural supercapacitors, J. Colloid Interface Sci. 395 (2013)
241–248, https://doi.org/10.1016/j.jcis.2012.12.015.
[29] N. Shirshova, H. Qian, M. Houll´
e, J.H.G. Steinke, A.R.J. Kucernak, Q.P.V. Fontana,
E.S. Greenhalgh, A. Bismarck, M.S.P. Shaffer, Multifunctional structural energy
storage composite supercapacitors, Faraday Discuss 172 (2014) 81–103, https://
doi.org/10.1039/C4FD00055B.
[30] H. Qian, A.R. Kucernak, E.S. Greenhalgh, A. Bismarck, M.S.P. Shaffer,
Multifunctional structural supercapacitor composites based on carbon aerogel
modied high performance carbon ber fabric, ACS Appl. Mater. Interfaces. 5
(2013) 6113–6122, https://doi.org/10.1021/am400947j.
[31] I. Corni, M.P. Ryan, A.R. Boccaccini, Electrophoretic deposition: from traditional
ceramics to nanotechnology, J. Eur. Ceram. Soc. 28 (2008) 1353–1367, https://
doi.org/10.1016/j.jeurceramsoc.2007.12.011.
[32] H.C. Hamaker, Formation of a deposit by electrophoresis, Trans. Faraday Soc. 35
(1940) 279–287, https://doi.org/10.1039/TF9403500279.
[33] J. Hagberg, H.A. Maples, K.S.P. Alvim, J. Xu, W. Johannisson, A. Bismarck,
D. Zenkert, G. Lindbergh, Lithium iron phosphate coated carbon ber electrodes
for structural lithium ion batteries, Compos. Sci. Technol. 162 (2018) 235–243,
https://doi.org/10.1016/j.compscitech.2018.04.041.
[34] C. Wang, J. Li, S. Sun, X. Li, F. Zhao, B. Jiang, Y. Huang, Electrophoretic deposition
of graphene oxide on continuous carbon bers for reinforcement of both tensile
and interfacial strength, Compos. Sci. Technol. 135 (2016) 46–53, https://doi.org/
10.1016/j.compscitech.2016.07.009.
[35] H. Diao, A. Bismarck, P. Robinson, M.R. Wisnom, Production OF continuous
intermingled CF/GF hybrid composite via bre tow spreading technology 9 (2014).
[36] K.K.C. Ho, S.-R. Shamsuddin, S. Riaz, S. Lamorinere, M.Q. Tran, A. Javaid,
A. Bismarck, Wet impregnation as route to unidirectional carbon bre reinforced
thermoplastic composites manufacturing, Plast. Rubber Compos. 40 (2011)
100–107, https://doi.org/10.1179/174328911X12988622801098.
[37] A.V. Delgado, F. Gonz´
alez-Caballero, R.J. Hunter, L.K. Koopal, J. Lyklema,
Measurement and interpretation of electrokinetic phenomena (IUPAC technical
report), Pure Appl. Chem. 77 (2005) 1753–1805, https://doi.org/10.1351/
pac200577101753.
[38] D.C. Henry, A. Lapworth, The cataphoresis of suspended particles. Part I.—the
equation of cataphoresis, Proc. R. Soc. Lond. Ser. Contain. Pap. Math. Phys.
Character. 133 (1931) 106–129, https://doi.org/10.1098/rspa.1931.0133.
[39] W. Weibull, Statistical Theory of Strength of Materials. Generalstabens Litograska
Anstalts F¨
orlag, Stockholm, 1939.
[40] W. Weibull, A statistical distribution function of wide applicability, ASME J. Appl.
Mech. 18 (1951) 293–297.
[41] M. Mehrali, E. Sadeghinezhad, S.T. Latibari, S.N. Kazi, M. Mehrali, M.N.B.M. Zubir,
H.S.C. Metselaar, Investigation of thermal conductivity and rheological properties
of nanouids containing graphene nanoplatelets, Nanoscale Res. Lett. 9 (2014) 15,
https://doi.org/10.1186/1556-276X-9-15.
[42] W. Yu, H. Xie, A review on nanouids: preparation, stability mechanisms, and
applications, J. Nanomater. 2012 (2011), e435873, https://doi.org/10.1155/
2012/435873.
[43] E. Fitzer, H. J¨
ager, N. Popovska, F. Von Sturm, Anodic oxidation of high modulus
carbon bres in sulphuric acid, J. Appl. Electrochem. 18 (1988) 178–182, https://
doi.org/10.1007/BF01009259.
[44] S. Dul, L. Fambri, C. Merlini, G.M.O. Barra, M. Bersani, L. Vanzetti, A. Pegoretti,
Effect of graphene nanoplatelets structure on the properties of
acrylonitrile–butadiene–styrene composites, Polym. Compos. 40 (2019)
E285–E300, https://doi.org/10.1002/pc.24645.
[45] J. Jiang, X. Yao, C. Xu, Y. Su, L. Zhou, C. Deng, Inuence of electrochemical
oxidation of carbon ber on the mechanical properties of carbon ber/graphene
oxide/epoxy composites, Compos. Part Appl. Sci. Manuf. 95 (2017) 248–256,
https://doi.org/10.1016/j.compositesa.2017.02.004.
[46] R. K¨
otz, M. Carlen, Principles and applications of electrochemical capacitors,
Electrochimica Acta 45 (2000) 2483–2498, https://doi.org/10.1016/S0013-4686
(00)00354-6.
[47] A. Burke, Ultracapacitors: why, how, and where is the technology, J. Power
Sources 91 (2000) 37–50, https://doi.org/10.1016/S0378-7753(00)00485-7.
O. Hubert et al.

Composites Science and Technology 217 (2022) 109126
11
[48] J. Zhao, A.F. Burke, Electrochemical capacitors: materials, technologies and
performance, Energy Storage Mater 36 (2021) 31–55, https://doi.org/10.1016/j.
ensm.2020.12.013.
[49] A. Javaid, K.K.C. Ho, A. Bismarck, J.H.G. Steinke, M.S.P. Shaffer, E.S. Greenhalgh,
Improving the multifunctional behaviour of structural supercapacitors by
incorporating chemically activated carbon bres and mesoporous silica particles as
reinforcement, J. Compos. Mater. 52 (22) (2018) 3085–3097, https://doi.org/
10.1177/0021998318761216.
[50] L.E. Asp, K. Bouton, D. Carlstedt, S. Duan, R. Harnden, W. Johannisson,
M. Johansen, M.K.G. Johansson, G. Lindbergh, F. Liu, K. Peuvot, L.M. Schneider,
J. Xu, D. Zenkert, A structural battery and its multifunctional performance, Adv.
Energy Sustain. Res. 2 (2021) 2000093, https://doi.org/10.1002/aesr.202000093.
[51] R. Reece, C. Lekakou, P.A. Smith, A structural supercapacitor based on activated
carbon fabric and a solid electrolyte, Mater. Sci. Technol. 35 (2019) 368–375,
https://doi.org/10.1080/02670836.2018.1560536.
[52] N. Shirshova, H. Qian, M.S.P. Shaffer, J.H.G. Steinke, E.S. Greenhalgh, P.T. Curtis,
A. Kucernak, A. Bismarck, Structural composite supercapacitors, Compos. Part
Appl. Sci. Manuf. 46 (2013) 96–107, https://doi.org/10.1016/j.
compositesa.2012.10.007.
[53] M.G. Huson, J.S. Church, A.A. Ka, A.L. Woodhead, J. Khoo, M.S.R.N. Kiran, J.
E. Bradby, B.L. Fox, Heterogeneity of carbon bre, Carbon 68 (2014) 240–249,
https://doi.org/10.1016/j.carbon.2013.10.084.
[54] J. Sun, F. Zhao, Y. Yao, Z. Jin, X. Liu, Y. Huang, High efcient and continuous
surface modication of carbon bers with improved tensile strength and interfacial
adhesion, Appl. Surf. Sci. 412 (2017) 424–435, https://doi.org/10.1016/j.
apsusc.2017.03.279.
O. Hubert et al.