nanomaterials
Review
Graphene/Reduced Graphene Oxide-Carbon Nanotubes
Composite Electrodes: From Capacitive to
Battery-Type Behaviour
Olena Okhay 1, * and Alexander Tkach 2






Citation: Okhay, O.; Tkach, A.
Graphene/Reduced Graphene
Oxide-Carbon Nanotubes Composite
Electrodes: From Capacitive to
Battery-Type Behaviour.
Nanomaterials 2021, 11, 1240. https://
doi.org/10.3390/nano11051240
Academic Editor: Albert G. Nasibulin
Received: 21 April 2021
Accepted: 4 May 2021
Published: 8 May 2021
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Nanomaterials 2021, 11, 1240. https://doi.org/10.3390/nano11051240
1 TEMA-Center for Mechanical Technology and Automation, Department of Mechanical Engineering,
University of Aveiro, 3810-193 Aveiro, Portugal
2 CICECO-Aveiro Institute of Materials, Department of Materials and Ceramic Engineering,
University of Aveiro, 3810-193 Aveiro, Portugal; atkach@ua.pt
* Correspondence: olena@ua.pt
Abstract: Thanks to the advanced technologies for energy generation such as solar cells and thermo-
or piezo-generators the amount of electricity transformed from light, heat or mechanical pressure
sources can be significantly enhanced. However, there is still a demand for effective storage devices
to conserve electrical energy which addresses the wide range of large stationary applications from
electric vehicles to small portable devices. Among the large variety of energy-storage systems
available today, electrochemical energy sources and, in particular, supercapacitors (SC), are rather
promising in terms of cost, scaling, power management, life cycle and safety. Therefore, this review
surveys recent achievements in the development of SC based on composites of such carbon-derived
materials as graphene (G) and reduced graphene oxide (rGO) with carbon nanotubes (CNT). Various
factors influencing the specific capacitance are discussed, while specific energy and power as well as
cycling stability of SC with G/rGO-CNT composite electrode materials are overviewed.
Keywords: energy storage; multifunctional materials; composites; graphene; nanostructures
1. Introduction
Besides the energy transformation, energy storage is one of the most important topics
of scientific research today. Similar to the existing commercial batteries, electrochemical
capacitors (often called supercapacitors (SC)) are widely studied for their commercial
application in electrical cars, portable electronics, etc. The most famous capacitive materials
are carbon-based compounds, particularly activated carbon (AC) that is already used for
energy storage due to its large surface area and low cost [1]. However, many carbon atoms
in AC cannot be accessed by electrolyte ions, thus being wasted in terms of activating their
electrochemical functions, due to a very irregular shape of AC, shown in Figure 1a [2].
That decreases the capacitance of the AC electrodes as well as reducing their electrical
conductivity (see Table 1).
In addition to AC, a graphene (G), being 2D sp2 -hybridized carbon sheet, also has a
large surface area as shown in Table 1. However, its electrical conductivity is significantly
higher, making graphene-related materials very promising for energy storage [3–9]. How-
ever, from G it is easy to form aggregates due to intensive π–π interaction or to restacking
forming graphite through van der Waals interactions. If G sheets are stacked together, the
electrolyte ions have difficulties gaining access to the inner layers to form electrochemical
double layers. Thus, the ions could be accumulated only on the top and bottom surfaces
of the sheet agglomerates that can lead to a lower specific capacitance. The problem of
graphene sheet agglomerations during the preparation process is valid also for multilayer
reduced graphene oxide (rGO) despite lower surface area compared to that of monolayer
G. However, the lower price and properties tunable by oxidation degree as well as high
https://www.mdpi.com/journal/nanomaterials
Nanomaterials 2021,
Nanomaterials 11,11,
2021, 1240
x FOR PEER REVIEW 2 of 2
30of 30

electrical
G. However,conductivity
the lowerand synergistic
price effectstunable
and properties in composite materials
by oxidation alsoas
degree make
well rGO useful
as high
electrical conductivity and
for electrode development. synergistic effects in composite materials also make rGO useful
for electrode development.
Table 1. Comparison of theoretical (theor.) and/or experimental (exp.) values of specific surface area
Table electrical
(SSA), 1. Comparison of theoretical
conductivity (theor.)
(σ) and and/or
specific experimental
capacitance (Csp )(exp.) values of
of different specific surface
carbonaceous materials
area (SSA), electrical conductivity (σ) and specific capacitance (Csp) of different carbonaceous ma-
such as activated carbon, graphene and single-wall carbon nanotubes (SWCNT).
terials such as activated carbon, graphene and single-wall carbon nanotubes (SWCNT).
ActivatedCarbon
Activated Carbon Graphene
Graphene SWCNT
SWCNT
m22/g
SSA,m
SSA, /g ~3000
~3000 (theor.[3])
(theor. [3]) ∼2600
∼2600(theor.
(theor. [5])
[5]) 1315
1315 (theor.
(theor. [5])
[5])
~10 8 (theor. [6])
S/m
σ,S/m 4.6××1010 −6 (exp. [4]) ~10 (theor. [6])
8 6
σ, 4.6 −6 (exp. [4]) ~103 3(exp.
(exp.[6])
[6]) ~10~10 (exp.
6 (exp. [8])[8])
~10
C , F/g 200 (exp. [3]) ~550 (theor. [7]) 180 (exp. [9])
Cspsp, F/g 200 (exp. [3]) ~550 (theor. [7]) 180 (exp. [9])

Figure1.1.Model
Figure Model of
ofactivated
activated carbon
carbon based
based on
on fullerene-like
fullerene-like elements
elements (a)
(a) (Reproduced
(Reproduced with
with permis-
permission
sion of [2]. Copyright Royal Society of Chemistry, 2007) and schematic model of reduced graphene
of [2]. Copyright Royal Society of Chemistry, 2007) and schematic model of reduced graphene oxide
oxide (rGO)/carbon nanotubes (CNT) hybrid structure (b) with blue and yellow spheres corre-
(rGO)/carbon nanotubes
sponding to carbon atoms(CNT)
of rGOhybrid structure
and CNT, (b) withasblue
respectively, welland yellow
as white spheres
and corresponding
red spheres repre- to
carbon atoms of rGO and CNT, respectively, as well as white and red spheres representing
senting hydroxyl group hydrogen and oxygen atoms, respectively (Reproduced with permission hydroxyl
of [10].hydrogen
group Copyrightand
Elsevier,
oxygen2015).
atoms, respectively (Reproduced with permission of [10]. Copyright
Elsevier, 2015).
As one of the possible ways to prevent the sheets from sticking together is a use of
As one of
1D material thecarbon
like possible ways to prevent
nanotubes (CNT) asthe sheets from
a spacer sticking
(see Figure 1b)together
[10]. CNT is ahave
use ofa 1D
readily accessible
material like carbon surface area as(CNT)
nanotubes well asashigh conductivity
a spacer (see 1b)
(see Figure Table 1). CNT
[10]. Since have
the surface
a readily
area is known
accessible to increase
surface area as with
well decrease of the CNT diameter,
as high conductivity (see Table single-wall
1). Since carbon nano-area
the surface
istubes
known (SWCNT) are expected
to increase to haveof
with decrease higher specific
the CNT capacitance
diameter, compared
single-wall to multiwall
carbon nanotubes
carbon nanotubes
(SWCNT) are expected(MWCNT)to have[5]. However,
higher specificCNT often stack
capacitance in bundles
compared and only carbon
to multiwall the
outermost (MWCNT)
nanotubes portion of CNT can function
[5]. However, CNT for often
ion absorption, whereasand
stack in bundles the only
innerthecarbon at-
outermost
oms are of
portion notCNT
involved
can in the process.
function From
for ion another side,
absorption, pristinethe
whereas CNT with carbon
inner preserved elec- are
atoms
tronic
not structures
involved canprocess.
in the be easilyFrom
dispersed
another in graphene oxide
side, pristine CNT(GO) solution
with without
preserved any
electronic
additives and generate clean, electrically addressable carbon–carbon
structures can be easily dispersed in graphene oxide (GO) solution without any additives interfaces. Thus, GO
is a “surfactant” to directly disperse CNT, while CNT can prevent
and generate clean, electrically addressable carbon–carbon interfaces. Thus, GO is a the aggregation of gra-
phene sheetstoasdirectly
“surfactant” a spacer, and has
disperse highwhile
CNT, conductivity,
CNT canhigh surface
prevent area, and prospective
the aggregation of graphene
mechanical properties. Moreover, non-conductive GO without
sheets as a spacer, and has high conductivity, high surface area, and prospective any thermal or chemical
mechanical
reduction was
properties. found tonon-conductive
Moreover, be a promising GO material for SC
without anybased
thermalon aorsimulation, which
chemical reduction
claimed its capacitance decreases with increasing oxidation state [11]. Currently, there are
was found to be a promising material for SC based on a simulation, which claimed its
a number of publications on G/rGO-CNT composites with different final parameters as
capacitance decreases with increasing oxidation state [11]. Currently, there are a number
promising electrode materials for energy storage applications, many of which are re-
of publications on G/rGO-CNT composites with different final parameters as promising
viewed and analysed in this work.
electrode materials for energy storage applications, many of which are reviewed and
analysed in this work.
2. General Information on Energy-Storage Materials
Capacitors
2. General and batteries
Information as well as supercapacitors
on Energy-Storage Materials (electrochemical capacitors) can
store a charge, possessing, however, different charge storage mechanisms. The dielectric
Capacitors and batteries as well as supercapacitors (electrochemical capacitors) can
store a charge, possessing, however, different charge storage mechanisms. The dielectric
capacitor has electrostatic storage, whereas a battery can be characterized by diffusion
Nanomaterials 2021, 11, x FOR PEER REVIEW 3 of 30
Nanomaterials 2021, 11, 1240 3 of 30

capacitor has electrostatic storage, whereas a battery can be characterized by diffusion

mechanism
mechanism of of storage
storage basedbased on on reduction-oxidation
reduction-oxidation (redox) (redox) processes
processes from from used
used faradaic
faradaic
materials. Moreover, there are clear differences in cyclic voltammograms
materials. Moreover, there are clear differences in cyclic voltammograms (CV), providing (CV), providing
the
the current
current response
response to to aa linearly
linearly cycled
cycled potential
potential sweep,
sweep, and and galvanostatic
galvanostatic charge charge and and
discharge (GCD) curves, defining how fast a cell is charged
discharge (GCD) curves, defining how fast a cell is charged or discharged, for these or discharged, for these elec-
trodes.
electrodes.Typically,
Typically,both battery
both battery electrodes,
electrodes, anode
anodeand andcathode,
cathode,include
includefaradaic
faradaic materials
materials
that result in strong redox peaks in CV, which are clearly visible
that result in strong redox peaks in CV, which are clearly visible both for a single electrode both for a single electrode
and
and inin the
the full
full cell
cell (battery)
(battery) measurements.
measurements. Moreover,Moreover, the the GCD
GCD process
process of of such
such electrodes
electrodes
as well as devices has long and wide plateaus that
as well as devices has long and wide plateaus that can be clearly observed. Bycan be clearly observed. By contrast
contrast
with a battery, a dielectric capacitor that stores energy by means
with a battery, a dielectric capacitor that stores energy by means of a static charge presents of a static charge presents
an
an ideal
ideal rectangular
rectangular shape shape of of CV, the area
CV, the area ofof which
which increases
increases withwith thethe potential
potential scan scan rate,
rate,
and
and triangular
triangular shapeshape of of GCD.
GCD. Moreover,
Moreover, in in dielectric
dielectric capacitors,
capacitors, the the current
current (i) (i) flowing
flowing
through a cell is proportional to the linear variation rate of the the voltage
voltage (v) (v) as
as i~v
i~v [12].
At the
the same
sametime,time,existing
existingsupercapacitors,
supercapacitors, which can store the energy
which can store the energy by electrochem- by electro-
chemical reactions, include electric double-layer capacitors
ical reactions, include electric double-layer capacitors (EDLC) and pseudocapacitors as (EDLC) and pseudocapacitors
as
cancan
be beseenseen in Figure
in Figure 2. In2.EDLC,
In EDLC, purepure electrostatic
electrostatic charge charge
storagestorage
occurs occurs
at theatelectrode–
the elec-
trode–electrolyte
electrolyte interface interface (see Figure
(see Figure 2a) [13]2a) and[13] and
their their rectangular
almost almost rectangular CV increases
CV increases linearly
linearly
with thewithscan the
ratescan
and rate
has and has symmetric
symmetric triangular triangular
GCD (seeGCD Figure(see3,Figure
left). In3,pseudocapaci-
left). In pseu-
docapacitors,
tors, which can which
involve canfaradaic
involve materials
faradaic materials
such as metal suchoxides
as metal oxides or conductive
or conductive polymers,
fast and reversible
polymers, fast and surface
reversible faradaic
surface process
faradaic such as electron
process such as charge-transfer between elec-
electron charge-transfer be-
trolyte and electrode occurs on/near to the electrode surface (see
tween electrolyte and electrode occurs on/near to the electrode surface (see Figure 2b) that Figure 2b) that can lead
to visible
can lead tosmall
visiblebulges
smallinbulges
CV and small
in CV and deviations in GCDin(see
small deviations GCD Figure 3, middle).
(see Figure More-
3, middle).
over, pseudocapacitive
Moreover, pseudocapacitive electrodes showshow
electrodes higher capacitance
higher capacitance valuevaluethan EDLC
than EDLC thanks
thanksto
thethe
to faradaic-electron
faradaic-electron transfer
transfer known
known as as
reversible
reversible surface
surface redox
redoxreactions
reactions in addition
in addition to
thethe
to non-faradaic
non-faradaic charge
chargestorage
storage on onthethe
surface
surfaceobserved
observed in EDLC.
in EDLC. However,
However, the theexcessive
exces-
increase
sive of theoffaradaic
increase the faradaiccontribution
contribution(deep(deep
intercalation in electrode
intercalation material
in electrode (see Figure
material 2c)
(see Fig-
can not only increase the capacitance but also dominate diffusion
ure 2c) can not only increase the capacitance but also dominate diffusion kinetics in the kinetics in the electrodes
that is manifested
electrodes as redox peaks
that is manifested shifted
as redox withshifted
peaks scan rate within scan
CV patterns,
rate in CV as patterns,
plateaus in as GCD
plat-
curves
eaus in (see
GCDFigure
curves 3, (see
right), and 3,
Figure in right),
longerand charge/discharge time [14].
in longer charge/discharge time [14].

Figure 2. Illustration
Illustrationofofthe
theelectrode
electrodeprocesses
processesoccurring
occurringatatelectrical double-layer
electrical double-layercapacitive (a),(a),
capacitive
pseudocapacitive
pseudocapacitive (b),
(b), and
and faradaic
faradaic(c)
(c)electrodes
electrodes(Reproduced
(Reproducedwith withpermission
permissionofof[13].
[13].Copyright
Copyright
Wiley, 2019).
Wiley, 2019).

Thus, according to Figure 3, the shape of CV can be rectangular-like for EDLC or can
show small redox peaks for pseudocapacitive composites including faradaic materials but
 with dominant EDLC charge storage. Faradaic capacitance electrodes, being also compo-
sites of capacitive materials with faradaic material similar to that of battery electrodes, are
characterized by distinct and widely separated peaks in CV with an increase of scan rate.
GCD curves of single battery-type electrodes are profoundly non-linear and characterized
Nanomaterials 2021, 11, 1240
by plateaus of nearly constant potential corresponding to the potentials, at which the far-
4 of 30
adaic reduction or oxidation of the metal centres, etc. is occurring in contrast to GCD of
pseudocapacitor with slight surface redox on the top of the electrode.

Figure 3. Schematic
Figure 3. Schematic view
view of
of typically
typicallyreported
reportedcyclic
cyclicvoltammograms
voltammograms(CV)
(CV)and
andgalvanostatic
galvanostaticcharge
and discharge (GCD) curves and applicable information for the G/rGO-CNT electrodes.
charge and discharge (GCD) curves and applicable information for the G/rGO-CNT electrodes.

Thus, according
The total currenttoofFigure 3, the shape of
CV measurements CV can
under be rectangular-like
a potential sweep rate for EDLC
(i(v)) or can
for compo-
show
sites with faradaic materials consists of two parts [15]. One part is a current related tobut
small redox peaks for pseudocapacitive composites including faradaic materials the
with dominant EDLC charge storage. Faradaic capacitance electrodes, being also compos-
double-layer charge at the electrolyte interface or to initial fast faradaic reactions on the
ites of capacitive materials with faradaic material similar to that of battery electrodes, are
exposed electrode surface (icap). Another part is the current related to the slow diffusion-
characterized by distinct and widely separated peaks in CV with an increase of scan rate.
controlled process (idif). The capacitive contribution and the diffusion-controlled contribu-
GCD curves of single battery-type electrodes are profoundly non-linear and characterized
tion can be calculated following a power-law relationship with the sweep rate (v) accord-
by plateaus of nearly constant potential corresponding to the potentials, at which the
ing to Equation (1):
faradaic reduction or oxidation of the metal centres, etc. is occurring in contrast to GCD of
pseudocapacitor with slight surface redox i(v) = icapon+the
idif =top
a ×ofvbthe
, electrode. (1)
The total current of CV measurements under a potential sweep rate (i(v)) for com-
where awith
posites and faradaic
b are variable parameters.
materials consistsbof values
two partscan be[15].
estimated
One part fromis athe slope of
current log(i)
related
vs. log(v) plot. The value b = 1 corresponds to the presence of the fast surface
to the double-layer charge at the electrolyte interface or to initial fast faradaic reactions redox reaction
andthe
on charge/discharge process
exposed electrode inherent
surface (icap ).to Another
EDLC when partdiffusion contribution
is the current relatedistoabsent and
the slow
CV show linear current
diffusion-controlled response
process (idif ). dependency
The capacitive oncontribution
the scan rateand (i~v)the
[14]. At the same time,
diffusion-controlled
the peak current
contribution response
can be of a following
calculated battery-type electrode with
a power-law strong redox
relationship with peaks
the sweepwill rate
be pro-
(v)
portional to the square
according to Equation (1): root of the scan rate (i~v 1/2) and in this case b = 0.5.

Thus, the number of the electrodes i(v) = icapcombining vb ,

+ idif = a ×capacitive and faradaic materials (1) re-
veals 0.5 < b < 1, exhibiting either strong capacitive or battery-type behaviour. Moreover,
where a and b are variable parameters. b values can be estimated from the slope of log(i) vs.
as also seen from Figure 3, electrodes with 0.8 < b < 1 are considered as pseudocapacitive
log(v) plot. The value b = 1 corresponds to the presence of the fast surface redox reaction
materials having predominantly capacitive storage [16,17] in contrast to the electrodes
and charge/discharge process inherent to EDLC when diffusion contribution is absent
with 0.5 < b < 0.8 with dominant faradaic (battery-type) behaviour. Thus, to understand
and CV show linear current response dependency on the scan rate (i~v) [14]. At the same
the dominant storage mechanism in composite electrode (capacitive or faradaic/battery-
time, the peak current response of a battery-type electrode with strong redox peaks will be
proportional to the square root of the scan rate (i~v1/2 ) and in this case b = 0.5.
Thus, the number of the electrodes combining capacitive and faradaic materials reveals
0.5 < b < 1, exhibiting either strong capacitive or battery-type behaviour. Moreover, as
also seen from Figure 3, electrodes with 0.8 < b < 1 are considered as pseudocapacitive
materials having predominantly capacitive storage [16,17] in contrast to the electrodes with
0.5 < b < 0.8 with dominant faradaic (battery-type) behaviour. Thus, to understand the
dominant storage mechanism in composite electrode (capacitive or faradaic/battery-type),
Nanomaterials 2021, 11, 1240 5 of 30
Nanomaterials 2021, 11, x FOR PEER REVIEW 5 of 30

the parameter
type), b needsb to
the parameter be calculated
needs in addition
to be calculated to correct
in addition andand
to correct deepdeep analysis
analysisofofCV
atCV
different scan rates, GCD curves, as well as electrochemical impedance
at different scan rates, GCD curves, as well as electrochemical impedance spectra (EIS) spectra (EIS)
including the slope of Nyquist plot,
including the slope of Nyquist plot, etc. etc.
Moreover,
Moreover,sincesincethe
the energy storagecell
energy storage cellconsists
consists of of
twotwo electrodes,
electrodes, different
different combi-
combina-
nation of electrodes can give a device with different charge storage
tion of electrodes can give a device with different charge storage mechanism as can mechanism as can
bebe
seen
seenininFigure
Figure4.4.According
According to to Figure
Figure 4,4, the combinationofoftwo
the combination twoEDLC
EDLCelectrodes
electrodes oror two
two
pseudocapacitive
pseudocapacitiveelectrodes
electrodesresults
resultsinina asymmetric
symmetriccapacitive
capacitivedevice.
device.At Atthe
thesame
sametime,
time,an
asymmetric
an asymmetric design of storage
design of storagedevice, consisting
device, consisting of the EDLC
of the EDLC electrode
electrode together
togetherwith
withthe
pseudocapacitive
the pseudocapacitive one, can
one,gain fromfrom
can gain the enlarged
the enlarged voltage window
voltage window covering
covering thethe
windows
win-
ofdows
two different electrodes,
of two different still keeping
electrodes, the capacitive
still keeping mechanism
the capacitive mechanism of charge
of charge storage as a
storage
dominant one [18].
as a dominant oneHowever, in the in
[18]. However, casetheofcase
combining two electrodes
of combining with different
two electrodes storage
with different
mechanism (EDLC and
storage mechanism battery-type
(EDLC or pseudocapacitor
and battery-type and battery-type)
or pseudocapacitor this mechanism
and battery-type) this
cannot be keptcannot
mechanism and thebe final hybrid
kept and the device will store
final hybrid energy
device in a energy
will store differentin way to capacitive
a different way
to capacitive
materials. materials.
Based on that,Based
energy onand
that,power
energydensities
and power densities
of hybrid of hybrid
devices must devices must
be compared
be compared
directly directly
with that withfor
reported thatbatteries,
reportedsince
for batteries, since their electrochemical
their electrochemical performanceperfor- must be
mance
higher inmust be higher
advance in advance
compared to thatcompared
of symmetric to that
andofasymmetric
symmetric and asymmetric super-
supercapacitors because
ofcapacitors because of
faradaic materials, faradaic
which materials, in
are dominant which are dominantelectrodes.
the battery-type in the battery-type
Thus, the elec-
charge
trodes.mechanism
storage Thus, the charge storage
of single mechanism
electrode needsof tosingle electrode
be clarified needs
before to be clarified
fabrication of thebefore
energy
fabrication
storage device.of The
the best
energy waystorage device. The
to understand it is best way to understand
the detailed it is characteristics,
analysis of their the detailed
analysis of CV,
particularly theirb characteristics,
parameter, GCD, particularly
Nyquist CV, plot.b parameter, GCD, Nyquist plot.

Figure
Figure 4. Schematic
4. Schematic view
view ofofpossible
possiblecombinations
combinationsof
of electrodes
electrodes and final
final devices
devicesfabricated
fabricatedon
onthem.
them.
Nanomaterials 2021, 11, 1240 6 of 30

3. Composite Capacitive Electrodes Based on Graphene/Reduced Graphene

Oxide-Carbon Nanotubes (G/rGO-CNT)
Both G/rGO and CNT are capacitive materials and their composite electrodes also
present EDLC type of energy storage with specific capacitance values varying up to
375 F/g [19] as shown in Table S1 in Supplementary Materials. Typically, rGO-CNT-based
electrodes show not an ideal but almost a rectangular shape of CV with slight deviation
due to an electrode resistance contribution, including ion transfer resistance or electrode
material resistance, or even some surface redox associated with oxygen functional groups
especially at high scan rate.
Enhancement of the electrochemical performance G/rGO-CNT electrodes and, respec-
tively, a final device, can be undertaken not only by addition of external components, but
also by modification/functionalization of G/rGO and/or CNT. Thus, the properties of
graphene-based materials are known to depend on the synthesis method and reduction
process. In particular, for the preparation of mixed G/rGO-CNT-based composites, carbon
nanotubes were combined with:
• commercially available graphene or graphene grown previously or directly in
the process;
• rGO obtained after reduction of GO by high temperature, or by hydrazine hydrate
(H6 N2 O), ammonium solution (NH4 OH), vitamin C, acids, etc.
Moreover, rGO-CNT-based electrodes were fabricated both on substrates, such as
glassy carbon electrode (GCE), Ni foams, metal foils, stainless steel; carbon-based supports
as carbon cloth/carbon paper, or graphite substrates; polyvinylidene fluoride (PVDF)-
treated paper; Si; plastic; indium tin oxide, etc. and without support as freestanding, i.e., by
vacuum filtration, frying or as fibres by injections, etc. Besides such important parameters
as the annealing temperature or chemicals used to control the oxygen functional groups
content, a ratio between G/rGO and CNT as well as their typology and electrolyte type
are also very important. Their influence will be overviewed in the subchapters below. In
addition, most of the studies involved not only G/rGO and CNT but also some conductive
additives and polymer binders in the similar to commercial battery fabrication process
(typically, in the 80:10:10 weight ratio between active material (AM), conductive carbon-
family-material and polymer).

3.1. Temperature Effect on G/rGO-CNT-Based Composite Electrodes

The freeze-dried GO aerogels reduced at 180 and at 700 ◦ C were used together with
SWCNT for the fabrication and comparative study of rGO-CNT composite electrodes with
PVDF binder on carbon cloth by Okhay et al. [20]. X-ray photoelectron spectroscopy (XPS)
data of the rGO aerogels shown in Figure 5a present a strongly reduced oxygen functional
group after heat treatment at 700 ◦ C that is supported by the almost ideal rectangular CV
curves for the high temperature-reduced aerogel (see Figure 5b) by contrast with that of
rGO aerogel annealed at 180 ◦ C that still show a bulge of surface redox on EDLC (see
Figure 5c). Moreover, the specific capacitance of 129 F/g reported for the electrodes made
of 180 ◦ C rGO is much higher than that of 41 F/g at 0.1 A/g for 700 ◦ C rGO [20].
A systematic study of temperature influence on functionalized graphene nanosheets/
carbon nanotubes networks (G/CNT) synthesized through chemical oxidation of CNT
followed by thermal reduction according to schematic illustration in Figure 6a was reported
by Ding et al. [21]. The external walls of CNT in the G/CNT structure annealed at 200 ◦ C
were observed as unzipped and transformed into functionalized graphene nanosheets,
while the inner walls were not unzipped and kept the tubular structure. Freeze-dried
G/CNT mixture was heat treated at 200, 300, 600 and 800 ◦ C for 2 h under N2 flow and
mixed with carbon black (CB) and poly (tetrafluoroethylene) (PTFE) before the covering on
Ni foam.
Nanomaterials 2021, 11, 1240 7 of 30
Nanomaterials 2021, 11, x FOR PEER REVIEW 7 of 30

Figure5.5.Carbon
Figure Carbon(C1s)
(C1s)(a,c)
(a,c)and
andoxygen
oxygen(O1s)
(O1s)(b,d)
(b,d)spectra
spectraobtained
obtainedby
byX-ray
X-rayphotoelectron
photoelectronspectroscopy
spectroscopy(XPS)
(XPS)on
ongraphene
gra-
phene oxide (GO) aerogel reduced◦at 180 °C in vacuum (a,b) and ◦at 700 °C in Ar (c,d). Cyclic voltammetry profiles at
oxide (GO) aerogel reduced at 180 C in vacuum (a,b) and at 700 C in Ar (c,d). Cyclic voltammetry profiles at different
different scan rates for the rGO-CNT-based composite electrodes on carbon cloth with 180 °C (e) and at 700 °C (f) rGO
scan rates for the
(Reproduced rGO-CNT-based
with composite
permission of [20]. electrodes
Copyright carbon cloth with 180 ◦ C (e) and at 700 ◦ C (f) rGO (Reproduced
on2020).
Elsevier,
with permission of [20]. Copyright Elsevier, 2020).
Based on Raman spectra, shown in Figure 6b, the ID/IG ratio was found to increase
after Based
chemicalon oxidation
Raman spectra, shown
from 0.93 in Figure
to 1.26, 6b, the
revealing thatIDa/I G ratio
large was found
number to increase
of oxygen func- af-
ter chemical oxidation from 0.93 to 1.26, revealing that a large number
tional groups were introduced during the unzipping process. However, after thermal re- of oxygen functional
groups
ductionwere
at 200introduced
°C partiallyduring
unstablethe oxygen
unzipping process.groups
functional However,wereafter thermal
removed reduction
as can be
at 200 ◦ C partially unstable oxygen functional groups were removed as can be also seen
also seen in Figure 6b, leading to an increase of the integrated area of CV in comparison
in
withFigure 6b, leading
as-prepared to an
material increase
(see of the
Figure 6c). Theintegrated areathermal
fact that after of CV reduction
in comparisonof G/CNTwith as-
prepared material (see
at 200 °C conjugated Figure
carbon 6c). Thewere
networks fact that afterwas
restored thermal reduction ofHowever,
also mentioned. G/CNT at 200 ◦ C
after
conjugated carbonatnetworks
thermal reduction T > 200 °C were
(seerestored
Figure 6b) wasthealso mentioned.
integrated area However,
of CV was after thermal
also sup-
reduction
pressed (see atFigure
T > 2006d)◦ C (see Figure
despite the fact6b)thatthe integrated
unstable oxygen area of CV was
functional groupsalsowere
suppressed
pro-
(see Figurereduced.
gressively 6d) despite the factthe
Therefore, thathighest
unstable oxygen
specific functional
capacitance ofgroups
202 F/gwere
at 0.5progressively
A/g was
reduced.
reported forTherefore,
electrodesthe highest
with G/CNT specific capacitance
after heat treatmentof at202
200 F/g at 0.5
°C (see A/g6e).
Figure was reported
Moreo-
for
ver,electrodes
CV curves with G/CNT
of G/CNTs andafter heat treatment
G/CNTs-200 at 200
electrodes show ◦ Cobvious
(see Figure
redox6e). Moreover,
humps in Fig- CV
ure 6c, meaning
curves of G/CNTs thatand
the G/CNTs-200
capacitances come both from
electrodes show EDLC
obviousand pseudocapacitance
redox humps in Figure due 6c,
to the reversible
meaning that theredox reactions come
capacitances amongboth the surface
from EDLC oxygen and functional groups [21]. due to the
pseudocapacitance
reversible redox reactions among the surface oxygen functional groups [21].
 Nanomaterials 2021, 11, x FOR PEER REVIEW 8 of 31

Nanomaterials 2021, 11, 1240 8 of 30

tance due to the reversible redox reactions among the surface oxygen functional groups
[21].

Figure 6. Schematic illustration of the synthetic route of reduced graphene oxide with carbon nanotubes (mentioned here
Figure 6. Schematic illustration of the synthetic route of (a).
reduced graphene oxide with (b).
carbon nanotubes Commented
here as [M2]: Please a
as G/CNTs) after annealing at 200 °C (G/CNTs-200) Raman spectra of studied electrodes CV curves measured (mentioned
at
Words in X-axis of subfigur
G/CNTs) after annealing at 200 ◦ C (G/CNTs-200) (a). Raman spectra of studied electrodes (b). CV curves measured at
50 mV/s on rGO, CNT, G/CNT and G/CNTs-200 (c), as well as G/CNT annealed at 300, 600 and 800 °C (G/CNTs-300,
G/CNTs-600, and G/CNTs-800, respectively) (d). Specific capacitance of rGO, CNTs, G/CNTs, G/CNTs-200, G/CNTs-300,
50 mV/s on rGO,G/CNTs-600
CNT, G/CNT and G/CNTs-200
and G/CNTs-800 as a function of (c), asdensity
current well as G/CNT annealed
(e) (Reproduced at 300,
with permission 600
of [21]. and 800
Copyright Else-◦ C (G/CNTs-300,
Commented [A3R2]: It is co
vier, 2018).
G/CNTs-600, and G/CNTs-800, respectively) (d). Specific capacitance of rGO, CNTs, G/CNTs, G/CNTs-200, G/CNTs-
300, G/CNTs-600 and G/CNTs-800 as a function of current density (e) (Reproduced with permission of [21]. Copyright
Elsevier, 2018).

3.2. Effect of CNT Length

The size of CNT can change the value of specific capacitance at least 3 times as was
reported by Zeng et al. [22]. To increase the utilization of closed pore volumes of CNT
and prevent the stacking of rGO, multiwall CNT (MWCNT) are tailored into super short
CNT (SSCNT) with aspect ratio of less than 5 by an ultrasonic oxidation-cut method (see
Figure 7a). After mixing SSCNT with rGO and PVDF, it was coated onto titanium plate.
The morphologies of the composite with MWCNTs and SSCNT were found very different
by scanning electron microscopy (SEM) as shown in Figure 7b,c. The length reduction from
5–15 µm to 10–300 nm led to rich structural features, such as nanoscale length, open ends,
abundant carbon atoms on the edge, quasi-0D characteristic and so on [22]. Moreover, the
introduction of SSCNT increased the specific surface area of rGO to 370 m2 /g that was
higher than 171 m2 /g after addition of MWCNT (see Figure 7d). Thus, shorter CNT can be
more uniformly distributed on the surface of rGO and form 3D multilayer architecture that
leads to an increase of specific capacitance from 88 F/g to 251 F/g at 50 mV/s [22].

Nanomaterials 2021, 11, 1240
phologies of the composite with MWCNTs and SSCNT were found very different by scan-
ning electron microscopy (SEM) as shown in Figure 7b,c. The length reduction from 5–15
μm to 10–300 nm led to rich structural features, such as nanoscale length, open ends,
abundant carbon atoms on the edge, quasi-0D characteristic and so on [22]. Moreover, the
introduction of SSCNT increased the specific surface area of rGO to 370 m2/g that was
higher than 171 m2/g after addition of MWCNT (see Figure 7d). Thus, shorter CNT can be 9 of 30
more uniformly distributed on the surface of rGO and form 3D multilayer architecture
that leads to an increase of specific capacitance from 88 F/g to 251 F/g at 50 mV/s [22].

Figure 7. Schematic illustration of the formation steps of the multilayer rGO/super short CNT (SSCNT) architecture (a).
Figure 7. Schematic illustration of the formation steps of the multilayer rGO/super short CNT (SSCNT) architecture (a).
Scanning electron microscope (SEM) images of raw multiwall CNT (MWCNT) (b) and SSCNT (c). Nitrogen sorption iso-
Scanning
therms electron
obtained microscope (SEM) images
at 77 K (d) (Reproduced withofpermission
raw multiwall
of [22].CNT (MWCNT)
Copyright (b)2013).
Elsevier, and SSCNT (c). Nitrogen sorption
isotherms obtained at 77 K (d) (Reproduced with permission of [22]. Copyright Elsevier, 2013).
3.3. CNT Concentration in G/rGO-CNT-Based Composite Electrodes
3.3. CNT Concentration in G/rGO-CNT-Based Composite Electrodes
Based on results reported by Lu et al. 16 wt.% CNT was the optimal concentration in
Based0on
the range ÷ 50results
wt.% reported
for rGO-CNTby Lu et al. 16 wt.%
freestanding CNT was
electrodes the optimal
to obtain concentration
the highest capaci-
intance
the ofrange 0 ÷ 50 wt.% for rGO-CNT freestanding electrodes to
265 F/g at 0.1 A/g [23]. A lower content of 10 wt.% CNT was chosen by Kumar obtain the highest
et
capacitance
al. for dried of 265 F/gpressed
composite at 0.1 A/g [23].
into Ni A lower
foam content concentration
as the optimal of 10 wt.% CNT wastochosen
of CNT obtain by
Kumar et al. for driedCV
the rectangular-like composite pressed
(see Figure 8a) andinto
theNi foamspecific
highest as the optimal concentration
capacitance (see Figureof8b)
CNT
to[24].
obtain
Among the studied CNT concentrations from 0 to 66.7 wt.%, 12.5 wt.% CNT was(see
the rectangular-like CV (see Figure 8a) and the highest specific capacitance
Figure
used to8b)get [24].
the Among
highest the studied CNT
capacitance concentrations
for rGO-CNT obtainedfrom
by 0vacuum
to 66.7 wt.%, 12.5(VF)
filtration wt.%
andCNT
was used to get the highest capacitance for rGO-CNT obtained by vacuum
pressed into Ni foam [25]. Specific capacitance of 132 F/g was reported by Lee et al. for filtration (VF)
and pressed into Ni foam [25]. Specific capacitance of 132 F/g was reported by Lee et al. for
rGO-CNT with 11 wt.% CNT prepared on glassy carbon electrode among other composites
with CNT concentration varied from 6 to 50 wt.% [26]. Thus, the optimal amount of CNT
for rGO-CNT mixed composites was found to be from 10 to 16 wt.%. However, in the
case of preparation of a CNT layered structure onto graphite paper by electrophoretic
deposition, significantly higher content of 40 wt.% CNT in suspension was used to obtain
the highest capacitance in that work of 87 F/g at 5 mV/s [27].
 rGO-CNT with 11 wt.% CNT prepared on glassy carbon electrode among other compo-
sites with CNT concentration varied from 6 to 50 wt.% [26]. Thus, the optimal amount of
CNT for rGO-CNT mixed composites was found to be from 10 to 16 wt.%. However, in
the case of preparation of a CNT layered structure onto graphite paper by electrophoretic
Nanomaterials 2021, 11, 1240 10 of 30
deposition, significantly higher content of 40 wt.% CNT in suspension was used to obtain
the highest capacitance in that work of 87 F/g at 5 mV/s [27].

Figure8.8.CV
Figure CVcurves
curvesatatthe
thescan
scan rate
rate of
of 50
50 mV/s (a)and
mV/s (a) andspecific
specificcapacitance
capacitance asas a function
a function of of current
current density
density determined
determined
fromGCD
from GCDcurves
curves(b)
(b)for
forrGO
rGO and
and rGO-CNT
rGO-CNT composites
compositeswith
with5,5,1010and
and1515
wt.%
wt.%of of
CNT electrodes
CNT (designated
electrodes as GP5C,
(designated as GP5C,
GP10C, and GP15C) in KOH electrolyte
GP10C, and GP15C) in KOH electrolyte [24]. [24].

3.4. Influence
3.4. Influence ofofElectrolyte
ElectrolyteType
Typeand
andPotential
PotentialWindow
Window
Based on the published research data, G/rGO-CNT-based
Based on the published research data, G/rGO-CNT-based electrodes were
electrodes studied
were in in
studied
both aqueous and non-aqueous electrolytes as well as in liquid and solid
both aqueous and non-aqueous electrolytes as well as in liquid and solid states. Aqueous states. Aqueous
electrolytes are
electrolytes are less
lessexpensive,
expensive,not
notflammable,
flammable,and andnot/less
not/less toxic in contrast
toxic to e.g.,
in contrast organic
to e.g., organic
electrolytes used in commercially available batteries/SC [28]. Many researchers
electrolytes used in commercially available batteries/SC [28]. Many researchers used used aque-
ous liquidliquid
aqueous and solid
andelectrolytes: acids such
solid electrolytes: as Hsuch
acids 2SO4 or H3PO4 or alkaline KOH or neutral
as H2 SO4 or H3 PO4 or alkaline KOH
Na2SO4, KCl, LiClO4, Li2SO4. Non-aqueous electrolytes were reported in the several pub-
or neutral Na2 SO4 , KCl, LiClO4 , Li2 SO4 . Non-aqueous electrolytes were reported in the
lications as organic electrolytes TEABF4 (tetraethylammonium tetrafluoroborate) and
several publications as organic electrolytes TEABF4 (tetraethylammonium tetrafluorobo-
Et4NBF4-AN (tetraethylammonium tetrafluoroborate—acetonitrile), and as ionic liquids
rate) and Et4 NBF4 -AN (tetraethylammonium tetrafluoroborate—acetonitrile), and as ionic
EMIM-BF4 (1-ethyl-3-methylimidazolium tetrafluoroborate) and EMI-TFSI (1-ethyl-3-me-
liquids EMIM-BF4 (1-ethyl-3-methylimidazolium tetrafluoroborate) and EMI-TFSI (1-ethyl-
thylimidazolium bis(trifluoromethylsulfonyl)imide). However, the results obtained, par-
3-methylimidazolium bis(trifluoromethylsulfonyl)imide). However, the results obtained,
ticularly CV form and size, indicate different behaviour of G/rGO-CNT in different elec-
particularly CV form and size, indicate different behaviour of G/rGO-CNT in different
trolytes. Cheng et al. studied supercapacitor fabricated on freestanding rGO-CNT in neu-
electrolytes. Cheng et al. studied supercapacitor fabricated on freestanding rGO-CNT in
tral electrolyte KCl, organic electrolyte TEABF4 in propylene carbonate (PC) and ionic liq-
neutral electrolyte
uid EMI-TFSI KCl,
[29]. CV organic
curve electrolyte
in the TEABF4 in
organic electrolyte propylene
TEABF carbonate (PC) and ionic
4/PC does not exhibit rectan-
liquid EMI-TFSI [29]. CV curve in the organic electrolyte
gular geometry due to the larger resistance in the organic electrolyte,4redoxTEABF /PC doesgroupsnotsuch
exhibit
rectangular geometry due to the larger resistance in the organic electrolyte,
as hydroxide group and carboxyl (see Figure 9b) by contrast with CV in KCl (see Figure redox groups
such
9a) andas in
hydroxide
EMI-TFSI.group and carboxyl
This corresponds (see Figure
to GCD 9b) have
curves that by contrast with CV shape
shown irregular in KCl in(see
Figure 9a) and in EMI-TFSI. This corresponds to GCD curves that
organic electrolyte (see Figure 9d) but triangular form in KCl (see Figure 9c) [29]. have shown irregular
shape in organic electrolyte (see Figure 9d) but triangular form in KCl (see Figure 9c) [29].
Nanomaterials 2021,
Nanomaterials 11,11,
2021, 1240
x FOR PEER REVIEW 11 of1130of 30

Figure
Figure 9. 9.Electrochemical
Electrochemical properties
properties of
of various
variouselectrodes made
electrodes made of of
CNTs, graphene,
CNTs, and and
graphene, graphene/CNT composite.
graphene/CNT (a)
composite.
Cyclic voltammetry curves in the aqueous KCl electrolyte at a scan rate of 10 mV/s. (b) Cyclic voltammetry curves in 1 M
(a) Cyclic voltammetry curves in the aqueous KCl electrolyte at a scan rate of 10 mV/s. (b) Cyclic voltammetry curves in
TEABF4/PC electrolyte at the same scan rate of 10 mV/s. (c) Galvanostatic charge/discharge curves in the aqueous electro-
1 Mlyte
TEABF 4 /PC electrolyte
at a charging
at the same scan rate of 10 mV/s. (c) Galvanostatic charge/discharge curves in the aqueous
current of 500 mA/g. (d) Galvanostatic charge/discharge curves in the organic electrolyte at the same
electrolyte
chargingat a charging
current of 500current of 500 mA/g.
mA/g (Reproduced (d)permission
with Galvanostatic charge/discharge
of [29]. Copyright Royalcurves
Societyinofthe organic electrolyte
Chemistry, 2011). at the
same charging current of 500 mA/g (Reproduced with permission of [29]. Copyright Royal Society of Chemistry, 2011).
A near rectangular shape of CV and slight asymmetry in GCD curves was observed
A nearetrectangular
by Kumar shape
al. for filtered outofrGO-CNT
CV and slight
pressed asymmetry in GCD
into Ni foam and curves
studiedwas observed
in such elec- by
Kumar
trolytesetas
al.KOH,
for filtered
NaOH out rGO-CNT
and LiOH (seepressed
Figure into
10)Ni foam
[24]. anddeviation
Small studied infromsuchthe
electrolytes
ideal
asrectangular
KOH, NaOH CV andand triangular
LiOH (seeGCDFigure 10) [24].
curves was Small deviation
explained by the from the ideal
occurrence rectangular
of some far-
CV andreaction
adaic triangular GCD
at the curves
surface was explained
ascribed by the occurrence
to the oxygen-containing of some
functional faradaic
groups reaction
attached
attothe
rGOsurface
sheetsascribed to the oxygen-containing
and functionalized CNT. As shownfunctional groups
in Figure 10a, attachedintegral
the highest to rGOareasheets
and
andfunctionalized
more rectangular CNT. As shown
CV curves were in Figureduring
observed 10a, the
thehighest integral area
test of electrodes in KOH andthat
more
rectangular CV curves
can be associated with awere observed
smaller hydratedduring the test
ionic radius andofhigher
electrodes in KOH thatofcan
ionic conductivity K+ be
associated with a smaller
ion in comparison hydrated
to that of Na+ andionic radius
Li+ ions [24].and
On higher
the otherionic
hand,conductivity of K+ ion
the ionic mobility
inenhanced
comparison to thathydrated
by a lower +
of Na ionic
and Li + ions
radius of [24].
K ionOn
+ theaccess
gains othertohand, the ionic
the electrode mobility
surface,
resulting in
enhanced byan improved
a lower electrochemical
hydrated ionic radius of K+ ion gains
performance of rGO-CNT
access toelectrode [24]. surface,
the electrode
resulting in an improved electrochemical performance of rGO-CNT electrode [24].
Nanomaterials 2021, 11, 1240 12 of 30
Nanomaterials 2021, 11, x FOR PEER REVIEW 12 of 30
Nanomaterials 2021, 11, x FOR PEER REVIEW 12 of 30

Figure
Figure 10.10.
CV CV curvesatat5050mV/s
curves mV/s (a)
(a) and
and GCD
GCD curves
curves at
at 33 A/g
A/g(b)
(b)ofofrGO-CNT
rGO-CNTfilm with
film 10 10
with wt.% of of
wt.% CNT (GP10C)
CNT in in
(GP10C)
Figure 10.
aqueous CV LiOH,
KOH, curvesand
at 50 mV/selectrolytes
NaOH (a) and GCD curves at 3 A/g (b) of rGO-CNT film with 10 wt.% of CNT (GP10C) in
[24].
aqueous KOH, LiOH, and NaOH electrolytes [24].
aqueous KOH, LiOH, and NaOH electrolytes [24].
Cuiet
Cui etal.
al.reported
reported thatthat the
the CV
CVshape
shape of ofrGO-CNT
rGO-CNT composite
composite studied
studied inin aa positive
positive po-
Cui window
potential et al. reported
is that the
different from CVthatshape
testedof inrGO-CNT
a negative composite
potential studied
window in(see
a positive
Figure 11).
tential window is different from that tested in a negative potential window (see Figure
potential window is different fromon that tested in airregular
negativeshape
potential window (see Figure
CV curves of rGO-CNT coated on Ti foil with irregular shape observed from 0 toto
11). CV curves of rGO-CNT coated Ti foil with observed from 0 +0.8V as
+0.8
V11).
as CV
wellcurves
as fromof rGO-CNT
−0.4 V to coated
+0.4 V on Tiinfoil
both Na with irregular shape observed from 0 to +0.8
2SO4 (see Figure 11a) and NaCl electrolyte
well aswell
from
V asFigure
−from
as11b)
0.4 V−0.4to +0.4 V
V tothe
both
+0.4
in Na
V both of
SO
in2 EDLC
(see(see
Na42SO4and
Figure 11a) and NaCl electrolyte
Figure 11a) and NaCl electrolyte
(see
(see
Figure 11b) indicated indicated
the existenceexistence
of EDLC pseudocapacitance
and pseudocapacitance in opposite
in opposite to
to almost
(see Figure
almost 11b) indicated
rectangular the existence
CV in potential window of EDLC
from 0and pseudocapacitance
to −0.8 V [10]. Moreover,in opposite
integrated to
rectangular
almost CV
rectangular in potential
CV in window
potential from
window 0 to
from − 0.8
0 to V [10].
−0.8 V Moreover,
[10]. integrated
Moreover, areas
integrated
areas of CV in Na2SO4 were obviously higher than that in NaCl. Correspondingly, the
ofareas
CV in Na CV2 SOin 4Nawere obviously higher than that in NaCl. Correspondingly, the specific
specificofcapacitance 2SO4 were obviously higher than that in NaCl. Correspondingly, the
was also different and the highest value was obtained for the compo-
capacitance
specific was
capacitancealso different
waspotential and the highest
also different andfromthe value
highest was obtained
value for the for
composite tested
site tested in a negative window 0 to −0.8 V inwas
Na2SOobtained the compo-
4. Thus, according to
insite
a negative
tested potential
inrGO-CNT window
a negativeelectrode from
potentialhas window0 to − 0.8 V in Na SO . Thus, according to Cui et al.
Cui et al. the shownfrom great 0 to −0.82V in
potential 4 Na2SO4. Thus, according to
to be used as a negative elec-
the rGO-CNT
Cui etfor electrode
al.energy-storage
the rGO-CNT electrodehas shown has shown great potential to be used as a negative elec-for
great potential to be used as a negative electrode
trode devices [10].
energy-storage devices [10].
trode for energy-storage devices [10].

Figure 11. CV curves of rGO-CNT electrode with ratio 10:1 (designated as GC01) obtained in different potential windows
Figure
Figure
at 5011. 11.
CVCV
mV/s in curves
curves
Na2SOof4 of rGO-CNT
rGO-CNT
electrolyte (a)electrode
electrode with
with
and NaCl ratio10:1
ratio 10:1(designated
electrolyte (designated
(b) asGC01)
as
(ReproducedGC01) obtained in
with obtained in different
permission different
of potential windows
[10]. Copyright windows at
Elsevier,
at 50
50 2015). mV/s in Na SO electrolyte (a) and NaCl electrolyte (b) (Reproduced with permission of [10].
mV/s in Na2 SO4 electrolyte (a) and NaCl electrolyte (b) (Reproduced with permission of [10]. Copyright Elsevier,
2 4 Copyright Elsevier,
2015).
2015).
Nanomaterials 2021, 11, 1240 13 of 30
Nanomaterials 2021, 11, x FOR PEER REVIEW 13 of 30

4. Modified G/rGO-CNT Electrodes with Faradaic Contribution

4. Modified G/rGO-CNT Electrodes with Faradaic Contribution
4.1. Nitrogen Doping
4.1. Nitrogen Doping
One of the popular directions today is modification of G/rGO-CNT by nitrogen (N)
One of the popular directions today is modification of G/rGO-CNT by nitrogen (N)
due to its atomic size and strong valence bonds, which are similar to those characteristics
due to its atomic size and strong valence bonds, which are similar to those characteristics
of carbon atoms [30,31]. Pyrolysis of GO with a low-cost N source is a versatile method
of carbon atoms [30,31]. Pyrolysis of GO with a low-cost N source is a versatile method
for large-scale production of N-doped graphene with flexible control over the N-bonding
for large-scale production of N-doped graphene with flexible control over the N-bonding
configurations. N-doped G/rGO-CNT structures on different substrates were reported
configurations. N-doped G/rGO-CNT structures on different substrates were reported as
as high-performance
high-performance supercapacitor
supercapacitor electrode
electrode materials.
materials. Different
Different nitrogen-containing
nitrogen-containing ma-
materials
terials such as polyacrylonitrile, acetonitrile, melamine, etc. are commonly used asused
such as polyacrylonitrile, acetonitrile, melamine, etc. are commonly as the
the ni-
nitrogen precursor.
trogen precursor.
Significant enhancementofofthe
Significant enhancement thespecific
specific capacitance
capacitance of rGO-CNT
of rGO-CNT composites
composites on GCE
on GCE
after
after N-doping was reported by Lin et al., when initial value of 10 F/g obtained for rGO-rGO-
N-doping was reported by Lin et al., when initial value of 10 F/g obtained for
CNT
CNT has grown to
has grown to 168
168 F/g
F/gatat0.5
0.5A/g
A/g forfor N-doped
N-doped rGO-CNT
rGO-CNT (designated
(designated as NGC)
as NGC) afterafter
addition of urea
addition of urea and
andlow-cost
low-costlignosulfonate
lignosulfonate (LS)
(LS) (see
(see Figure
Figure 12)12) [32].
[32]. Adding
Adding onlyonly
LS toLS to
the
the mixture of GO
mixture of GOandandCNTs
CNTs(designated
(designatedasas LGC),
LGC), thethe obtained
obtained graphene
graphene sheets
sheets in LGC
in LGC
composite were very
composite were verythick
thick(see
(seeFigure
Figure12b).
12b).

Figure12.
Figure 12.Schematic
Schematicrepresentation
representation of
of preparation
preparationprocedure
procedureofofporous
porousN-doped
N-dopedrGO-CNT
rGO-CNT (PNGC)
(PNGC)composite (a). (a).
composite SEMSEM
images of lignosulfonate (LS) modified rGO-CNT (b) and of that after the addition of urea and heat treatment at 800 °C ◦
images of lignosulfonate (LS) modified rGO-CNT (b) and of that after the addition of urea and heat treatment at 800 C
(c) (Reproduced with permission of [32]. Copyright Elsevier, 2015).
(c) (Reproduced with permission of [32]. Copyright Elsevier, 2015).
The result may be attributed to the three-dimensional structure of macromolecular
The result may be attributed to the three-dimensional structure of macromolecular
LS. However, after the addition of urea into the mixture of N-doped GO and CNTs, the
LS. However, after the addition of urea into the mixture of N-doped GO and CNTs, the
graphene sheets in NGC were found to be thinner and looser. Moreover, more porous N-
graphene sheets in NGC were found to be thinner and looser. Moreover, more porous
doped rGO-CNT (PNGC) obtained after addition of both LS and urea with further heat
N-doped
treatmentrGO-CNT
at 800 °C (PNGC)
have shownobtained
CV withafter
theaddition of both LSarea
highest integrated andthat
ureacorresponds
with furthertoheat
treatment at 800 ◦ C have shown CV with the highest integrated area that corresponds to
the highest capacitance of 246 F/g at 0.5 A/g (see Table S2 in Supplementary Materials for
the highest
details). capacitance
During of 246
the annealing F/g atexcess
process, 0.5 A/gurea(see Table S2
molecules in Supplementary
decompose Materials
abundant gases,
for details). During the annealing process, excess urea molecules decompose
which open the space between graphene sheets and prevent graphene sheets from stack- abundant
gases, whichFurthermore,
ing tightly. open the space afterbetween graphene
the addition of LSsheets andinto
and urea prevent graphene
the mixture sheets
of GO andfrom
stacking tightly. Furthermore, after the addition of LS and urea into
CNTs, many pores on the surface of thin graphene sheets in PNGC are observed, as shown the mixture of GO
and
by arrows in Figure 12c. In addition, the CV curve of rGO-CNT shows a pair of redox as
CNTs, many pores on the surface of thin graphene sheets in PNGC are observed,
shown by arrows
peaks, which may bein attributed
Figure 12c. to In
theaddition, the CV curve
residual carboxyl of rGO-CNT
and hydroxyl groupsshows
of CNT a via
pair of
redox peaks,
oxidation which
process. CVmay be of
curves attributed
NGC andto the residual
PNGC carboxyl
can also be seen to and hydroxyl
exhibit groups of
nearly rectan-
CNT
gular via oxidation
shapes and haveprocess. CV curves
some peaks as well,of NGC and
ascribed PNGC
to the can also
combination ofbe seen todou-
electrical exhibit
nearly rectangular
ble-layer capacitanceshapes and have
and faradic some peaks asfrom
pseudocapacitance well,nitrogen
ascribeddoping
to theand
combination
residual of
electrical double-layer capacitance and faradic pseudocapacitance from nitrogen doping
Nanomaterials 2021, 11, 1240 14 of 30
Nanomaterials 2021, 11, x FOR PEER REVIEW 14 of 30

and residual carboxyl and hydroxyl groups [32]. Close capacitance value of 176 F/g at
carboxyl and hydroxyl groups [32]. Close capacitance value of 176 F/g at 0.5 A/g was re-
0.5 A/g was reported for composite electrodes on Ni foam made of N-doped rGO-CNT by
ported for composite electrodes on Ni foam made of N-doped rGO-CNT by addition of
addition of polydopamide (PDA), acetylene black (AB) and PVDF [33].
polydopamide (PDA), acetylene black (AB) and PVDF [33].
4.2. Addition of Conductive Polymers
4.2. Addition of Conductive Polymers
4.2.1. G/rGO-CNT with Polypyrrole
4.2.1. G/rGO-CNT with Polypyrrole
Polypyrrole (PPy) has been extensively studied by many research groups due to its
Polypyrrole (PPy) has been extensively studied by many research groups due to its
particular advantages with regard to low cost, environmental friendliness, high capacitive
particular advantages with regard to low cost, environmental friendliness, high capacitive
capability and easy processing. Typically G/rGO-CNT-PPy composite electrodes were
capability and easy processing. Typically G/rGO-CNT-PPy composite electrodes were
fabricated by the in situ polymerization method. Pseudocapacitive composites of rGO-
fabricated by the in situ polymerization method. Pseudocapacitive composites of rGO-
CNT with PPy were obtained by Wang et al. as fibre electrode [34] and by Lu et al. as
CNT with PPy were obtained by Wang et al. as fibre electrode [34] and by Lu et al. as
freestanding electrode [35] as well as a composite with PTFE onto graphite substrate [36].
freestanding electrode [35] as well as a composite with PTFE onto graphite substrate [36]. ◦
InInthe work of Wang et al., GO-CNT fibres (see Figure13a)
the work of Wang et al., GO-CNT fibres (see Figure
13a)reduced
reducedbybyvitamin
vitamin C at 90 C
C at 90 °C
have shown specific capacitance of 10.8 F/cm 3 at 0.01 V/s in LiCl electrolyte that was
have shown specific capacitance3
of 10.8 F/cm at 0.01 V/s in LiCl electrolyte that was in-
3
increased
creased up uptoto25.9
25.9 F/cm
F/cm3 afterafter covering
covering by[34]
by PPy PPy(see
[34] (seeS3
Table Table S3 in Supplementary
in Supplementary Mate-
Materials for details).
rials for details).

Figure 13. Cross-sectional SEM image of the rGO/MWCNT fibre at low (a) and high (b) magnifications (Reproduced with
Figure 13. Cross-sectional SEM image of the rGO/MWCNT fibre at low (a) and high (b) magnifications (Reproduced
permission of [34]. Copyright American Chemical Society, 2017). Schematic representation of the microstructure and en-
with permission of [34]. Copyright American Chemical Society, 2017). Schematic representation of the microstructure and
ergy storage characteristics of the rGO-CNT-polypyrrole (PPy) film (c) (Reproduced with permission of [35]. Copyright
energy storage
Elsevier, characteristics of the rGO-CNT-polypyrrole (PPy) film (c) (Reproduced with permission of [35]. Copyright
2011).
Elsevier, 2011).
At the same time Lu et al. measured composite made of rGO, poly(sodium 4-sterene
At the same
sulfonate) time Lu et al. measured
(PSS) functionalized CNT and PPy composite
prepared made of rGO, poly(sodium
as freestanding electrode [35]4-sterene
and
sulfonate) (PSS)
as electrode onfunctionalized
graphite substrateCNTmixed
and PPy withprepared
CB and as freestanding
PTFE (see Figureelectrode
13b) [36].[35]
PSSand
ascontaining
electrode aon graphite substrate
hydrophilic group (–SO mixed
3) waswith CB and PTFE
demonstrated to be(see Figure
strongly and13b) [36]. PSS
uniformly
containing
adsorbed on a hydrophilic
the surface of group
rGO-CNT(–SO3during
) was demonstrated
the modified processto be strongly
that leadsand uniformly
to high sta-
adsorbed
bility andon the surface
dispersion of theoffunctionalized
rGO-CNT during rGO-CNTthe modified
within theprocess
aqueousthat leadsSimul-
solution. to high
stability andthedispersion
taneously, of thewith
sulfonic groups functionalized
negative chargesrGO-CNTextendingwithin
in thethe aqueous
solution solution.
provide a
Simultaneously,
number of coordinating sites onto rGO-CNT surface. Such coordinating sites can be provide
the sulfonic groups with negative charges extending in the solution used
a to
number of coordinating
effectively tether and absorbsites more
onto rGO-CNT
monomer PPy surface. Such coordinating
and facilitate the following sites can be
“homo-
used to effectively
geneous” deposition tether
of PPy and absorbonmore
particles monomer
the electrode PPy and
surface. facilitate the
Corresponding following
rGO-PSS-
CNT-based composite
“homogeneous” electrodes
deposition of PPypresented
particles oncapacitive behaviour
the electrode withCorresponding
surface. specific values rGO-
of
211 F/g and 361 composite
PSS-CNT-based F/g at 0.2 A/g for freestanding
electrodes presentedandcapacitive
graphite substrate
behaviour supported electrode,
with specific values
ofrespectively
211 F/g and (see361
Table
F/gS3atin 0.2
Supplementary Materials forand
A/g for freestanding details). That substrate
graphite could be explained
supported
by an increasing
electrode, amount
respectively (seeofTable
PPy from
S3 in~40 wt.% for the freestanding
Supplementary Materials for electrode
details).toThat
morecould
than be
70 wt.% that
explained by anon increasing
graphite substrate.
amount In of the
PPycase
from of ~40
further
wt.%PPyforconcentration
the freestanding increase the
electrode
tospecific capacitance
more than 70 wt.%can thatcontinue to grow
on graphite to 453 F/g
substrate. Inatthe
5 mV/s
case as
of reported
further PPy by Aphale et al.
concentration
for rGO-CNT-PPy
increase the specificelectrode
capacitance using
canmore than 99
continue to wt.%
growPPy [37].
to 453 F/g at 5 mV/s as reported by
Aphale et al. for rGO-CNT-PPy electrode using more than 99 wt.% PPy [37].
4.2.2. G/rGO-CNT with Polyaniline
4.2.2. G/rGO-CNT with Polyaniline
Polyaniline (PANI) is a main conductive polymer with high environmental stability,
redox reversibility,
Polyaniline electroactivity
(PANI) and unusualpolymer
is a main conductive doping/de-doping chemistry. PANIstability,
with high environmental as a
redox reversibility, electroactivity and unusual doping/de-doping chemistry. PANI as a
component of rGO-CNT composite can initiate the pseudocapacitance from the faradaic
Nanomaterials 2021,
Nanomaterials 11,11,
2021, 1240x FOR PEER REVIEW 1530
15 of of 30

component of
contribution of its
rGO-CNT composite
redox nature can initiate
that, together withtheEDLC
pseudocapacitance fromto
of rGO-CNT, leads thethefaradaic
electrode
contribution of its redox nature that, together with EDLC of rGO-CNT,
capacitance enhancement. Typically, PANI can be obtained by an in situ polymerization leads to the elec-
trode capacitance enhancement. Typically, PANI can be obtained by an
process using the dissolved aniline monomer [38–43]. By this method the total surface in situ polymeri-
zation process using the dissolved aniline monomer [38–43]. By this method the total sur-
of freestanding rGO-CNT paper [43] and fibre electrode [42] was covered by PANI that
face of freestanding rGO-CNT paper [43] and fibre electrode [42] was covered by PANI
resulted in the capacitance of 138 F/g at 0.2 A/g and 193 F/cm33 at 1 A/cm 3 , respectively
that resulted in the capacitance of 138 F/g at 0.2 A/g and 193 F/cm at 1A/cm3, respectively
(see Table S4 in Supplementary Materials for details). Higher specific capacitance of 359 F/g
(see Table S4 in Supplementary Materials for details). Higher specific capacitance of 359
at 1 A/g was obtained by Huang et al. for electrodes prepared by mixing hydrazine-
F/g at 1 A/g was obtained by Huang et al. for electrodes prepared by mixing hydrazine-
reduced GO, CNT and aniline to obtain a composite with 80 wt.% PANI [38] and with
reduced GO, CNT and aniline to obtain a composite with 80 wt.% PANI [38] and with
larger intercalation compared with PANI coating just mentioned above. Then slight surface
larger intercalation compared with PANI coating just mentioned above. Then slight sur-
redox including response from PANI can be observed in CV curves but the corresponding
face redox including response from PANI can be observed in CV curves but the corre-
peaks
spondingsymmetrical
are and do notand
peaks are symmetrical shift
dowith the increasing
not shift scan ratescan
with the increasing (seerate
Figure
(see14a). That
Figure
fact
14a). That fact together with symmetric and triangular CGD curves (see Figure 14b) sup-the
together with symmetric and triangular CGD curves (see Figure 14b) supported
dominant
ported thecapacitive behaviourbehaviour
dominant capacitive in these electrodes [38]. In[38].
in these electrodes addition, according
In addition, to our
according
estimation the b
to our estimation of the b parameter for these electrodes it was found to be ~0.9 that is to
of parameter for these electrodes it was found to be ~0.9 that is close
b close
= 1 associated with capacitor
to b = 1 associated behaviour.
with capacitor behaviour.

Figure
Figure 14.14.
CV CVcurves
curvesatatdifferent
different scan rates
rates(a)
(a)and
andGCD
GCD curves at different
curves current
at different densities
current (b) of(b)
densities rGO/CNT/PANI (Re-
of rGO/CNT/PANI
produced with
(Reproduced withpermission
permissionofof[38]. Copyright
[38]. CopyrightElsevier, 2018).
Elsevier, 2018).

AtAtthe
the same
same time,
time,similar
similarmixture
mixture of hydrazine
of hydrazine reduced GO (marked
reduced as GNSasinGNS
GO (marked Fig- in
ure 15),
Figure CNT
15), CNTandandaniline, to obtain
aniline, final final
to obtain composite including
composite also CBalso
including andCBPTFE
andbut withbut
PTFE
fraction
with of PANI
fraction lowered
of PANI to ~64 to
lowered wt.%,
~64 was
wt.%,found
wastofound
presentto significantly higher capaci-
present significantly higher
tance of 1035
capacitance of F/g
1035at F/g
1 mV/s
at 1[41].
mV/s However, the appearance
[41]. However, of strong of
the appearance redox peaks
strong redoxin CV,
peaks
shifted with the increasing scan rate (see Figure 15a), and far from symmetrical
in CV, shifted with the increasing scan rate (see Figure 15a), and far from symmetrical triangular
GCD curve
triangular GCDshapes,
curveapproaching that with plateau
shapes, approaching that with (see Figure(see
plateau 15b), are expected
Figure 15b), arenot for
expected
materials with capacitive storage mechanism but rather for faradaic materials
not for materials with capacitive storage mechanism but rather for faradaic materials [41]. [41].
Moreover,also
Moreover, alsohigh
high specific
specific capacitance
capacitance of of 987
987F/gF/gatat0.5
0.5A/g
A/gwas reported
was reportedby by
Tran
Tran
et al. for rGO-CNT-PANI prepared by the hydrothermal (HT) method at 180 °C◦and mixed
et al. for rGO-CNT-PANI prepared by the hydrothermal (HT) method at 180 C and mixed
with mesoporous carbon (MC) and Nafion (with PANI content lowered to ~26 wt.% in
with mesoporous carbon (MC) and Nafion (with PANI content lowered to ~26 wt.% in
final composite) before covering onto carbon paper [39], while value of 638 F/g at 0.5 A/g
final composite) before covering onto carbon paper [39], while value of 638 F/g at 0.5 A/g
was measured by Liu et al. for freestanding electrodes fabricated by mixing and filtration
was measured by Liu et al. for freestanding electrodes fabricated by mixing and filtration
of CNT with graphene nanosheets already covered by PANI in an autoclave at 250 °C to
of CNT with graphene nanosheets already covered by PANI in an autoclave at 250 ◦ C to
form nanorods (with >50 wt.% PANI in composite) [40]. However, these rGO-CNT-PANI
form nanorods (with >50 wt.% PANI in composite) [40]. However, these rGO-CNT-PANI
composites with high specific capacitance have shown b value much lower than 1, being
composites with high specific capacitance have shown b value much lower than 1, being
thus not associated with capacitive behaviour in pseudocapacitive materials. In addition,
thus
the slope of the EIS with
not associated capacitive
Nyquist behaviour
plot presented in pseudocapacitive
by Tran materials. In addition,
et al. for rGO-CNT-PANI-MC-Nafion
the slope of the EIS Nyquist plot presented by Tran et al. for rGO-CNT-PANI-MC-Nafion
electrode on carbon paper was closer to 45° than to 90° [39] which means a strong faradaic
electrode on carbon paper was closer to 45◦ than to 90◦ [39] which means a strong faradaic
contribution in the analysed electrodes. Thus, rGO-CNT-PANI electrodes reported by Liu
et al. [40], Tran et al. [39], and Yan et al. [41] (see Table S4 in Supplementary Materials for
 Nanomaterials 2021, 11, x FOR PEER REVIEW 16 of 30

Nanomaterials 2021, 11, 1240 16 of 30

contribution in the analysed electrodes. Thus, rGO-CNT-PANI electrodes reported by Liu

et al. [40], Tran et al. [39], and Yan et al. [41] (see Table S4 in Supplementary Materials for
details)
details) have
have shown
shown a a dominant diffusion-controlledmechanism
dominant diffusion-controlled mechanism of of energy
energy storage
storage thatthat
explains the obtained high value of specific capacitance.
explains the obtained high value of specific capacitance.

Figure
Figure 15.15.
CV CV curvesofofrGO-CNT-PANI
curves rGO-CNT-PANI(GNS/CNTs/PANI)
(GNS/CNTs/PANI) composite
compositeatatscan
scanrates
ratesofof1,1,10,
10,2020
and
and5050
mV/s (a) (a)
mV/s andand
GCDGCD
curves
curves of of pure
pure PANI
PANI andand itsits compositeswith
composites withCNT
CNT(CNTs/PANI),
(CNTs/PANI), rGO
rGO (GNS/PANI)
(GNS/PANI)and andboth bothofofthem
them(GNS/CNTs/PANI)
(GNS/CNTs/PANI)
at 2atmA/cm
2 mA/cm (b) (Reproduced with permission of [41]. Copyright Elsevier, 2009).
2 2(b) (Reproduced with permission of [41]. Copyright Elsevier, 2009).

4.3.Influence
4.3. Influence of of Metal
Metal Catalysts,
Catalysts,Metal
MetalOxides
OxidesandandHydroxides
Hydroxides
4.3.1. G/CNT Grown with
4.3.1. G/CNT Grown with Me-Catalysts Me-Catalysts
Co,Mo,
Co, Mo, Al/Fe
Al/Fe22OO3,3 Au
, Auand
andother
othermetals
metals were
were reported
reportedas as
catalysts
catalystsused for for
used G orGCNT
or CNT
growth for G/CNT electrodes [44–46]. Seo et al. formed vertical
growth for G/CNT electrodes [44–46]. Seo et al. formed vertical graphene nanosheets graphene nanosheets
(VGNS)by
(VGNS) bythe
theplasma
plasma transformation
transformation of of aa commercially
commerciallyavailable
availablenatural
naturalprecursor
precursorbut-butter
ter as illustrated in Figure 16a [44]. The plasma was essential
as illustrated in Figure 16a [44]. The plasma was essential in the process to in the process to break
breakdown
down
thecarbon-containing
the carbon-containing molecules
moleculesin inbutter
butterandandreconstruct
reconstruct them
them into ordered
into ordered andandvertical
vertical
graphitic structures (see Figure 16c). The growth of CNT was then
graphitic structures (see Figure 16c). The growth of CNT was then performed in a thermal performed in a thermal
chemical vapour
chemical vapour deposition
deposition (CVD) (CVD)process
processafter
afterthe thedeposition
deposition ofof
a Co/Mo
a Co/Mo catalyst on on
catalyst
VGNS. The as-grown VGNS/CNTs hybrid structure on a flexible graphite substrate is pre-
VGNS. The as-grown VGNS/CNTs hybrid structure on a flexible graphite substrate is
sented in Figure 16b. The SEM and transmission electron microscopy (TEM) images of
presented in Figure 16b. The SEM and transmission electron microscopy (TEM) images
pure VGNS and the VGNS-CNTs obtained after the direct growth process are shown in
of pure VGNS and the VGNS-CNTs obtained after the direct growth process are shown
Figure 16d,f. An inherently open, 3D network with dense and uniform graphene
in Figure 16d,f. An inherently open, 3D network with dense and uniform graphene
nanosheets was clearly observed to cover the entire surface of the graphite paper [44].
nanosheets was clearly observed to cover the entire surface of the graphite paper [44].
Measured VGNS-CNTs electrodes have shown a specific capacitance of 278 F/g at 10 mV/s
Measured VGNS-CNTs electrodes have shown a specific capacitance of 278 F/g at 10 mV/s
(see Table S5 in Supplementary Materials for details) and CV curves with typical shape
(see Table S5 in Supplementary Materials for details) and CV curves with typical shape
for EDLC material without redox peaks from Co and Mo used as catalysts. Moreover,
for EDLC
neither Comaterial
nor Mowithout redox by
was detected peaks
XPSfrom Co and
analysis Mostructure.
of this used as catalysts.
GCD curves Moreover,
and
neither Co nor Mo was detected by XPS analysis of this structure.
Nyquist plot also indicated capacitive energy storage mechanism in the current electrode GCD curves and Nyquist
plot
withalso
the indicated
negligible capacitive
electrochemicalenergy storage mechanism
contributions of Co andinMo thenanoparticles
current electrode[44]. with the
negligible electrochemical contributions of Co and Mo nanoparticles
At the same time, Fan et al. mixed GO with Co(NO3)2 before growing vertical CNT [44].
At the
by CVD same
at 750 °Ctime, Fan et
with Fe/Al al. mixed GO with Co(NO )2 before
2O3 as catalyst (see Figure 17a)3[45].
growing
In this case thevertical
sandwich CNT
by CVD at 750 ◦ C with Fe/Al O as catalyst (see Figure 17a) [45]. In this case the sandwich
structure was reported with vertical 2 3 CNT grown between graphene sheets as can be seen
structure
in Figurewas reported
17b,c. with to
In opposite vertical
work byCNTSeogrown between
et al. where graphene
Co/Mo sheets
catalyst wereasnot can be seen in
detected
Figure
by XPS or in an electrochemical study [44], Fan et al. have shown visible Co-based cata- by
17b,c. In opposite to work by Seo et al. where Co/Mo catalyst were not detected
XPS
lystsorresided
in an electrochemical
at the top of CNT study
(see[44],
FigureFan17e,f).
et al. Moreover,
have shown visible
strong Co-based
redox peaks in catalysts
CV
resided at the top
curves shown of CNT
in Figure 17d(see Figure
as well 17e,f). Moreover,
as nascent plateau in GCDstrong redoxinpeaks
shown Figurein17e
CVwere
curves
shown
visible in Figure 17d
suggesting theas wellpseudocapacitance
high as nascent plateau of in GCDhydroxide
cobalt shown inthat Figure 17e were
resulted visible
in meas-
suggesting
ured specific the high pseudocapacitance
capacitance of 385 F/g at 10 mV/s of cobalt
[45]. hydroxide that resulted in measured
specific capacitance of 385 F/g at 10 mV/s [45].
Nanomaterials 2021, 11, 1240 17 of 30
Nanomaterials 2021, 11, x FOR PEER REVIEW 17 of 30
Nanomaterials 2021, 11, x FOR PEER REVIEW 17 of 30

FigureFigure 16. Schematic for the direct growth of CNT onto vertical graphene nanosheets (VGNS) (a). Photograph of the as-
16. Schematic for the direct growth of CNT onto vertical graphene nanosheets (VGNS) (a). Photograph of the
grown16.
Figure VGNS/CNTs
Schematic on foratheflexible
directgraphite
growth substrate (b). vertical
of CNT onto SEM micrograph
grapheneof pristine VGNS
nanosheets (VGNS)prior
(a).toPhotograph
CNT growth of(c).
theSEM
as-
as-grown
grown
VGNS/CNTs
micrograph of the final
VGNS/CNTs
on aflexible
flexible
on ahybrid
graphite nanoarchitecture
VGNS/CNTs substrate
graphite substrate
(b). SEM micrograph
in which
(b). SEM micrograph the of VGNS
graphene
of pristine
pristine VGNS
nanosheets prior
prior to were
to CNTSEM
CNT decorated
growth
growth (c).with a (c).
SEM micrograph
micrograph
high densityofof
of the
theCNT final
final (d).hybrid VGNS/CNTs
Cross-sectional
hybrid VGNS/CNTs andnanoarchitecture
(e)nanoarchitecture
high-resolution inSEM
(f)
in which which the graphene
the images
graphene nanosheets
of nanosheets
the VGNS/CNT were decorated
were (Reproduced
decorated with a with
with
a highpermission
density
high ofofof
density [44].
CNTCNT Copyright
(d). Wiley, 2014).
(d).Cross-sectional
Cross-sectional (e) and
(e) andhigh-resolution
high-resolution (f)(f)
SEMSEM images
images of the
of the VGNS/CNT
VGNS/CNT (Reproduced
(Reproduced with with
permission
permission of [44].
of [44]. Copyright
Copyright Wiley,
Wiley, 2014).
2014).

Figure 17. Illustration of the formation of hybrid materials with CNT grown in between graphene nanosheets, showing
Figurestacked
Figure 17.layers
17. Illustration of graphene
Illustration of the
of the oxide withof
formation
formation Co catalystmaterials
ofhybrid
hybrid particles with
materials adhered
withCNT onto
CNT the layer
grown
grown in surface
in between after
graphene
between deposition
graphene (white
nanosheets, points)
showing
nanosheets, showing
and CNT
stacked in between
layers graphene
of graphene oxidelayers
with after growthparticles
Co catalyst (black tubes) (a). SEM
adhered onto images
the layer (b)surface
and transmission electron
after deposition microscopy
(white points)
stacked layers
(TEM) imageof graphene oxide (c).
of Co-rGO-CNT with Co
CVafter catalyst
results particles
measured adhered
at scan rates onto
of the50,
10, 20, layer surface after deposition (white
of com-points)
and CNT in between graphene layers growth (black tubes) (a). SEM images (b)andand 100 mV/s
transmission(d). electron
GCD curves
microscopy
and CNT
(TEM) inimage
posite between
electrode graphene
at
of Co-rGO-CNT layers
various current after
(c). CV growth
densities
results (5-50(black
measured attubes)
mA/cm (a). of
2) in KOH
scan rates SEM images
solution
10, 20, (e)and
50, (b)100
andmV/s
transmission
(Reproduced with
(d). GCD electron
permission
curves microscopy
ofofcom-
[45].
(TEM) Copyright
image
posite ofWiley,
electrode at2010).
Co-rGO-CNT (c). CV
various current results(5-50
densities measured
mA/cm2at scan
) in KOH rates of 10,(e)20,
solution 50, and 100
(Reproduced mV/s
with (d). GCD
permission curves of
of [45].
Copyright
composite Wiley,at
electrode 2010).
various current densities (5–50 mA/cm2 ) in KOH solution (e) (Reproduced with permission of [45].
Copyright Wiley, 2010).
Nanomaterials 2021, 11, 1240 18 of 30

Nanomaterials 2021, 11, x FOR PEER REVIEW 18 of 30

Very interesting results were obtained by Li et al. for electrodes made on the core-
shell structure of G grown on CNT preliminarily covered by Au nanoparticles as catalyst
(CNT@Au)Very [46].
interesting
Figureresults were aobtained
18 shows schematic by Li et al. forand
diagram electrodes
obtainedmade on the core-
structures at various
shell structure of G grown on CNT preliminarily covered by Au nanoparticles as catalyst
stages during the formation of CNT@Au composite as a function of the deposition time.
(CNT@Au) [46]. Figure 18 shows a schematic diagram and obtained structures at various
Elemental Au originating from the catalyst nanoparticles was also detected by XPS [46].
stages during the formation of CNT@Au composite as a function of the deposition time.
CNT@G powder
Elemental with graphene
Au originating growth
from the time
catalyst of 5 min being
nanoparticles was pressed into Ni
also detected by foam exhibited
XPS [46].
theCNT@G
largestpowder
CV with withredox peaksgrowth
graphene and, correspondingly,
time of 5 min being thepressed
highest specific
into Ni foam capacitance
exhib- in
comparison to other
ited the largest CV CNT@G.
with redoxHowever,
peaks and, thecorrespondingly,
reported value the of the capacitance
highest was strongly
specific capaci-
dependent on at leastto
tance in comparison two factors
other suchHowever,
CNT@G. as the mass load andvalue
the reported the width of the usedwas
of the capacitance potential
window.
stronglyFigure 19a on
dependent illustrates that
at least two the integral
factors area
such as the became
mass load andsignificantly
the width oflarger,
the usedbut the
potential
redox peaks window. Figurewith
associated 19a illustrates that the
Au catalyst integral
became area became significantly
inconspicuous larger, mass
with increasing
but theBased
loading. redox onpeaks
CV associated
measuredwith fromAu−1catalyst
V to +1became inconspicuous
V presented in Figurewith
19aincreasing
the gravimetric
(Cmmass loading.
) and areal (CBased on CV measured from −1 V to +1 V presented in Figure 19a the grav-
a ) capacitance values at different scanning rates for CNT@G electrodes
imetric (C m) and areal (Ca) capacitance values at different scanning rates for CNT@G elec-
with different mass loadings were deduced (see Figure 19b). The highest gravimetric (or
trodes with different mass loadings were deduced (see Figure 19b). The highest gravimet-
specific) capacitance of 218 F/g was obtained for the electrode with the lowest G@CNT
ric (or specific) capacitance of 218 F/g was obtained for the electrode with the lowest
mass loading of 0.5 mg/cm2 at 10 mV/s, but the highest areal capacitance of 281 mF/cm2
G@CNT mass loading of 0.5 mg/cm2 at 10 mV/s, but the highest areal2 capacitance of 281
was obtained
mF/cm 2 was for the highest
obtained studiedstudied
for the highest mass loading of 5–6
mass loading ofmg/cm
5–6 mg/cm also atat1010mV/s.
2 also mV/s.

FigureFigure 18. Schematic

18. Schematic diagram
diagram of formationofofCNT@Au
of formation CNT@Au composite
compositewith
with Au catalyst
Au (a) (a)
catalyst andand
its SEM images
its SEM after growth
images after growth
Nanomaterials
periods of2021,
about11,3x min
FOR (b),
PEER5 REVIEW 19 of 31
min (c), 6 min (d), 7 min (e) and 10 min (f) (Reproduced with permission of [46]. Copyright
periods of about 3 min (b), 5 min (c), 6 min (d), 7 min (e) and 10 min (f) (Reproduced with permission of [46]. Copyright
Elsevier, 2019).
Elsevier, 2019).

19. CV
FigureFigure 19.curves of CNT@G
CV curves of CNT@GFigure
with 19.
with CV curves
different
different of CNT@G
mass
mass loadingswith
loadings atdifferent
at 50mV/s
50 mV/s mass loadings
(a)(a)
andand at 50 mV/s
calculated
calculated (a) and calculated
capacitance
capacitance capac-
of theofCNT@G
the CNT@G Commented
electrode itance of the CNT@G electrode on Ni foam as a function of mass loading (b) (Reproduced with
electrode on Nionfoam
Ni foam
as aas a function
function of of mass
mass loading(b)
loading (b) (Reproduced
(Reproduced with
withpermission
permission of [46]. Copyright
of [46]. CopyrightElsevier, 2019).2019).
Elsevier, image of this
permission of [46]. Copyright Elsevier, 2019).
These two im
Regarding
Regardingthe
thepotential windoweffect,
potential window effect,
Li Li et studied
et al. al. studied it on CNT@G
it on CNT@G electrodes
electrodes
withwith
mass loadings 2 and 5 mg/cm 2
mass loadingsofof33mg/cm
mg/cm2 and 5 mg/cm2. A CV. A CVexample
curve curve example at arate
at a scanning scanning
of rate
Commented
20 mV/s for 3 mg/cm2 mass loading can be seen in Figure 20a. CV curves of both elec- high quality
trodes show a pair of redox peaks in the negative potential range, and an additional pair
brightness /
of redox peaks appears with an increasing potential window on the positive side. These
peaks are related to the trace amount of Au catalyst distributed on the graphene sheets.
In addition, both kinds of the specific capacitance increased with the potential window
Nanomaterials 2021, 11, 1240 Figure 19. CV curves of CNT@G with different mass loadings at 50 mV/s (a) and calculated capac- 19 of Commented
30
itance of the CNT@G electrode on Ni foam as a function of mass loading (b) (Reproduced with
permission of [46]. Copyright Elsevier, 2019). image of thi
These two im
Regarding the potential 2 window effect, Li et al. studied it on CNT@G electrodes
of 20with
mV/smassfor 3 mg/cm
loadings of 3 mg/cmmass loading
2 and 5 mg/cm can
2. A be
CVseen
curvein Figure
example at 20a. CV curves
a scanning rate of of both
Commented
electrodes
20 mV/s for 3 mg/cm mass loading can be seen in Figure 20a. CV curves of bothan
show a pair 2 of redox peaks in the negative potential range, and additional
elec- high quality
pair trodes
of redox
showpeaks appears
a pair of withinan
redox peaks theincreasing potential
negative potential window
range, on the positive
and an additional pair side.
brightness /
These peakspeaks
of redox are related
appears towiththe
antrace amount
increasing of Au
potential catalyst
window distributed
on the on These
positive side. the graphene
peaks
sheets. Inare related toboth
addition, the trace
kindsamount of Au
of the catalyst
specific distributed on
capacitance the graphene
increased withsheets.
the potential
In addition, both kinds of the specific capacitance increased with the
window width up to the maximum of 1.8 V corresponding to the range of −0.9 to 0.9 V potential window
width up to the maximum of 1.8 V corresponding to the range of −0.9 to 0.9 V (see Fig-
(see Figure 20b).
ure 20b).

Figure 20. CV curves at different Figure 20. CVwindows

potential curves at different
measuredpotential windows measured
for CNT@G electrodes forwith
CNT@G electrodes
3 mg/cm with 3loading at
2 mass
Commented
mg/cm2 mass loading at 20 mV/s (a) and relationship between the potential window and gravi-
20 mV/s (a) and relationship between
metric (orthe potential
specific) window
capacitance andareal
(Cm) and gravimetric
capacitance(or
(Caspecific)
) for CNT@Gcapacitance (Cmmass
electrodes with ) and areal image of thi
capacitance (Ca ) for CNT@G electrodes with mass loading of 23(b)
mg/cm 2 and 5 mg/cm2 (b) (Reproduced with permission
loading of 3 mg/cm 2 and 5 mg/cm (Reproduced with permission of [46]. Copyright Elsevier, These two im
of [46]. Copyright Elsevier, 2019).2019).
Commented
At the
At the same
same time,the
time, the detailed
detailed study of of
study CVCV
curves recorded
curves in the in
recorded widest potentialpotential
the widest high quality
window covering the range of −0.9 V to +0.9 V (see Figure 21a), covered two smaller
window covering the range of −0.9 V to +0.9 V (see Figure 21a), covered two smaller brightness /
windows such as between 0 and +0.9 V (see Figure 21b) and between −0.9 V and 0 V (see
windows such as between 0 and +0.9 V (see Figure 21b) and between −0.9 V and 0 V (see
Figure 21c) with completely different forms of CV. The negative potential window electrode
works obviously as EDLC (see Figure 21c) opposite to the positive range with a visible
Faradaic response (see Figure 21b). There is also a correlation with the calculated values of
the specific capacitance for CNT@G/Ni electrode with mass loading of 3 mg/cm2 . This
value reaches only 51.3 F/g at 1 mV/s for EDLC in a potential window of −0.9 V ÷ 0 V,
achieving very high 620 F/g for battery-like behaviour in the range 0 V ÷ +0.9 V, and the
middle value of 373 F/g for EDLC with a Faradaic impact in the widest potential window
−0.9 V ÷ +0.9 V (see Figure 21d). In addition, the estimated b parameter was different
for each of all three diapasons and can be presented as 0.5 < b(0 ÷ +0.9 V) < b(−0.9 V÷ +0.9 V)
< b(−0.9 V ÷ 0) ~1. Moreover, as can be seen in Figure 21d, the faradaic impact to specific
capacitance seen at a low scan rate disappeared with the rate increase. Furthermore, at
a high scan rate >0.02 mV/s the value of the specific capacitance in all three measured
potential windows becomes the same and does not exceed the lowest capacity of 51.3 F/g
(see Figure 21d) [46].
Nanomaterials 2021, 11, 1240 20 of 30
Nanomaterials 2021, 11, x FOR PEER REVIEW 20 of 30

Figure 21. CV curves obtained at different

different scanning
scanningrates
ratesfor
forCNT@G
CNT@Gwithwithmass
massloading
loadingof
of33mg/cm
mg/cm2 measured
measured in potential
window from −0.9 V to +0.9 V (a), from 0 to +0.9 V (b), from 0 to −0.9 V (c), and corresponding specific capacitances
window from −0.9 V to +0.9 V (a), from 0 to +0.9 V (b), from 0 to −0.9 V (c), and corresponding specific capacitances (d)
(d)
(Reproduced with permission of [46]. Copyright Elsevier,
(Reproduced with permission of [46]. Copyright Elsevier, 2019).2019).

4.3.2. MnO22 Induced

Induced Pseudocapacitance
Pseudocapacitance
MnO22 isis widely
widely used
used for for energy storage because of its high theoretical theoretical pseudocapaci-
pseudocapac-
itance, widepotential
tance, wide potential range,
range, and and low
low toxicity
toxicity and
and cost
cost (natural
(natural abundance).
abundance). The fact that
MnO22-based
-based composites
composites are are widely applied with neutral aqueous electrolytes being well
correlated with thethe current
current environmental
environmentalrequirements
requirementsofof“green “greenelectrolytes”
electrolytes”ininsuperca-
super-
capacitors is also
pacitors is also important.
important. Indeed,
Indeed, all reported
all reported composite
composite electrodes
electrodes basedbased on G/rGO-
on G/rGO-CNT
CNT MnO
with with 2MnO
were2 were
tested tested
in theinNa the2 SO
Na42SO 4 electrolyte
electrolyte as seen
as seen in Table
in Table S6 inS6Supplementary
in Supplemen-
Materials.
tary Comparing
Materials. Comparing the thevalues
valuesforfor
thethecomposites
composites prepared
preparedwith withand andwithout
withoutMnO MnO22
(also shown
shownin inTable
TableS6)S6)it it
can canbebeseen thatthat
seen MnO MnO
2 as 2 a redox
as a oxide
redox can
oxide significantly
can increase
significantly in-
the specific capacitance of rGO/CNT composite
crease the specific capacitance of rGO/CNT composite [47–51]. [47–51].
highest enhancement
The highest enhancementwas wasreported
reportedbybyBiBi et et
al. al.
for for
the the
layeredlayered structure
structure of
of gra-
graphene
phene andand
CNT CNT decorated
decorated by MnO by MnO
2 on 2Cuonfoil
Cu[47].
foil [47]. The and
The long longcomplicated
and complicated prepa-
preparation
ration
of of layered
layered G/CNT G/CNT
with MnO with MnO2 included
2 structure structureCVD, included CVD, immersion,
immersion, a thermal
a thermal decomposi-
decomposition
tion process, etc.process,
However, etc. it However,
resulted itinresulted in the
the specific specific capacitance
capacitance increase from increase from
42 F/g to
42 F/g to 365 F/g at 1 A/g before and
365 F/g at 1 A/g before and after MnO2 deposition, after MnO deposition,
2 respectively [47].respectively [47].
A more
A more popular
popular and
and simple
simple method
method is is the
the use
use of of KMnO
KMnO44 to to obtain
obtain MnOMnO22 during
during thethe
processing. In this way Ramezani et al. obtained the capacitance of 367 F/g at 20 20
processing. In this way Ramezani et al. obtained the capacitance of 367 F/g at mV/s
mV/s for
for the
the composite
composite of hydrazine
of hydrazine reduced
reduced rGO,rGO, CNT, CNT,MnO MnO 2 , graphite
2, graphite powder powder
and PVDFand PVDF cov-
covering
ering graphite
graphite paper
paper andand it wasit was twice
twice higher
higher than than
150150F/gF/g mentioned
mentioned in the
in the same sameworkwork
for
for rGO-CNT without MnO [48]. Electrodes on Ni foam
rGO-CNT without MnO2 [48]. Electrodes on Ni foam with rGO, CNT, MnO2, AB, PTFE
2 with rGO, CNT, MnO 2 , AB, PTFE
were
were fabricated
fabricated byby Liu
Liu etet al.
al. [49]
[49] and
and Deng
Deng et et al.
al. [51].
[51]. However,
However, Liu Liu etet al.
al. reported
reported the the
Nanomaterials 2021, 11, 1240 21 of 30

Nanomaterials 2021, 11, x FOR PEER REVIEW 21 of 30

increase from 35 F/g at 5 mV/s for rGO-CNT-AB-PTFE to 133 F/g for rGO-CNT-MnO2 -
AB-PTFE [49], using GO aerogel reduced at 800 ◦ C. At the same time, Deng et al. used
increase from 35 F/gGO,
hydrazine-reduced at 5CNT,
mV/sMnOfor rGO-CNT-AB-PTFE
2 AB, PTFE and reportedto 133 F/g for rGO-CNT-MnO
specific capacitance of2-AB-
91 F/g
PTFE
and 126[49],
F/gusing GOA/g
at 0.25 aerogel reduced at 800
for electrodes °C. Atand
without the with
sameMnO time,2Deng
[51]. et al. used
Using ureahydra-
for GO
zine-reduced
reduction and GO, CNT, MnO2 AB, PTFE and reported
poly(1,5-diaminoanthraquinone) (PDAA) specific capacitance of 91 F/g
for functionalization and 2 -
of MnO
126 F/g at 0.25 A/g for electrodes without and with MnO [51].
CNT Lei et al. obtained 80 F/g and 193 F/g at 0.2 A/g for rGO, CNT, PDAA, CB, PTFE
2 Using urea for GO reduc-
tionrGO,
and and poly(1,5-diaminoanthraquinone)
CNT, MnO2 , PDAA, CB, PTFE composites, (PDAA) forrespectively
functionalization [50]. of MnO2-CNTofLei
Preparation rGO-
et al. obtained 80 F/g and 193 F/g at ◦
0.2 A/g for rGO, CNT,
CNT-MnO2 by the HT method at 150 C was used by Li et al. for the fabrication PDAA, CB, PTFE and rGO,
of rGO,
CNT,MnO
CNT, MnO 2, PDAA, CB, PTFE composites, respectively [50]. Preparation of rGO-CNT-
2 , AB, PTFE composite electrode on Ni foam with the final specific capacitance
MnO 2 by the HT method at 150 °C was used by Li et al. for the fabrication of rGO, CNT,
336 F/g at 0.5 A/g [52]. As can be seen, all the aforementioned composite electrodes with
MnO2used
MnO , AB, additives
PTFE composite electrode
and binders on Ni
[48–52] foam with
otherwise the final
resulting specific capacitance
in fabrication 336[47].
difficulties
2
F/g at 0.5 A/g [52]. As can be seen, all the aforementioned composite
However, Cheng et al. were able to prepare rGO-CNT-MnO2 freestanding electrode electrodes with MnO 2

used additives and binders [48–52] otherwise resulting in fabrication

by simple filtration [53]. Although for that electrode, GO was reduced by hydrazine difficulties [47].
However,
and ammonium Chengsolutions,
et al. weretheablespecific
to prepare rGO-CNT-MnO
capacitance equal to 2 freestanding electrode by
372 F/g at 10 mV/s was
simple filtration [53]. Although for that electrode, GO was reduced by hydrazine and am-
measured [53].
monium solutions, the specific capacitance equal to 372 F/g at 10 mV/s was measured [53].
It needs to be stressed here that CV and GCD curves of reported electrodes with MnO2
It needs to be stressed here that CV and GCD curves of reported electrodes with
presented shapes typical for materials with dominant EDLC energy storage behaviour as
MnO2 presented shapes typical for materials with dominant EDLC energy storage behav-
can be seen in Figure 22 for G/rGO-CNT-MnO2 -AB-PTFE reported by Deng et al. [51].
iour as can be seen in Figure 22 for G/rGO-CNT-MnO2-AB-PTFE reported by Deng et al.
There are no significant redox peaks appearing in CV curves even at a high scan rate
[51]. There are no significant redox peaks appearing in CV curves even at a high scan rate
(see Figure 22a) and no plateau in GCD curves for these electrodes (see Figure 22b) [51].
(see Figure 22a) and no plateau in GCD curves for these electrodes (see Figure 22b) [51].
Moreover, the b parameter for all aforementioned G/rGO-CNT-based composites with
Moreover, the b parameter for all aforementioned G/rGO-CNT-based composites with
MnO
MnO2 wasestimated
2 was estimated by by us to be
us to be ~0.8
~0.8that
thatalso
alsocorresponds
correspondstoto dominant
dominant capacitive
capacitive type
type of of
storage in these electrode materials.
storage in these electrode materials.

Figure
Figure 22.22.
CVCV curves
curves ofof electrodesmade
electrodes madeofofrGO
rGO(mentioned
(mentioned as GR),
GR), on
onmultiwall
multiwallCNTCNT(here
(hereasasMCNT),
MCNT),their combination
their combination
GR/MCNT
GR/MCNT (a,(a, left),
left), and and their
their composite
composite with
with MnO
MnO 2
2 (GR/MCNT/MnO2)) as
(GR/MCNT/MnO 2 as well
well as
as GR/MnO
GR/MnO 2 and
2
MCNT/MnO
and MCNT/MnO 2 at aatscan
2 a scan
rate of 5 mV/s (a). GCD curves of GR/MCNT/MnO 2 at different current densities (b) (Reproduced with permission of [51].
rate of 5 mV/s (a). GCD curves of GR/MCNT/MnO2 at different current densities (b) (Reproduced with permission of [51].
Copyright Elsevier, 2012).
Copyright Elsevier, 2012).
4.3.3.Effect
4.3.3. Effectof
ofOther
Other Metal
Metal Oxides
Oxides
InInaddition
additionto
toMnO
MnO2,, the influence of other metal oxides on G/rGO-CNT-based com-
2 the influence of other metal oxides on G/rGO-CNT-based com-
posites has been also reported [54–56]
posites has been also reported [54–56] and
andsummarized
summarizedininTable
TableS7S7
inin
thethe
Supplementary
Supplementary
Materials. Ramesh et al. mixed CNT, ammonium reduced GO, and
Materials. Ramesh et al. mixed CNT, ammonium reduced GO, and cellulose cellulose fibres simul-
fibres si-
taneously with Co 3O4 and SnO2, added AB, PTFE and covered Ni foam with it [54]. CV
multaneously with Co3 O4 and SnO2 , added AB, PTFE and covered Ni foam with it [54].
curves of such electrodes studied in KOH electrolyte presented strong redox peaks at −0.1
CV curves of such electrodes studied in KOH electrolyte presented strong redox peaks at
V ÷ −0.2 V, which, as well as GCD curve shape, cannot be attributed to EDLC (see Figure
−0.1 V ÷ −0.2 V, which, as well as GCD curve shape, cannot be attributed to EDLC (see
23a) but correlated well with the faradaic impact from Co3O4 and SnO2 [54]. The reported
Figure 23a) but correlated well with the faradaic impact from Co3 O4 and SnO2 [54]. The
specific capacitance of 215 F/g at 0.2 A/g was obtained for electrodes studied in a negative
reported specific capacitance of 215 F/g at 0.2 A/g was obtained for electrodes studied in a
potential window from 0 to −1.0 V.
negative potential window from 0 to −1.0 V.
Nanomaterials 2021, 11, 1240 22 of 30
Nanomaterials 2021, 11, x FOR PEER REVIEW 22 of 30

Figure
23.23.
CVCV curves
curves obtainedatatdifferent
obtained differentscanning
scanningrates
ratesfor
for rGO-CNT-Co
rGO-CNT-Co3O 4-SnO2-AB-PTFE composite on Ni foam (a)
Figure 3 O4 -SnO2 -AB-PTFE composite on Ni foam (a)
(Reproduced with permission of [54]. Copyright Elsevier, 2017). Electrochemical performance of LiMn2O4-CNT-graphene
(Reproduced with permission of [54]. Copyright Elsevier, 2017). Electrochemical performance of LiMn2 O4 -CNT-graphene
nanocomposite: GCD at different current densities (b), EIS Nyquist plots of the electrode after cycling (c). Contribution
nanocomposite: GCD at different current densities (b), EIS Nyquist plots of the electrode after cycling (c). Contribution ratios
ratios of capacitive and diffusion-controlled processes at various scan rates (d) (Reproduced with permission of [56]. Cop-
of capacitive
yright Elsevier,diffusion-controlled
and 2019). processes at various scan rates (d) (Reproduced with permission of [56]. Copyright
Elsevier, 2019).
A similar diffusion-dominated energy-storage mechanism can be observed in CV
A similar
curves reported diffusion-dominated energy-storage
by Trian et al. for electrodes mechanism
on Ni foam made of can be observed
rGO-CNT with Fein 2O3
CV
curves
and mixed with CB and PTFE and supported by GCD curves (see Figure 23b) [55] or by O3
reported by Trian et al. for electrodes on Ni foam made of rGO-CNT with Fe 2
and
Chenmixed
et al.with CB and PTFE
for rGO-CNT with and
LiMn supported by GCD
2O4 and mixed withcurves
AB, PTFE (see[56].
Figure 23b)
In all the[55] or by
cases,
Chen et al. for rGO-CNT with LiMn
strong redox peaks in CV from metal oxides O and mixed with AB, PTFE [56]. In
2 4 indicated the significant impact from faradaicall the cases,
strong redox
materials peaks
as can be in CVinfrom
seen Figuremetal oxides
23a for LiMn indicated the significant
2O4 [56]. Moreover, GCDimpact
curves from faradaic
(see Figure
materials as can be seen in Figure 23a for LiMn
23b) as well as the EIS Nyquist plot with the slop close O [56]. Moreover, GCD curves
2 4 to 45° (see Figure 23c) was remi- (see
Figure ◦
niscent23b)
the as well as theelectrodes.
battery-type EIS Nyquist In plot withChen
addition the slop
et al.close to 45 (see
calculated Figure 23c)
the parameter b =was
reminiscent
0.689 that isthe battery-type
closer to b = 0.5 electrodes.
characteristicInfor
addition Chen et
battery-type al. calculated
energy the parameter
storage, especially at
b low
= 0.689
scanthat
rateisthat
closer bebseen
can to = 0.5incharacteristic for battery-type energy storage, especially at
Figure 23d [56].
low scan rate that can be seen in Figure 23d [56].
4.3.4. G/rGO-CNT with Ni(OH)2
4.3.4. G/rGO-CNT
Nickel hydroxide withisNi(OH) 2
an attractive material for supercapacitor applications because of
Nickel
its high hydroxide
theoretical is an capacitance,
specific attractive material for supercapacitor
well-defined applications
redox behaviour because
and low cost. The of
itsavailable data for rGO-CNT
high theoretical before and well-defined
specific capacitance, after modification
redoxbybehaviour
Ni(OH)2 are
andpresented
low cost.inThe
available data for rGO-CNT before and after modification by Ni(OH)2 are presented in
Table S8 in Supplementary Materials. The reported specific capacitance of composites with
Nanomaterials 2021, 11, x FOR PEER REVIEW 23 of 30
Nanomaterials 2021, 11, 1240 23 of 30

Table S8 in Supplementary Materials. The reported specific capacitance of composites

with
Ni(OH)Ni(OH) has significantly
2 has2 significantly higherhigher
values values in comparison
in comparison with other
with other electrodes
electrodes de-
described
scribed
above. above.
Moreover,Moreover,
all the all the reported
reported electrodes
electrodes based based on G/rGO-CNT
on G/rGO-CNT with Ni(OH)
with Ni(OH) 2 were 2

were
testedtested
in KOH in KOH electrolyte,
electrolyte, showingshowing
ratherrather close capacitance
close capacitance values values independent
independent of G/rGO of
G/rGO processing
processing temperatures.
temperatures. According
According to FantoetFanal., et al., simple
simple mixing mixing of Ni(NO)
of Ni(NO) 3 ·6H23O ·6H 2O
with
with
ureaurea
and and
withwith rGO-CNT
rGO-CNT aerogel
aerogel reduced
reduced at 800 ◦ C °C
at 800 resulted
resulted in ainstable
a stable electrode
electrode onon Ni
Ni foam
foam withwith specific
specific capacitance
capacitance of 1208
of 1208 F/gF/gat 1atA/g,
1 A/g, although
although a Ni-free
a Ni-free rGO-CNT
rGO-CNT elec-
electrode
trode prepared
prepared in theinsame
the same way showed
way showed onlyF/g
only 149 149atF/g at 1 A/g
1 A/g [57].[57]. A similar
A similar valuevalue of 1320
of 1320 F/g
at 6atA/g
F/g 6 A/gwaswasreported
reported byby
Chen
Chenet al. forfor
et al. composite
composite electrodes
electrodes mademadeof slurry including
of slurry includingAB,
PTFE andand rGO-CNT-Ni(OH) ◦ C [58]. A more complicated
AB, PTFE rGO-CNT-Ni(OH) 2 obtained
2 obtained in autoclave
in autoclave at 120
at 120 °C [58]. A more compli-
method
cated was used
method was by Duby
used et al.
Dufor et the preparation
al. for of vertically
the preparation aligned CNT
of vertically aligned(VACNT)
CNT
structure from highly ordered pyrolytic graphite at 1200 ◦ C and G growing by pyrolysis of
(VACNT) structure from highly ordered pyrolytic graphite at 1200 °C and G growing by
iron phthalocyanine ◦ C (see Figure 24a) with the following Ni(OH) coating
(FePc) at 1000(FePc)
pyrolysis of iron phthalocyanine at 1000 °C (see Figure 24a) with the following 2
by electrochemical
Ni(OH) 2 coating bydeposition (see Figure
electrochemical 24b). Specific
deposition capacitance
(see Figure of 110 F/g
24b). Specific at 10 mV/s
capacitance of
for F/g
110 G onata10CNTmV/s structure
for G on and 1384 F/g
a CNT at 5 mV/s
structure for GF/g
and 1384 on CNT and covered
at 5 mV/s for G onbyCNT Ni(OH)and2
was measured
covered by Ni(OH) [59].2 was measured [59].

Figure
Figure24.24.Typical
TypicalSEMSEMimages
imagesof
of3D 3Dpillared
pillaredvertically
verticallyaligned
alignedCNTCNT(VACNT)-graphene
(VACNT)-graphenearchitectures
architectureswithout
without(a,b)
(a,b)and
and
with
with Ni(OH)2 (c,d). Electrochemical impedance spectra (EIS) Nyquist plots of VACNT-graphene electrode before(e)
Ni(OH) 2 (c,d). Electrochemical impedance spectra (EIS) Nyquist plots of VACNT-graphene electrode before (e)and
and
after
after(f)
(f)modification
modificationby byNi(OH)
Ni(OH)2.2 .CV
CVatatdifferent
differentscan
scanrates
rates(g)
(g)and
andGCD
GCDcurves
curvesatatvarious
variousdischarge
dischargecurrent
currentdensities
densities(h)
(h)
(Reproduced with permission of [59]. Copyright American Chemical Society, 2011).
(Reproduced with permission of [59]. Copyright American Chemical Society, 2011).

However,
However,allallthese
theseelectrodes
electrodeswith
withNi(OH)
Ni(OH)22presented
presentedCV CVand
andCGD
CGDcurves
curvesvery
veryfar
far
from those for typical pseudocapacitive and particularly for EDLC materials. As
from those for typical pseudocapacitive and particularly for EDLC materials. As presented pre-
sented in Figure
in Figure 24c,
24c, CV CV curves
curves with strong
with strong redoxredox
peakspeaks increasing
increasing and shifting
and shifting withwith the
the scan
scan
rate,rate, typical
typical for for
all all G/rGO-CNTelectrodes
G/rGO-CNT electrodeswith
withNi(OH)
Ni(OH)22,, were
were reported
reported by
by Du
Duetetal.
al.
Moreover, such electrodes have shown clearly visible plateaus typical for battery electrodes
Nanomaterials 2021, 11, 1240 24 of 30

(see Figure 24d). A significant impact of faradaic contribution to G/rGO-CNT electrodes

with Ni(OH)2 is easy to detect in EIS Nyquist plots of the G-CNT electrode before (see
Figure 24e) and after (see Figure 24f) Ni(OH)2 deposition [59]. The slope changes from
almost 90◦ for G-CNT with pure EDLC behaviour to almost 45◦ was associated with the
battery. In addition, b parameters estimated by us are close to 0.6 that means the diffusion
controlled mechanism (typical for battery-type electrodes) as the dominant one in these
composites. Based on that and according to requirements from many research papers (i.e.,
references [14,15], etc.), other units and calculations associated with batteries (i.e., mAh
instead F) need to be used for the characterization of such electrode materials. Moreover,
comparison of these high specific capacitance values for such hybrid materials as well as
other their parameters with that of really capacitive materials are incorrect and speculative.

5. Specific Energy and Power of Supercapacitors with Electrodes Based on

G/rGO-CNT and Their Cycling Stability
Energy density (in Wh/cm3 ) and power density (in W/cm3 ) are known to be among
the main characteristic parameters of SC for their commercial application. Therefore, the
goal of research is to achieve high energy density at high power density, although in the
case of the electrodes based on G/rGO-CNT these values are rarely presented, being always
substituted by specific energy and power. SC specific energy (E in Wh/kg) and specific
power (P in W/kg) can be calculated by using the following expressions:

1 1
E= Ctotal ∆V 2 or E = C ∆V 2 , (2)
2 × 3.6 8 × 3.6 single el.
E
P= , (3)
∆t
where Ctotal and Csingle el. are the measured capacitance of full SC and that of single elec-
trode, respectively, ∆V is the operating voltage window, ∆t is the discharge time in hours.
Thus, although the values of capacitance are very important for the SC performance, the
electrolyte voltage window plays also a major role for the enhancement of specific energy
as well as specific power.
However, in the case of devices with strong redox peaks in CV and plateaus in GCD
curves (e.g., references [46,55–57]), the calculation of the specific energy cannot be done
using Equations (2) and (3) valid only for capacitive materials characterized by rectangular
CV and triangular GCD. That is mainly because of the non-triangular shaped GCD that is
used for the calculation of energy. Whereas the specific power and energy calculation in
capacitive materials is based on the area under the triangular GCD during charge discharge
time, the actual energy in the battery-type materials is the area under the curved lines
with plateaus. Thus, it appears that the charging energy is larger than the discharging one,
reflecting the electrode reaction being not fully reversible. Hence only a portion of the
energy used during the charging period was released during the discharging period. In
this case, the energy efficiency considered as the ratio of discharging energy to charging
energy is far smaller than 1, in contrast to that for capacitive energy storage.
Thus, the Ragone plot shown in Figure 25 presents only the available data of several
symmetric supercapacitors made of the capacitive materials (EDLC and pseudocapacitors)
and one asymmetric device that used rGO-CNT-AB-PTFE (EDLC electrode) and rGO-CNT-
MnO2 -AB-PTFE (pseudocapacitive electrode). As a result, the highest value was calculated
for the asymmetric supercapacitor supporting the importance of the enlarged voltage
window according to Equations (2) and (3). However, if we consider only symmetric SC
with capacitive electrodes (open circle in Figure 25), a significant performance is evident
to be achieved by Ding et al. for EDLC without addition of faradaic materials [21]. It is a
surprise that a network of functionalized graphene nanosheets and CNT (fG/CNT) was
synthesized by chemical oxidation of KMnO4 simultaneously with CNT showing after
low temperature treatment specific energy of 11.7 Wh/kg. This value of energy density is
higher than that for devices using faradaic materials, i.e., MnO2 [51,53], Fe3 O4 catalyst [60],
Nanomaterials 2021, 11, x FOR PEER REVIEW 25 of 30

Nanomaterials 2021, 11, 1240 25 of 30

showing after low temperature treatment specific energy of 11.7 Wh/kg. This value of en-
ergy density is higher than that for devices using faradaic materials, i.e., MnO2 [51,53],
Fe3O4 catalyst [60], PANI [38], PDA [33], and it is significantly higher than that for two
other
PANISCs [38],with
PDAEDLC electrodes
[33], and [51,61]. Moreover,
it is significantly specific
higher than capacitance
that for two otherofSCs202with
F/g EDLC
at 0.5
A/g that was not too high was reported for these single fG/CNT electrodes
electrodes [51,61]. Moreover, specific capacitance of 202 F/g at 0.5 A/g that was not too in three-elec-
trode
high configuration
was reported and the widest
for these singlevoltage
fG/CNT window was not
electrodes used for the electrochemical
in three-electrode configuration
test.
andThus, if Ding
the widest et al. did
voltage not find
window wasin not
the analysed composite
used for the the tracestest.
electrochemical of MnO
Thus,2 [21], for
if Ding
the preparation
et al. did not find ofinwhich
the analysed is usually the
KMnO4 composite used [49,50,52]
traces of MnO and
2
with
[21], which
for the such an
preparation im-of
provement could be associated, this is the best result for the G/rGO-CNT-based
which KMnO4 is usually used [49,50,52] and with which such an improvement could be electrodes
reported to date.
associated, this is the best result for the G/rGO-CNT-based electrodes reported to date.

Figure
Figure 25.
25. Ragone plots for
Ragone plots forreported
reportedenergy
energystorage
storage devices
devices using
using G/rGO-CNT-based
G/rGO-CNT-based electrodes
electrodes preliminary
preliminary tested
tested in
in three
three electrode system.
electrode system.

Another
Another important characteristicfor
important characteristic forthethepractical
practicalapplication
application of of
thethe electrodes/full
electrodes/full SC
SC is their
is their stability
stability after
after charging/dischargingfor
charging/discharging foraalong
longtime.
time. The cycling stability
stability can
can bebe
seen from Table 2 to be rather independent of the type of electrolyte or electrode
seen from Table 2 to be rather independent of the type of electrolyte or electrode substrate substrate
for
forall
all the
the reported devices,
devices,although
althoughseveral
severalofofthemthemreported
reported some
some fluctuations
fluctuations during
during the
the measurements
measurements [21,60].
[21,60]. While
While mostmost of the
of the valuesvalues
are are close
close to 100%,
to 100%, therethere are also
are also rela-
relatively
tively low values
low values of 75% of and
75%80.5%
and 80.5% reported
reported for rGO-CNT
for rGO-CNT withwith
MnOMnO2 , AB2, and
AB and
PTFE PTFE
[51] [51]
and
and rGO-CNT
rGO-CNT withwith
PANI PANI
[38], [38], respectively.
respectively. On the On the hand,
other other hand, thestability
the cycle cycle stability meas-
measurement
urement
of energyofstorage
energydevices
storagehasdevices
to beginhas to after
only beginstable
only operation
after stable hasoperation has been
been demonstrated
and the abnormal results reported sometimes are not from the
demonstrated and the abnormal results reported sometimes are not from the stabile stabile cycling but rather
cy-
frombut
cling conditioning.
rather fromInconditioning.
this case, electrode
In this stabilization
case, electrodemust be performed
stabilization mustbefore the cycle
be performed
stability
before themeasurements
cycle stability can be properlycan
measurements made.be properly made.
Nanomaterials 2021, 11, 1240 26 of 30

Table 2. Values of the initial specific capacitance (Cin ) of G/rGO-CNT-based single electrodes measured in three-electrode
configuration before long-term test and capacitance retention after long cycling (ordered with number of cycles increase)
reported for G/rGO-CNT-based single electrodes as well as for full devices using them.

Capacitance Retention
Electrolyte, Cin of Single at Current Density or Scan
Electrode Number of Rate Ref.
Sub-Strate Processing Details Voltage Electrode at Current
Materials Cycles
Window Density or Scan Rate Single
Full Cell
Electrode
rGO, CNT, - VF, H6 N2 O, NH4 OH, Na2 SO4 275 F/g 95% -
1000 [53]
+MnO2 KMnO4 1V 50 mV/s 50 mV/s
GO freeze-drying,
rGO, CNT, Na2 SO4 129 F/g 90%
C-cloth 180 ◦ C 1000 - [61]
80%rGO + 10%CNT +
+PVDF 1.6 V 0.1 A/g 0.1 A/g
10%PVDF
rGO, CNT, H6 N2 O, 95 ◦ C H2 SO4 359 F/g 80.5% -
GCE 2000 [38]
+PANI in-situ polymerization 1V 1 A/g 50 mV/s
rGO, CNT, ◦ C, Na2 SO4 120 F/g - 75%
Ni foam H6 N2 O, 95 KMnO4 2500 [51]
+MnO2 , AB,
2V 1 A/g 1 A/g
PTFE
rGO, CNT, HT, 180 ◦ C, coating KOH 165 F/g - 98.9%
Ni foam 10,000 [33]
+PDA, AB, 80%AM + 10%PVDF +
0.8 V 1 A/g 1 A/g
PTFE 10%AB
Fe3 O4 on G by
CNT/G balls, KOH - - 107.7%
Ni foam aerosolization 10,000 [60]
+Fe3 O4 , PVDF CNT by CVD, 700 ◦ C 0.9 V 3.25 A/g
rGO, CNT, chemical oxidation, Na2 SO4 202 F/g 103% 102%
Ni foam 20,000 [21]
+CB, PTFE 200 ◦ C 1.6 V 0.5 A/g 200 mV/s 200 mV/s

6. Conclusions and Perspective of G/rGO-CNT-Based Composite Electrodes

The high-quality monolayer of graphene shows great potential for different applica-
tions such as miniaturized and precise micro/nano electronics, while chemically or/and
thermally reduced graphene oxide provides a practical route towards lower-cost produc-
tion of different rGO-based devices, particularly supercapacitors. Because GO is easily
dissolved in a variety of solvents and due to high solubility of CNT achieved in GO so-
lution, the combination of G/rGO and CNT is widely studied in SC as mixed or layered
electrode materials. An amount of around 10 wt.% of CNT is generally sufficient to obtain
the maximum value of the specific capacitance in case of the two-component rGO-CNT
electrodes. Therefore, both freestanding and substrate supported electrodes can show high
specific capacitance but only under certain/individual conditions such as the method of
fabrication, type of substrate, reduction temperature or chemicals, ratio between compo-
nents of the electrodes, etc. Thus, the addition of faradaic materials always increases the
capacitance. However, although the compatibility of G/rGO-CNT with different polymers,
metals, etc. opens up a route for a wide practical realization of functional composite
materials for SC fabricated using commercial, mainly slurry-based battery technology,
determining the mechanism of energy storage for each individual electrode immediately
before the manufacture of the final device is essential. At the same time, talking not about
electrodes only but about supercapacitors in general, high specific energy and power are
crucial parameters for commercialization as well as capacitance retention. Thus, until now
the highest energy and power densities were associated with hybrid devices. Based on the
information described above, combining G/rGO-CNT with different additives and binders,
a number of perspective composite electrodes both of capacitive and battery type can be
achieved. Therefore, particular requirements to the performance of electric storage devices
determine the need for detailed understanding of the relationship between the fabrication,
structure and final properties of composite electrodes. In turn, the understanding of the
importance of the results achieved through their comparison can be greatly facilitated if
the result presentation is unified/standardized. In addition, it is obvious that the future of
supercapacitors is in the asymmetric configuration, because the symmetric one has more
limited behaviour that does not correspond to future needs.
Nanomaterials 2021, 11, 1240 27 of 30

Supplementary Materials: The following are available online at https://www.mdpi.com/article/

10.3390/nano11051240/s1, Table S1: Published details of capacitive G/rGO-CNT-based single elec-
trodes measured in three-electrode configuration, Table S2: Published details of N-doped G/rGO-
CNT single electrodes measured in three-electrode configuration, Table S3: Published details of
single G/rGO-CNT electrodes modified by PPy measured in three-electrode configuration, Table S4:
Published details of G/rGO-CNT single electrodes modified by PANI measured in three-electrode
configuration, Table S5: Published details of G/rGO-CNT-based single electrodes prepared with
metal catalysts measured in three-electrode configuration, Table S6: Published details of G/rGO-
CNT-based single electrodes modified by MnO2 measured in three-electrode configuration, Table S7:
Published details of G/rGO-CNT-based single electrodes modified by other metal oxides measured
in three-electrode configuration, Table S8: Published details of G/rGO-CNT-based single electrodes
modified by Ni(OH)2 measured in three-electrode configuration.
Author Contributions: Both authors carried out the data analysis as well as writing, reviewing and
editing of the paper. Both authors have read and agreed to the published version of the manuscript.
Funding: This article is supported by national funds, through FCT—Fundação para a Ciência e
a Tecnologia, in the scope of the framework contract foreseen in the numbers 4, 5 and 6 of the
article 23 of the Decree-Law 57/2016, of 29 August, and UID/EMS/00481/2019, UIDB/50011/2020,
UIDP/50011/2020 and by Centro 2020, through the European Regional Development Fund (ERDF),
in the scope of the project CENTRO-01-0145-FEDER-022083.
Data Availability Statement: No new data were created or analyzed in this study. Data sharing is
not applicable to this article.
Conflicts of Interest: The authors declare no conflict of interest.

Glossary

AB acetylene black
AC active carbon
AM active material
CB carbon black
CNT carbon nanotubes
CV cyclic voltammogram
CVD chemical vapor deposition
EDLC electric double-layer capacitors
EMI-TFSI 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide
fG functionalized graphene nanosheets
G graphene
GCD galvanostatic charge and discharge
GCE glassy carbon electrode
GO graphene oxide
HOPG highly ordered pyrolytic graphite
LS lignosulfonate
MC mesoporous carbon
MWCNT multiwall carbon nanotubes
NGC Nitrogen-doped reduced graphene oxide with carbon nanotubes
PANI polyaniline
PC propylene carbonate
PDA polydopamine
PDAA poly(1,5-diaminoanthraquinone)
PNGC porous nitrogen-doped reduced graphene oxide with carbon nanotubes
PPy polypyrrole
PSS poly(sodium 4-sterene sulfonate)
PTFE polytetrafluoroethylene
PVDF polyvinylidene fluoride
rGO reduced graphene oxide
SC supercapacitors
Nanomaterials 2021, 11, 1240 28 of 30

SEM scanning electron microscopy

SSA specific surface area
SSCNT super short carbon nanotubes
SWCNT single-wall carbon nanotubes
TEABF4 tetraethylammonium tetrafluoroborate
TEM transmission electron microscopy
VA vertically aligned
VGNS vertical graphene nanosheets
VF vacuum filtration
XPS X-ray photoelectron spectroscopy

References
1. Inagaki, M.; Konno, H.; Tanaike, O. Carbon materials for electrochemical capacitors. J. Power Sources 2010, 195, 7880–7903.
[CrossRef]
2. Terzyk, A.P.; Furmaniak, S.; Harris, P.J.F.; Gauden, P.A.; Włoch, J.; Kowalczyk, P.; Rychlicki, G. How realistic is the pore size
distribution calculated from adsorption isotherms if activated carbon is composed of fullerene-like fragments? Phys. Chem. Chem.
Phys. 2007, 9, 5919–5927. [CrossRef] [PubMed]
3. Simon, P.; Burke, A. Nanostructured carbons: Double-layer capacitance and more. Electrochem. Soc. Interface 2008, 17, 38–43.
[CrossRef]
4. Sánchez-González, J.; Stoeckli, F.; Centeno, T.A. The role of the electric conductivity of carbons in the electrochemical capacitor
performance. J. Electroanal. Chem. 2011, 657, 176–180. [CrossRef]
5. Peigney, A.; Laurent, C.; Flahaut, E.; Bacsa, R.R.; Rousset, A. Specific surface area of carbon nanotubes and bundles of carbon
nanotubes. Carbon 2001, 39, 507–514. [CrossRef]
6. Marinho, B.; Ghislandi, M.; Tkalya, E.; Koning, C.E.; With, G. Electrical conductivity of compacts of graphene, multi-wall carbon
nanotubes, carbon black, and graphite powder. Powder Technol. 2012, 221, 351–358. [CrossRef]
7. Xia, J.; Chen, F.; Li, J.; Tao, N. Measurement of the quantum capacitance of graphene. Nat. Nanotechchnol. 2009, 4, 505–509.
[CrossRef] [PubMed]
8. Ebbesen, T.W.; Lezec, H.J.; Hiura, H.; Bennett, J.W.; Ghaemi, H.F.; Thio, T. Electrical conductivity of individual carbon nanotubes.
Nature 1996, 382, 54–56. [CrossRef]
9. An, K.H.; Kim, W.S.; Park, Y.S.; Moon, J.-M.; Bae, D.J.; Lim, S.C.; Lee, Y.S.; Lee, Y.H. Electrochemical properties of high-power
supercapacitors using single-walled carbon nanotube electrodes. Adv. Funct. Mater. 2001, 11, 387–392. [CrossRef]
10. Cui, X.; Lv, R.; Sagar, R.U.R.; Liu, C.; Zhang, Z. Reduced graphene oxide/carbon nanotube hybrid film as high performance
negative electrode for supercapacitor. Electrochim. Acta 2015, 169, 342–350. [CrossRef]
11. DeYoung, A.D.; Park, S.W.; Dhumal, N.R.; Shim, Y.; Jung, Y.J.; Kim, H.J. Graphene oxide supercapacitors: A computer simulation
study. J. Phys. Chem. C 2014, 118, 18472–18480. [CrossRef]
12. Chen, G.Z. Understanding supercapacitors based on nano-hybrid materials with interfacial conjugation. Prog. Nat. Sci. Mater.
2013, 23, 245–255. [CrossRef]
13. Mathis, T.S.; Kurra, N.; Wang, X.; Pinto, D.; Simon, P.; Gogotsi, Y. Energy storage data reporting in perspective guidelines for
interpreting the performance of electrochemical energy storage systems. Adv. Energy Mater. 2019, 9, 1902007. [CrossRef]
14. Gogotsi, Y.; Penner, R.M. Energy Storage in Nanomaterials—Capacitive, pseudocapacitive, or battery-like? ACS Nano 2018, 12,
2081–2083. [CrossRef] [PubMed]
15. Liu, J.; Wang, J.; Xu, C.; Jiang, H.; Li, C.; Zhang, L.; Lin, J.; Shen, Z.X. Advanced Energy Storage Devices: Basic Principles,
Analytical Methods, and Rational Materials Design. Adv. Sci. 2018, 5, 1700322. [CrossRef] [PubMed]
16. Wang, J.; Polleux, J.; Lim, J.; Dunn, B. Pseudocapacitive contributions to electrochemical energy storage in TiO2 (anatase)
nanoparticles. J. Phys. Chem. C 2007, 111, 14925–14931. [CrossRef]
17. Girard, H.-L.; Wang, H.; d’Entremont, A.; Pilon, L. Physical interpretation of cyclic voltammetry for hybrid pseudocapacitors. J.
Phys. Chem. C 2015, 119, 11349–11361. [CrossRef]
18. Conway, B.E.; Pel, W.G. Double-layer and pseudocapacitance types of electrochemical capacitors and their applications to the
development of hybrid devices. J. Solid State Electrochem. 2003, 7, 637–644. [CrossRef]
19. Zhou, Y.; Hu, X.C.; Guo, S.; Yu, C.; Zhong, S.; Liu, X. Multi-functional graphene/carbon nanotube aerogels for its applications in
supercapacitor and direct methanol fuel cell. Electrochim. Acta 2018, 264, 12–19. [CrossRef]
20. Okhay, O.; Tkach, A.; Gallo, M.J.H.; Otero-Irurueta, G.; Mikhalev, S.; Staiti, P.; Lufrano, F. Energy storage of supercapacitor
electrodes on carbon cloth enhanced by graphene oxide aerogel reducing conditions. J. Energy Storage 2020, 32, 101839. [CrossRef]
21. Ding, B.; Guo, D.; Wang, Y.; Wu, X.; Fan, Z. Functionalized graphene nanosheets decorated on carbon nanotubes networks for
high performance supercapacitors. J. Power Sources 2018, 398, 113–119. [CrossRef]
22. Zeng, F.; Kuang, Y.; Zhang, N.; Huang, Z.; Pan, Y.; Hou, Z.; Zhou, H.; Yan, C.; Schmidt, O.G. Multilayer super-short carbon
nanotube/reduced graphene oxide architecture for enhanced supercapacitor properties. J. Power Sources 2014, 247, 396–401.
[CrossRef]
Nanomaterials 2021, 11, 1240 29 of 30

23. Lu, X.; Dou, H.; Gao, B.; Yuan, C.; Yang, S.; Hao, L.; Shen, L.; Zhang, X. A flexible graphene/multiwalled carbon nanotube film as
a high performance electrode material for supercapacitors. Electrochim. Acta 2011, 56, 5115–5121. [CrossRef]
24. Kumar, A.; Kumar, N.; Sharma, Y.; Leu, J.; Tseng, T.Y. Synthesis of free-standing flexible rGO/MWCNT films for symmetric
supercapacitor application. Nanoscale Res. Lett. 2019, 14, 266. [CrossRef]
25. Huang, Z.D.; Zhang, B.; Liang, R.; Zheng, Q.B.; Oh, S.W.; Lin, X.Y.; Yousefi, N.; Kim, J.K. Effects of reduction process and carbon
nanotube content on the supercapacitive performance of flexible graphene oxide papers. Carbon 2012, 50, 4239–4251. [CrossRef]
26. Lee, B.R.; Chang, D.W. Graphene/multi-walled carbon nanotubes hybrid materials for supercapacitors. Clean Technol. 2015, 21,
62–67. [CrossRef]
27. Lu, T.; Pan, L.; Li, H.; Nie, C.; Zhu, M.; Sun, Z. Reduced graphene oxide–carbon nanotubes composite films by electrophoretic
deposition method for supercapacitors. J. Electroanal. Chem. 2011, 661, 270–273. [CrossRef]
28. Ramachandran, R.; Wang, F. Chapter 3: Electrochemical Capacitor Performance: Influence of Aqueous Electrolytes. In
Supercapacitors—Theoretical and Practical Solutions, in Supercapacitors: Theoretical and Practical Solutions; Liudvinavičius, L., Ed.;
IntechOpen: London, UK, 2017. [CrossRef]
29. Cheng, Q.; Tang, J.; Ma, J.; Zhang, H.; Shinya, N.; Qin, L.C. Graphene and carbon nanotube composite electrodes for supercapaci-
tors with ultra-high energy density. Phys. Chem. Chem. Phys. 2011, 13, 17615–17624. [CrossRef]
30. Zhang, X.; Samorì, P. Graphene/polymer nanocomposites for supercapacitors. ChemNanoMat 2017, 3, 362–372. [CrossRef]
31. Huang, Z.; Li, L.; Wang, Y.; Zhang, C.; Liu, T. Polyaniline/graphene nanocomposites towards high-performance supercapacitors:
A review. Compos. Commun. 2018, 8, 83–91. [CrossRef]
32. Lin, T.T.; Lai, W.H.; Lü, Q.F.; Yu, Y. Porous nitrogen-doped graphene/carbon nanotubes composite with an enhanced supercapaci-
tor performance. Electrochim. Acta 2015, 178, 517–524. [CrossRef]
33. Zeng, R.; Deng, H.; Xiao, Y.; Huang, J.; Yuan, K.; Chen, Y. Cross-linked graphene/carbon nanotube networks with polydopamine
“glue” for flexible supercapacitors. Compos. Commun. 2018, 10, 73–80. [CrossRef]
34. Wang, S.; Liu, N.; Su, J.; Li, L.; Long, F.; Zou, Z.; Jiang, X.; Gao, Y. Highly stretchable and self-healable supercapacitor with
reduced graphene oxide based fiber springs. ACS Nano 2017, 11, 2066–2074. [CrossRef]
35. Lu, X.; Dou, H.; Yuan, C.; Yang, S.; Hao, L.; Zhang, F.; Shen, L.; Zhang, L.; Zhang, X. Polypyrrole/carbon nanotube nanocomposite
enhanced the electrochemical capacitance of flexible graphene film for supercapacitors. J. Power Sources 2012, 197, 319–324.
[CrossRef]
36. Lu, X.; Zhang, F.; Dou, H.; Yuan, C.; Yang, S.; Hao, L.; Shen, L.; Zhang, L.; Zhang, X. Preparation and electrochemical capacitance
of hierarchical graphene/polypyrrole/carbon nanotube ternary composites. Electrochim. Acta 2012, 69, 160–166. [CrossRef]
37. Aphale, A.; Maisuria, K.; Mahapatra, M.K.; Santiago, A.; Singh, P.; Patra, P. Hybrid electrodes by in-situ integration of graphene
and carbon-nanotubes in polypyrrole for supercapacitors. Sci. Rep. 2015, 5, 14445. [CrossRef]
38. Huang, Y.; Zhou, J.; Gao, N.; Yin, Z.; Zhou, H.; Yang, X.; Kuang, Y. Synthesis of 3D reduced graphene oxide/unzipped carbon
nanotubes/polyaniline composite for high-performance supercapacitors. Electrochim. Acta 2018, 269, 649–656. [CrossRef]
39. Tran, V.C.; Nguyen, V.H.; Nguyen, T.T.; Lee, J.H.; Huynh, D.C.; Shim, J.J. Polyaniline and multi-walled carbon nanotube-
intercalated graphene aerogel and its electrochemical properties. Synth. Met. 2016, 215, 150–157. [CrossRef]
40. Liu, P.; Yan, J.; Gao, X.; Huang, Y.; Zhang, Y. Construction of layer-by-layer sandwiched graphene/polyaniline nanorods/carbon
nanotubes heterostructures for high performance supercapacitors. Electrochim. Acta 2018, 272, 77–87. [CrossRef]
41. Yan, J.; Wei, T.; Fan, Z.; Qian, W.; Zhang, M.; Shen, X.; Wei, F. Preparation of graphene nanosheet/carbon nanotube/polyaniline
composite as electrode material for supercapacitors. J. Power Sources 2010, 195, 3041–3045. [CrossRef]
42. Liu, D.; Du, P.; Wei, W.; Wang, H.; Wang, Q.; Liu, P. Skeleton/skin structured (RGO/CNTs)@PANI composite fiber electrodes
with excellent mechanical and electrochemical performance for all solid-state symmetric supercapacitors. J. Colloid Interface Sci.
2018, 513, 295–303. [CrossRef]
43. Huang, Z.D.; Liang, R.; Zhang, B.; He, Y.B.; Kim, J.K. Evolution of flexible 3D graphene oxide/carbon nanotube/polyaniline
composite papers and their supercapacitive performance. Compos. Sci. Technol. 2013, 88, 126–133. [CrossRef]
44. Seo, D.H.; Yick, S.; Han, Z.J.; Fang, J.H.; Ostrikov, K. Synergistic fusion of vertical graphene nanosheets and carbon nanotubes for
high-performance supercapacitor electrodes. ChemSusChem 2014, 7, 2317–2324. [CrossRef]
45. Fan, Z.; Yan, J.; Zhi, L.; Zhang, Q.; Wei, T.; Feng, J.; Zhang, M.; Qian, W.; Wei, F. A three-dimensional carbon nanotube/graphene
sandwich and its application as electrode in supercapacitors. Adv. Mater. 2010, 22, 3723–3728. [CrossRef]
46. Li, K.; Li, H.; Li, M.; Li, C.; Su, L.; Qian, L.; Yang, B. Carbon-nanotube@graphene core-shell nanostructures as active material in
flexible symmetrical supercapacitors. Compos. Sci. Technol. 2019, 175, 92–99. [CrossRef]
47. Bi, T.; Fang, H.; Jiang, J.; He, X.X.; Zhen, X.; Yang, H.; Wei, Z.; Jia, Z. Enhance supercapacitive performance of MnO2 /3D carbon
nanotubes graphene as a binder-free electrode. J. Alloys Compd. 2019, 787, 759–766. [CrossRef]
48. Ramezani, M.; Fathi, M.; Mahboubi, F. Facile synthesis of ternary MnO2 /graphene nanosheets/carbon nanotubes composites
with high rate capability for supercapacitor applications. Electrochim. Acta 2015, 174, 345–355. [CrossRef]
49. Liu, Y.; He, D.; Duan, J.; Wang, Y.; Li, S. Synthesis of MnO2 /graphene/carbon nanotube nanostructured ternary composite for
supercapacitor electrodes with high rate capability. Mater. Chem. Phys. 2014, 147, 141–146. [CrossRef]
50. Lei, Z.; Shi, S.; Lu, L. Incorporation of MnO2 -coated carbon nanotubes between graphene sheets as supercapacitor electrode. ACS
Appl. Mater. Interfaces 2012, 4, 1058–1064. [CrossRef] [PubMed]
Nanomaterials 2021, 11, 1240 30 of 30

51. Deng, L.; Hao, Z.; Wang, J.; Zhu, G.; Kang, L.; Liu, Z.-H.; Yang, Z.; Wang, Z. Preparation and capacitance of graphene/multiwall
carbon nanotubes/MnO2 hybrid material for high-performance asymmetrical electrochemical capacitor. Electrochim. Acta 2013,
89, 191–198. [CrossRef]
52. Li, W.; Xu, H.; Cui, M.; Zhao, J.; Liu, F.; Liu, T. Synthesis of sulfonated graphene/carbon nanotubes/manganese dioxide composite
with high electrochemical properties. Ionics 2019, 25, 999–1006. [CrossRef]
53. Cheng, Y.; Lu, S.; Zhang, H.; Varanasi, C.V.; Liu, J. Synergistic effects from graphene and carbon nanotubes enable flexible and
robust electrodes for high-performance supercapacitors. Nano Lett. 2012, 12, 4206–4211. [CrossRef]
54. Ramesh, S.; Khandelwal, S.; Rhee, K.Y.; Hui, D. Synergistic effect of reduced graphene oxide, CNT and metal oxides on cellulose
matrix for supercapacitor applications. Compos. Part B Eng. 2018, 138, 45–54. [CrossRef]
55. Tian, Y.; Hu, X.; Wang, Y.; Li, C.; Wu, X. Fe2 O3 nanoparticles decorated on graphene-carbon nanotubes conductive networks for
boosting the energy density of all-solid-state asymmetric supercapacitor. ACS Sustain. Chem. Eng. 2019, 7, 9211–9219. [CrossRef]
56. Chen, L.; Li, D.; Zheng, X.; Chen, L.; Zhang, Y.; Liang, Z.; Feng, J.; Si, P.; Lou, J.; Ci, L. Integrated nanocomposite of
LiMn2 O4 /graphene/carbon nanotubes with pseudocapacitive properties as superior cathode for aqueous hybrid capacitors. J.
Electroanal. Chem. 2019, 842, 74–81. [CrossRef]
57. Fan, W.; Shi, Y.; Gao, W.; Sun, Z.; Liu, T. Graphene-carbon nanotube aerogel with a scroll-interconnected-sheet structure as an
advanced framework for a high-performance asymmetric supercapacitor electrode. ACS Appl. Nano Mater. 2018, 1, 4435–4441.
[CrossRef]
58. Chen, X.; Chen, X.; Zhang, F.; Yang, Z.; Huang, S. One-pot hydrothermal synthesis of reduced graphene oxide/carbon nanotube/α-
Ni(OH)2 composites for high performance electrochemical supercapacitor. J. Power Sources 2013, 243, 555–561. [CrossRef]
59. Du, F.; Yu, D.; Dai, L.; Ganguli, S.; Varshney, V.; Roy, A.K. Preparation of tunable 3D pillared carbon nanotube-graphene networks
for high-performance capacitance. Chem. Mater. 2011, 23, 4810–4816. [CrossRef]
60. Mao, S.; Wen, Z.; Bo, Z.; Chang, J.; Huang, X.; Chen, J. Hierarchical nanohybrids with porous CNT-networks decorated crumpled
graphene balls for supercapacitors. ACS Appl. Mater. Interfaces 2014, 6, 9881–9889. [CrossRef]
61. Okhay, O.; Tkach, A.; Staiti, P.; Lufrano, F. Long term durability of solid-state supercapacitor based on reduced graphene oxide
aerogel and carbon nanotubes composite electrodes. Electrochim. Acta 2020, 353, 136540. [CrossRef]