Transcript I - e

12.3 Flexible transparent conductive thin films for electrode [12-8, 12-11]
To produce conductive electrodes, the authors in this work grew large-area
graphene on Cu foils by taking advantage of the low C solubility in Cu and used
polymethyl-methacrylate (PMMA) to aid the transfer of graphene to other
substrates. This kind of transfer-printing method has also been used to transfer
and align carbon nanotube arrays. However,
they found that the transfer process caused
the graphene to form cracks upon transfer,
due to the intrinsic mechanical properties
on monolayer graphene and the method
of transfer initially used.
Ideally, if the graphene films were
perfectly flat, the old process would be
efficient in transferring large-area
graphene; however, because the surface
of the metal goes through significant
surface reconstruction at high
temperatures the resulting metal surface
tends to be rough and graphene tends to
follow the surface of the underlying metal.
1
When graphene is removed from the “rough” metal surface it does not lie flat on
top of the target surface, as a result, there are always some small gaps between the
graphene and the substrate surface, that is, the graphene does not make full
contact with the SiO2/Si substrate and the unattached regions tend to break easily
and cracks are formed when the PMMA film is dissolved away (top-right inset in
Figure 2). Part of the problem is that the PMMA is a hard coating after curing and
when it is dissolved away, the graphene does not relax.
We found that the graphene transfer process was improved by introducing a
second PMMA coating step after the PMMA/graphene was placed on the SiO2/Si
substrate. After placing the PMMA/graphene stack on the target substrate, an
appropriate amount of liquid PMMA solution was dropped on the cured PMMA layer
thus partially or fully dissolving the precoated PMMA. The redissolution of the
PMMA tends to mechanically relax the underlying graphene, leading to a better
contact with the substrate.
Figure 3a shows an optical micrograph of a
50 × 50 μm2 graphene film. The uniform color
contrast of the optical micrograph indicates that
the film has excellent thickness uniformity but it
also shows the presence of dark lines that are
associated with wrinkles which are believed to
form during cool-down as a result of the
difference in coefficient of thermal expansion
between the graphene and the Cu substrate.
Fig. 3a
2
Figure 3. Micro-Raman spectroscopy images. (b-d) Raman maps of integrated intensity of the D (13001400 cm-1) and G (1560-1630 cm-1) bands, and fwhm of 2D (2620-2740 cm-1) bands, respectively,
corresponding to the area in (a).
A Raman spectrum with peaks typical for graphene, including a 2D-band with a
full width half-maximum (fwhm) of ∼28 cm-1 located at ∼2680 cm-1. In addition, the
intensity of the D-band at ∼1350 cm-1, a measure of defects in the graphene, is
below the Raman detection limit. This is further demonstrated by a low D-band
intensity map (Figure 3b) across most of the film except at wrinkled regions. It
should be noted that the D-band map is more sensitive to wrinkles than the optical
micrograph, and the G- and 2D-band maps. The G- and 2D-band maps (Figure
3
3c,d) show that the film is graphene and that it is continuous
The sheet resistance is calculated using the formula:
Rs 
 ( Rx  Ry )
ln 2
2
f
where f is a factor that is a function only of the ratio of Ry/Rx. Rx and Ry are the
resistance of the film in the x and y direction, respectively. For a
flat graphene layer on PMMA, the sheet resistance, Rs, is
∼350 Ω/□. When the film was bent, Rx is
independent of the bending radius while
Ry is nearly invariant for r = 3 mm
(approximate tensile strain of 5%) even
after 100 bending cycles. But when r was
decreased to about 1 mm (approximate tensile
strain of 15%), Ry increased by about a factor
of 2.
The increase in Ry may be attributed to
excess stretching of graphene but even for
this case, the sheet resistance is still low,
<500 Ω/ □.
4
(a)Photographs of 1 cm2
films with 1 up to 4 layers of
stacked graphene films on
cover glass slips.
(b) Transmittance of n-layer
graphene films shown in (a).
The inset is the relationship
between the transmittance,
T (%), at λ ) 550 nm as a
function of the number of
stacked graphene layers, n.
(c) Sheet resistance of n-layer graphene films as a function of the
number of stacked graphene layers, n.
(d) Comparison of transparent conductive films,
5
In the work performed by Wang [12-8], they prepared conductive, transparent,
and ultrathin graphene films from exfoliated graphite oxide, followed by thermal
reduction. The obtained graphene films with a thickness of 10 nm exhibit a high
conductivity of 550 S/cm, which is comparable to that of polycrystalline graphite
(1250 S/cm), and a transparency of more than 70% over 1000-3000 nm. The
application of graphene films as window electrodes in solid-state dye-sensitized
solar cells is demonstrated.
In their work, GO was produced by the Hummers method through acid oxidation of
flake graphite. The processes included:
(1) The primary product was suspended in water under ultrasonication for half an
hour, followed by centrifuged at 4000 rpm for 30 min.
(2) The obtained supernate was dried via evaporation of water under vacuum. Then,
the solid were dispersed again in water (1.5 mg/mL) by ultrasonicated for 2 h and
centrifuged at 10 000 rpm for 15 min to further remove aggregates.
(3) Finally, the supernate was collected and ready for use.
Such aqueous dispersion of exfoliated GO could stay stable for several weeks,
free of any obvious precipitates. The exfoliated graphene sheets with lateral
dimensions of several tens to hundreds of nanometers were observed under
6
scanning electron microscopy (SEM) (Figure 1a).
However, the obtained GO are electrically insulating due to the heavy
oxygenation of graphene sheets. Reduction of GO, either by chemical reaction
using reducing agent, such as NaBH4 or dimethylhydrazine, or by pyrolysis at
high temperatures, has been reported to render the material electrically
conductive.
However, to avoid agglomeration
of graphene sheets after reduction,
other host molecules such as
polymers must be used, which
hamper the electron-transfer
property of the graphenes. In this
work, GO sheets were first
deposited on the surface of the
substrate and then reduced into
graphenes, which afforded ultrathin
and homogeneous graphene films.
Figure 1. Morphology of GO films. (A) SEM image of exfoliated graphite oxide (GO). (B) SEM image of
GO film prepared from dip coating. (C) AFM height image (3.2 3.2 mm2) (color scale: black to bright
7
yellow, 30 nm). (D) AFM phase image (color scale: black to bright yellow, 15°) of the obtained GO film.
Typically, GO sheets were deposited on hydrophilic substrates (such as
pretreated quartz) by dip coating of a hot, aqueous GO dispersion and
subsequent temperature controlled drying of the film. The thickness of the film
was tuned by changing the temperature of GO dispersion as well as the dipping
repetition.
For example, 2-fold dip coating of the GO dispersion at 70 °C resulted in a ca.
10 nm thick, continuous and homogeneous film. Quasi-one-dimensional
creases with a length of 0.2-2.5 mm and a height of 5-20 nm were observed with
SEM (Figure 1b) and atomic force microscopy (AFM), which was formed by the
overlap of GO sheets where some of the graphene edges were scrolled
or folded(Figure 1c,d) during film fabrication.
Reduction of the GO film into a graphene film was achieved via thermal
treatment under protection of Ar and/or H2 flow. Color change from light brown
to light gray of the GO film on quartz indicated the formation of graphene.
The obtained graphene film displayed similar morphology to GO film and
creases were occasionally observed . However, the surface roughness (Ra)
has been improved significantly after thermal treatment. The average Ra of the
as prepared graphene film over a 10x10 mm2 area is ca. 0. 78 nm. It is widely
accepted that the Ra of the electrodes is crucial in optoelectronic devices. In
contrast to the rough FTO surface, which might short-circuit the cells, an
8
ultrasmooth surface is a prominent characteristic of the graphene films.
The electrical conductivity of the as-prepared graphene films is closely related
to the annealing temperature and the thickness of the film. At a given film
thickness of ca. 10 nm, concomitant increase of film conductivity was observed
with an increase in the heating temperatures from 550 to 1100 °C.
The sheet resistance (Rs) of a 10.1 ± 0.76 nm thick graphene film prepared by
1100 °C thermal treatment was 1.8 ±0.08 kΩ/sq with the calculated average
conductivity of 550 S/cm. In addition, the conductivity of graphene film increased
to 727 S/cm when the film thickness was increased to 29.9 ± 1.1 nm, for which
the Rs is 0.46 ± 0.03 kΩ/sq . The high conductivity of the graphene film,
comparable to that of polycrystalline graphite (1250 S/cm), results from the
effective recovering and subsequent annealing of continuous and overlapped
graphene sheets by removal of oxygenated groups in GO films as shown by IR
spectroscopy
The transmittance of the graphene film depends on the film thickness. At a
wavelength of 1000 nm, a ca. 10 nm thick film has a transmittance of 70.7%, which
is lower than that of FTO of 82.4% and ITO (with an thickness of about 120 nm and
sheet sheet resistance <15 Ω cm-2) of 90.0%. However, decreasing the film
thickness leads to an improvement of the transmittance to over 80.0%. Most
interestingly, in contrast to FTO and ITO, which show strong absorptions in the
region of near- (0.75-1.4 mm) and short-wavelength infrared (1.4-3 mm), the
9
graphene films remain transparent in these regions.
Films obtained from lower temperatures such as 700 and 500 °C show stronger
absorption in the range 200-1000 nm, compared to the films obtained from 1100 °C,
suggesting that the transparency of graphene films is also correlated to their
structures. A higher extent of graphitization of GO film is important to improve not
only the transmittance but also the conductivity of the finally formed graphene films.
Besides low resistance, high transparency, and a smooth surface, a high thermal
stability is necessary for application as electrodes in dye-sensitized TiO2 solar cells to
replace FTO. Furthermore, in contrast to metal oxide coatings, graphene films are
chemically stable and resistant to strong acids such as hydrochloride acid. It is also
notable that the surface wettability of graphene film is tunable. The contact angle of
graphene films was changed from 66.5-69° to 2.2-8.6° by exposure to argon plasma
treatment for 30 s. Thereby, the sheet resistance of the film changed little from 10
0.9 to 1.08 kΩ/sq.
A dye-sensitized solid solar cell based on
spiro-OMeTAD1 (as a hole transport material)
and porous TiO2 (for electron transport) was
fabricated using the graphene film as anode
and Au as cathode (Figure 3a)
In optoelectronic devices, proper contact
between electrode and p/n type material is
essential for charge collection. Figure 3b
shows the energy level diagram of a graphene/
TiO2/dye/spiro-OMeTAD/Au device.
Because the calculated work function of
graphene is 4.42 eV and the mostly reported
work function of HOPG is 4.5 eV,17 it is
reasonable to presume that the work function
of as prepared graphene film is close to that of
FTO electrode (4.4 eV). The electrons are first
injected from the excited state of the dye into the
conduction band of TiO2 and then reach the
graphene electrode via a percolation mechanism
inside the porous TiO2 structure. Meanwhile, the
photo oxidized dyes are regenerated by the
spiro-OMeTAD hole conducting molecules.
11
Because spiro-OMeTAD and graphite have been proved to form ohmic contact, a
blocking TiO2 layer is used to prohibit the possible recombination of the charge
carriers at the interface of graphene electrode/spiro-OMeTAD. Therefore, the devices
were prepared by first depositing a compact blocking TiO2 layer via spin-coating Ti-IV
tetra-isopropoxide (TTIP) ethanol solution on electrodes of graphene films or F-doped
SnO2 (FTO).
Anode :
Do + energy (hν) → D*
ground
excited
state
state
＋
－
D* → D ＋e (TiO2)
2D+ + 3I－ (electrolyte) → 2Do + I3－
eFlexible PET
e-
Electrolyte
e-
I3I
ITOTiO2 Dye
Cathode :
I3－ + 2e－(Pt) → 3I－
-
e
-
Pt
Journal of Photochemistry and
Photobiology C: Photochemistry
Reviews 4 (2003) 145–153
Rh : Impedance of substrate(TCO)
R1: Impedance of Pt/I–/I3– redox couple
R2: Impedance of TiO2/dye/ I–/I3– redox couple
R3: Diffusion of electrolyte
12
C. R. Chimie 9 (2006) 645–651
Afterward, the substrate was sintered at 400 °C for 30 min, followed by fabrication
of a 2.5-3 mm thick mesoporous film of TiO2 via doctor-blading a TiO2 paste. The
substrate was sintered at 400 °C again for 1 h and then sensitized by soaking in an
ethanol solution of ruthenium dye N3 (Ru(4,4’-dicarboxy-2,2’-bipyridine) 2-(NCS)2) for
7 h and then washed with ethanol.
The hole transporter matrix layer was prepared by spin-coating a solution of
spiro-OMeTAD (0.17 M) in chlorobenzene, containing tert-butylpyridine (0.13 M),
N(PhBr)3SbCl6 (0.3 mM), and Li(CF3SO2)2N (0.013 M). Finally, a 50 nm thick
Au electrode was evaporated on top of substrates.
Ref: B. Li et al. / Solar Energy Materials &
Solar Cells 90 (2006) 549–573
13
12.4 Graphene-based nanocomposites for energy storage [12-10]
There are three major ways that energy is stored: chemically, electrochemically,
and electrically. There is an abundant array of materials that can be used for
energy storage. However, if one is concerned about the energy/weight ratio and
cost of the material, the choice of materials shrinks. The lightest element used for
energy storage that can be structured into various forms to provide high surface
area and energy capacity is carbon.
In the past decade, there has been strong focus on the use of carbon nanotubes
(CNTs) for energy storage device fabrication. Besides being lightweight, CNTs pose
many advantages. They have a large surface area, up to 1315 m2 /g for singlewalled carbon nanotubes (compare to graphite, which exhibits a typical surface area
of 10–20 m2 /g), can be nanostructured, and mass-produced. However, CNTs
exhibit disadvantages, such as the presence of toxic residual metallic impurities,
which are very difficult to remove, and a high manufacturing cost.
The surface area of graphene is 2630 m2 /g, which is hugely favorable for energy
storage applications. Graphene is conductive and easy to functionalize with
other molecules.8 A family of graphene-related materials, called ‘‘graphenes’’ by the
research community, consists of structural or chemical derivatives of graphene.
14
The most important chemically derived graphene is graphene oxide (defined as
single layer of graphite oxide), which is usually prepared from graphite by
oxidization to graphite oxide and consequent exfoliation (i.e. by simple sonication,
long-time stirring of water/graphite oxide mixtures or thermal exfoliation) to
graphene oxide.
Information as to how the graphene was prepared is crucial because the
properties of graphene strongly depend on the method of fabrication. For example,
reduction of graphene oxide to graphene results in a graphene structure that is also
one-atom thick but which contains large numbers of defects, such as nanoholes
and Stone–Wales defects.
A Stone–Wales defect is a crystallographic defect that occurs on carbon
nanotubes and graphene and is thought to be responsible for nanoscale plasticity
and the plasticity and the brittle-ductile transitions in carbon nanotubes.
15
Nature Nanotechnology 2, 358 - 360 (2007)
Chemical storage: hydrogen
Hydrogen can be stored as a liquid, as pressurized gas, or under pressure
adsorbed on a substrate (sorbent).
The target specification of the US Department of Energy (DOE) for the use of
hydrogen for energy storage for mobile automotive applications is 6.5 wt% of
hydrogen in sorbent with a volume density of 62 kg H2 m3 at ambient temperature.
Such high capacity is challenging, when we consider that the stoichiometric ratio of
C–H is 7.7 wt% of hydrogen. There are two main approaches for hydrogen storage in
sorbents: (i) storage of molecular hydrogen or (ii) storage of atomic hydrogen based
on hydrogen spillover.
(a) storage of molecular hydrogen
Hydrogen is a non-polar molecule and its interactions with graphene-based
systems are based on instantaneous dipole–dipole induced forces, called
London dispersion forces (LDF).
London dispersion forces are a type of force acting between atoms and molecules.
They are part of the van der waals forces. The LDF is a weak intermolecular force
arising arising from quantum induced instantaneous polarization multipoles in
molecules. They can therefore act between molecules without permanent
16
multipole moments.
It has been shown that the spatial distribution of hydrogen adsorbed on graphene
is delocalized and molecular hydrogen exhibits free lateral movement.
There is a slight attractive force of -1.2 kJ mol-1 in a H2–graphene system at room
temperature. The free energy of physisorption corresponds to the equilibrium
constant of 1.6, which means that at room temperature, a single layer of graphene
increases the abundance of hydrogen by only 60%. Therefore using only the
surface of graphene/graphite for a H2 storage has no meaning. To improve the
binding capacity, it is possible to create a sandwich structure and take a H2 inbetween the graphene layers. Theoretical work shows that for graphene layers
separated by a distance of 6 A ° , one monolayer of a H2 can be accommodated
within the intergraphene structure (yielding a 2–3 wt% storage capacity at 5 MPa).
It is an attractive possibility to store two
layers of hydrogen molecules in a single
graphene–graphene interlayer. This is possible
by having two graphene layers separated by 8
A ° and would lead to gravimetric storage
capacities of 5.0–6.5% of H2 achievable under
technologically acceptable conditions (Fig. 1,
top). This method would require appropriately
separated graphene sheets using a nanoarchitectonics approach. Examples of stacked
graphene sheets separated by short carbon
nanotube pillars and by fullerenes.
17
(Fig. 1, top).
Another possibility is to use graphene as a
support material for metallic/metal oxide nanoparticles. It has been predicted that Kubas
interactions between H2 molecules and transition
metal are strong enough to provide significant
capacity for the binding of molecular hydrogen.
[see appendix]
Storing hydrogen on Ti atoms
(Fig. 1, bottom) deposited on graphene support.
Graphene oxide with its functional groups is an ideal material for anchoring
metals. It has been shown that titanium strongly bonds with the oxygen groups of
graphene oxide sheets (Fig. 1, bottom). Because each Ti atom can bind with
several H2 molecules, the theoretical capacity of such a lightweight metal–
graphene hybrid is 4.9 wt% and 64 g /L.
There have been several theoretical works showing that doped graphene,
especially with boron or aluminum, shows significantly improved binding capacity for
the adsorption of H2. Planar graphene, graphene-like C3N4 sheets, and pure boron
sheets decorated with alkali metals were also suggested in theoretical studies as
favorable materials for hydrogen storage with theoretical loading capacities of 7.8
wt% and 10.7 wt% of H2. Calcium-decorated graphene exhibits a storage capacity of
5 wt%. Ca preferentially adsorbs on the zigzag edge of graphene with a Ca–Ca
distance of 10 A ° ; each Ca atom can adsorb six molecules of H2.
18
Hornekaer et al. elucidated structure of hydrogen adsorbate on graphene by
STM. It was found that at low coverage, the formation of hydrogen dimers occurs
preferentially on protruding areas. At high coverage, random adsorption into large
hydrogen cluster was observed (see Fig. 2). As the hydrogen storage capacity
measurements are prone to artefacts, it is crucial to have method for investigation
of the space-size and hydrogen storage capacity of nanostructured carbon
materials.
Anderson et al. showed
that NMR can be used for carbon
nanospace characterization and
demonstrated that for nanopores
of size of less than 2.4 nm, uptake
up to 2.3 wt% of H2 at 0.2 MPa and
100 K were observed. Lueking et al.
demonstrated that it is possible to
store hydrogen in nanoporous
carbon and produce it with
subsequent carbon crystallization.
19
(b) storage of atomic hydrogen based on hydrogen spillover.
Hydrogen spillover arises in hydrogen catalysed reactions on supported metal
catalysts. Dihydrogen molecules dissociate on the metal part of the catalyst.
Some hydrogen atoms remain attached to the metal, whilst others diffuse to the
support and are said to spillover.
In the case of graphene, the
catalysts are typically Pt, Pd,
or Ni (or other transition metals)
and the support is graphene.
Hydrogen spillover can
be significantly enhanced
by building ‘‘bridges’’
between the catalyst and
the graphene/graphite
surface as this facilitates
the spillover. The
theoretical binding capacity can be as high as 7.7 wt% (having a stoichiometric
ratio of C–H as 1 : 1 for fully hydrogenated graphene), meeting DOE’s target.
Even though the spillover was experimentally observed some time ago, it has
never been clear why it is favorable for molecular hydrogen to dissociate on a
catalyst into its atomic form and further to spill over graphene sheets.
20
Hydrogen spillover is feasible even
before the metal clusters are fully
saturated with hydrogen. It has also been
demonstrated that it is thermodynamically
unfavorable for the spillover to occur on
pristine graphene but that hydrogenated
or doped graphene significantly
improves the hydrogen binding (Fig. 3).
Hydrogen spillover is feasible even
before the metal clusters are fully
saturated with hydrogen. It has also been
demonstrated that it is thermodynamically
unfavorable for the spillover to occur on
pristine graphene but that hydrogenated
or doped graphene significantly improves
the hydrogen binding (Fig. 3). Furthermore,
Stone–Wales types of defects on
graphene should significantly improve
binding capacity.
Fig. 3 Schematics of hydrogen spillover process in
real space (left). The inequality shows the range of
chemical potential mH favorable for spillover.
Right: model of spillover in energy space indicates
relative energy (chemical potential) of H in different
states. The gray, dark-red, and blue lines show the
mH in fully hydrogenated graphene (CH), in metal
hydride (PdH0.75), and in the H2 molecule,
respectively. The pink and dark-red blocks show the
range of energies of H at the catalyst and at the
Pd(111) surface with the H coverage varying from
0.25 to 1 ML. The family of thin, dark-blue lines
corresponds to the energies of H bound to
graphene.
21
Electrochemical storage: batteries
For the past several decades, there has been intensive research on the insertion of
Li+ ions into the lattice of graphite and, recently, there has been strong interest in
research on electrochemical energy storage in graphene-based systems. It is well
known that graphitic carbon can form LiC6 structures. The relatively low density
of lithium in graphite leads to the relatively low specific capacity of graphite, 372 mA
h/g. By employing individual graphene sheets, the storage capacity limit is 744 mA h /g
when lithium is stored on both sides of the graphene sheet, creating LiC3 structures.
From Wiki
22
Graphite-based electrodes suffer from the large lateral size of the graphite and the
long diffusion pathways of lithium into the material. The solution here is to minimize
the lateral (x–y axes) dimension so that Li+ can diffuse into the interlayer space
much easier with higher reversibility. Such material, which has lateral dimensions of
tens or hundreds of nanometres and a length of several micrometres is called
stacked platelet graphene nanofibers. Such materials exhibit outstanding
electrochemical properties because their surfaces exhibit practically only edge-like
sites and interlayer spacing of the graphite while basal planes are located only at the
ends of the nanofibers. Such small lateral sizes lead to an improved energy storage
capacity of 461 mA h/g.
The smaller the stacked graphene platelet nanofiber, the higher the energy storage
capacity. To further increase the specific capacity of Li+-based batteries, graphene
sheet-based electrodes were employed and achieved a capacity of 540 mA h/g1. It
was recognized that graphene sheets are prone to restacking, so when carbon
nanotubes or fullerenes were employed as spacers of individual graphene sheets, the
capacity of the Li+ battery was increased even more, to 730 or 784 mA h/g,
respectively (for an example, see Fig. 4).
23
Takamura et al. observed that nanosized holes in graphene sheets are
responsible for the high-rate discharge
capability of Li ion battery anodes.
This was a very important discovery
for further development of Li+
graphene-based batteries as the
graphene nanopores/nanoholes can
be functionalized to facilitate
selective ion passage.
Fig. 4 Graphene for batteries. (A) Relationship
between the d-spacing and the charge capacity of
graphene nanosheet (GNS) families and graphite.
(B) Cross-sectional TEM images of GNS families
with almost the same numbers (5–6) of graphene
stacking layers for (a) GNS, (b) GNS + CNT, and
(c) GNS + C60.
Note that capacity of the C60 separated
graphene nanosheets was higher than
theoretical capacity of LiC3 model. Pan
et al. demonstrated that beyond LiC3
model enhanced reversible capacities
observed in the highly disordered
graphene nanosheets may arise from
frequent defects in edge sites and
internal basal plane’s of multilayer
graphenes.
24
It was suggested that even though some Li+ ions may be stored in interfacial
layers between (002) graphene planes, the majority of the Li+ is stored in such
graphene defects (see Fig. 5).
The reaction of Li with the active
defects in discharge processes
proceeds at relatively low
potentials. However, release of Li
from the relatively strong bonds
with the defects during the charge
processes demands higher
voltages, and leads to the large
voltage hysteresis.
Fig. 5 (a) Irreversible lithium storage at the interface
between the graphene nanosheets and electrolyte; (b)
irreversible Li storage at edge sites and internal defects
of nanodomains embedded in graphene nanosheets;
(c) reversible Li storage between (002) planes.
25
Pan et al. found in their work capacity of high reversible capacities (794–1054
mA h/g) for disordered graphene. These high capacities are close to Li2 covalent
model proposed by Sato et al. for disordered carbon (theoretically up to 1116 mA
h/g for LiC2). Cyclic performance is crucial parameter for energy storage
materials. Even though initially the performance of the graphene based batteries
were not outstanding, current devices exhibit very good cycling performance
(90–95%). Here it should be mentioned that it was suggested that high specific
surface area leads to higher irreversible capacity as the solid electrolyte interface
between the graphene and electrolyte forms.
Metal/graphene nanocomposites are another route to improve the capacity of
lithium batteries.
26
Electrical storage: supercapacitors
Supercapacitors (also called ultracapacitors) are devices that are able to store
energy and release it within a short time interval with high power capability and
large current density. They exhibit high dynamic charge propagation and long
cycle life (over 100 000 cycles).
The principle of energy storage in a supercapacitor can be either (i) pure charge
storage on an electrode/ electrolyte interface by electrochemical double layer
capacitance (EDLC) or (ii) it can transfer a charge to the layer of redox molecules
on the surface of the electrode. The EDLC mechanism is directly proportional to
the surface area in contact with the solution. The latter mechanism is called
pseudo-capacitance and, even though it is possible to store more energy by
pseudo-capacitance than by EDLC, it suffers from drawbacks such as lower power
density and lower stability during cycling.
The typical materials used for pseudo-capacitance are oxides and nitrides such
as MnOx, RuOx, and VN, as well as conducting polymers such as polyaniline. It
should be noted that many supercapacitors use both mechanisms.
27
Graphene with its maximal surface area of 2630 m2/g is an ideal medium for
supercapacitors as the EDLC is directly proportional to the surface area. This was
first explored by Ruoff and co-workers who found that chemically modified
graphene (CMG) exhibits large capacitances of 135 and 99 F/g for aqueous and
organic electrolytes, respectively (see Fig. 6).
Fig. 6 (a) SEM image of chemically
modified graphene (CMG) particle
surface, (b) TEM image showing
individual graphene sheets extending
from CMG particle surface, (c) low
and high (inset) magnification SEM
images of CMG particle electrode
surface, and (d) schematic of test cell
assembly.
28
The different studies exhibit different values because the exfoliated chemically
reduced graphene used tends to restack and to create graphitic structures, thus
reducing the maximum theoretical surface as only the surfaces in contact with the
solution contribute to overall capacitance. It was suggested that spacing the
graphene layers with Pt nanocrystals 4 nm in diameter would improve the
capacitance from 14 F/g for dried restacked ‘‘graphene’’ to 269 F/g for Pt-separated
graphene sheets.
Graphene sheets can be used to support materials having large pseudo
capacitance in order to minimize their clustering and maximize the electrochemically
accessible area. MnO2 nanocrystals were synthesized on graphene oxide sheets and
provided a large capacitance of 197 F/g, which was significantly higher than that of
graphene oxide (10.9 F/g) and bulk MnO2 (6.8 F/g). A much higher capacitance of
1335 F/g was exhibited by Ni(OH)2 nanoplates grown on graphene sheets. It was
shown that Ni(OH)2 nanoplates grown on graphene sheets significantly outperform
Ni(OH)2 nanoparticles grown on heavily oxidized graphene oxide sheets, which are
electrically insulating.
29
Graphene/polymer nanocomposites are very attractive for use in supercapacitors.
Graphene/polyaniline-based nanomaterials are easy to prepare electrochemically or
chemically (for schematics, see Fig. 7) and they can exhibit capacitances ranging
from 233 F/g to 1046 F/g, depending on the nanostructure of the composite. Very
high cycle stability was achieved on poly(sodium 4-styrensulfonate) (PSS)–graphene
nanocomposite supercapacitor.
After 14 860 cycles, the specific
capacitance (of 190 F/g) decreased
only by 12%. The graphene/polymer
composites are ideally suitable for
portable and wearable electronics
as they are flexible and they retain
their properties even if they are
under mechanical stress. Wu et al.
demonstrated flexible polyaniline/
graphene composite with capacity
of 210 F/g and cycling stability of
94%.
30