Jincy Parayangattil Jyothibasu1,2, Ming-Zhu Chen1, Rong-Ho Lee1. 1. Department of Chemical Engineering, National Chung Hsing University, Taichung 402, Taiwan. 2. Department of Environmental Engineering, National Chung Hsing University, Taichung 402, Taiwan.
Abstract
In this study, a facile and environmentally friendly method was used to prepare a freestanding supercapacitor electrode displaying excellent areal capacitance and good cycle life performance. First, we prepared polypyrrole nanoparticles (PPyNP) through a simple in situ chemical polymerization using the plant-derived material curcumin as a bioavailable template. A PPyNP/f-CNT freestanding composite electrode of high mass loading (ca. 14 mg cm-2) was prepared after blending the mixtures of the prepared PPyNP and functionalized CNTs (f-CNTs). The performance of the as-prepared material as a supercapacitor electrode was evaluated in a three-electrode setup using aqueous 1 M H2SO4 as the electrolyte. The PPyNP/f-CNT freestanding composite electrode exhibited a high areal capacitance of 4585 mF cm-2 and a corresponding volumetric capacitance of 176.35 F cm-3 at a current density of 2 mA cm-2. A symmetric all-solid-state supercapacitor assembled using two identical pieces of PPyNP/f-CNT composite electrodes exhibited maximum areal energy and power density of 129.24 μW h cm-2 and 12.5 mW cm-2, respectively. Besides, this supercapacitor device exhibited good cycle life performance, with 79.03% capacitance retention after 10,000 charge/discharge cycles. These results suggest practical applications for these PPyNP/f-CNT freestanding composite electrode-based symmetric all-solid-state supercapacitors.
In this study, a facile and environmentally friendly method was used to prepare a freestanding supercapacitor electrode displaying excellent areal capacitance and good cycle life performance. First, we prepared polypyrrole nanoparticles (PPyNP) through a simple in situ chemical polymerization using the plant-derived material curcumin as a bioavailable template. A PPyNP/f-CNT freestanding composite electrode of high mass loading (ca. 14 mg cm-2) was prepared after blending the mixtures of the prepared PPyNP and functionalized CNTs (f-CNTs). The performance of the as-prepared material as a supercapacitor electrode was evaluated in a three-electrode setup using aqueous 1 M H2SO4 as the electrolyte. The PPyNP/f-CNT freestanding composite electrode exhibited a high areal capacitance of 4585 mF cm-2 and a corresponding volumetric capacitance of 176.35 F cm-3 at a current density of 2 mA cm-2. A symmetric all-solid-state supercapacitor assembled using two identical pieces of PPyNP/f-CNT composite electrodes exhibited maximum areal energy and power density of 129.24 μW h cm-2 and 12.5 mW cm-2, respectively. Besides, this supercapacitor device exhibited good cycle life performance, with 79.03% capacitance retention after 10,000 charge/discharge cycles. These results suggest practical applications for these PPyNP/f-CNT freestanding composite electrode-based symmetric all-solid-state supercapacitors.
The
ever-growing market for wearable electro-optical devices has
intensified the need for inexpensive, reliable, lightweight, eco-friendly,
and safe energy storage devices.[1,2] Electrochemical supercapacitors
have several attractive features, including large power density, fast
charging/discharging rates, high cycling stability, low fabrication
costs, and environmental friendliness, which make them promising energy
sources for such electronic devices.[3,4] Nevertheless,
the low energy densities of conventional supercapacitors have limited
their practical applications.[5] To enhance
the energy densities, it will be necessary to design electrode materials
to have high capacitances, large electrochemically accessible surface
areas, suitable porous structures, and high electrical conductivities.[6] During the fabrication of conventional supercapacitor
electrodes, the active materials are usually mixed with binders and
conductive additives, with the resulting slurries pressed onto metallic
current collectors. Adding these inactive components enhances the
cost of electrode fabrication and the total weight of the devices,
making them less compatible with portable and wearable electronics.
Moreover, the inherent resistance of the binder materials decreases
the electrode’s surface area and conductivity, leading to a
low electrochemical reaction rate, further hindering the applicability
in high-performance supercapacitors.[7] Thus,
to achieve high performance, it will be necessary to fabricate freestanding
supercapacitor electrodes that do not involve binders, conductive
additives, or current collectors.Polypyrrole (PPy) is a promising
candidate for supercapacitor electrodes
because of its large theoretical capacitance, good redox properties,
superior conductivity, ease of synthesis, nontoxicity, biocompatibility,
and high thermal and environmental stability.[8−10] Unfortunately,
pristine PPy usually undergoes structural damage when subjected to
long-term cycling because of the repeated swell–shrink behavior
of the polymer backbone during the ion doping and dedoping processes
of PPy, resulting in more unsatisfactory electrochemical performance
and lower cyclic stability when compared with carbon-based materials.[11−14] Significant improvements in the electrochemical energy storage ability
and cycling charge–discharge stability of PPy-based electrodes
can be realized by properly tuning the size and morphology of the
PPy to enhance the specific surface area and porosity.[15] Compared with their bulk counterparts, nanostructured
forms of PPy (e.g., nanospheres,[4,16,17] nanosheets,[18] nanorods/tubes/wires,[10,19−22] and nanobrushes[23]) can have significantly
greater effective surface areas and porosities that, in turn, result
in greater use of active materials by shortening the diffusion path
for ions to maximize electrolyte exposure.[4,7] The
morphologies and thereby the subsequent electrochemical property of
the PPy nanostructures are highly related to the synthetic methods
used for their preparation. Such nanoscale PPy structures are commonly
prepared using chemical oxidative polymerization,[5,17,24,25] vapor phase
polymerization,[23,26] electropolymerization,[27] and interfacial polymerization[15,22,28,29] in the presence of hard or soft templates to fine-tune the final
morphologies. The syntheses involving the hard templates require harsh
post-treatment processes to etch away the hard templates and recover
the pristine PPy nanostructures, making such processes complex and
time-consuming.[4] Therefore, soft-template
methods, involving surfactant micelles, functionalized organic acids,
or polymeric stabilizers, are preferred over hard-template methods
for the formation of desired morphologies of PPy. Although the syntheses
of PPy nanostructures using various soft templates (e.g., cetyltrimethylammonium bromide,[22] sodium
dodecylbenzene sulfonate,[5,24] and N-methylene phosphonic chitosan[17]) can
be simple and effective, they can be expensive and possibly environmentally
contaminating when prepared on bulk scales. Therefore, for the bulk
production of PPy nanostructures, the quest remains to develop synthetic
methods that are less complicated, more environmentally friendly,
less time-consuming, less expensive, cleaner, and more efficient.To improve its electrochemical properties and cycling stability,
PPy is often combined with carbon-based electrical double-layer capacitance
(EDLC) materials to form nanocomposite electrodes. Carbon nanotubes
(CNTs) are particularly effective additives for forming highly efficient
supercapacitor electrodes with PPy because of their exceptional mechanical
properties, large surface areas, and superior electrical conductivities.
Composite electrodes developed from combinations of PPy and CNTs have
displayed enhanced electrochemical properties in terms of improved
cycling stabilities, power densities, and rate capabilities. In addition,
CNTs can also increase the mechanical strength of PPy/CNT-based composites
by serving as structural-reinforcing agents. Taking advantage of the
synergy between CNTs with high conductivity and PPy with high energy
density, PPy/CNT nanocomposites can effectively mitigate the drawbacks
of the low specific capacitance of CNTs and the poor cycling life
of PPy. Nevertheless, the synthetic routes for the preparation of
PPy/CNT nanocomposites can have a great influence on their structural
and electrochemical properties. For example, most previously reported
PPy/CNT nanocomposites have been synthesized through in situ chemical polymerization of the pyrrole in a dispersion containing
CNT; they have been obtained in powder form and thereby have required
the use of binders, conductive additives, and current collectors to
fabricate supercapacitor electrodes.[30−32] Accordingly, efforts
have been made to fabricate freestanding PPy/CNT nanocomposites involving
the direct growth of PPy through chemical polymerization or electrodeposition
on preformed CNT films, thereby enabling high interfacial conductance
between PPy and CNT and resulting in improved supercapacitor performance.
It has been difficult, however, to deposit large amounts of PPy on
CNTs without blocking the electrolyte channels when using such methods.[33,34] This limited mass loading results in poor areal capacitance—an
undesirable feature for practical applications.[35] In general, electrodes with very low active-material mass
loadings often result in overestimated gravimetric capacitance values.[36] In reality, the gravimetric capacitance will
not increase upon increasing the active-material loading; instead,
performance matrices of the supercapacitor will all tend to decrease
accordingly. Therefore, to achieve a true estimation of the electrochemical
properties of the supercapacitors for practical applications, the
supercapacitor electrodes must be prepared with high mass loadings
of the active materials (>10 mg cm–2).[37] Nevertheless, achieving a high energy/power
density, rate performance, and cycling charge–discharge stability
at a high mass loading remains a scientific challenge.[38]In this study, a composite film based
on PPy nanoparticles and f-CNTs was fabricated through
a green route using the plant-derived
material curcumin as a template. The as-prepared composite film was
then directly used as a freestanding electrode for supercapacitors.
The electrode with high mass loading achieved a superior areal capacitance
of 4585 mF cm–2 and an excellent volumetric capacitance
of 176.35 F cm–3. Furthermore, the symmetric supercapacitor
cell assembled using the binder-free electrodes exhibited maximum
areal energy and power density of 129.24 μW h cm–2 and 12.5 mW cm–2, respectively, and excellent
cycle life (retained 79.03% capacitance after 10,000 cycles).
Results and Discussion
The fabrication process of the
PPyNP/f-CNT nanocomposite
film is illustrated in Figure . First, curcumin and the pyrrole monomer were dissolved in
ethanol. Curcumin is a natural polyphenol produced by the plant Curcuma longa of the ginger family.[39] Curcumin is poorly soluble in water because of its aromatic
rings. Therefore, the addition of water resulted in the precipitation
of curcumin crystals, which could serve as anchoring sites for the
pyrrole monomer due to the hydrophobic nature of its phenyl rings
and the potential for hydrogen bonding of its side groups with the
NH unit of pyrrole.[40,41] The addition of aqueous ferric
chloride solution initiated the polymerization of pyrrole. After the
polymerization was complete, anhydrous ethanol was used to remove
the curcumin template and obtain the interconnected PPy nanoparticles.
Figure 1
Schematic
representation of the preparation of PPy nanoparticles
(PPyNP).
Schematic
representation of the preparation of PPy nanoparticles
(PPyNP).SEM and TEM were used to study
the nanostructure of the PPy samples
prepared with and without curcumin. The presence of curcumin in the
polymerization medium had a strong influence on the morphology and
size of the PPy nanoparticles. In the absence of the curcumin template,
PPyC was formed as dense irregular agglomerates with sizes ranging
from 200 to 300 nm (Figure a). These agglomerated structures hinder the movement of electrolyte
ions. When the polymerization was performed in the presence of the
curcumin template, the PPyNP structures were considerably smaller
(60–80 nm) and the interconnected nanoparticles formed a loosely
porous morphology (Figure b). The porous morphology of PPyNP would enhance the interfacial
area between the PPy nanoparticles and the electrolyte, promoting
maximum active-material utilization. Moreover, a ready infiltration
of electrolytes into the porous composite film matrix would facilitate
the rapid movement of ions during charge/discharge processes. The
TEM images of PPyC and PPyNP were in agreement with the SEM images.
PPyNP in Figure d
reveals interconnected PPy nanoparticles forming a porous structure;
in contrast, a dense agglomerated morphology was observed in the TEM
image of PPyC (Figure c).
Figure 2
(a,b) SEM and (c,d) TEM images of (a,c) PPyC and (b,d) PPyNP.
(a,b) SEM and (c,d) TEM images of (a,c) PPyC and (b,d) PPyNP.Figure presents
the SEM images of freestanding films of f-CNT, PPyC/f-CNT, and PPyNP/f-CNT, respectively. The
image in Figure a
reveals a porous conductive network of entangled f-CNTs. For both the PPyC/f-CNT and PPyNP/f-CNT freestanding electrodes, the f-CNTs
provided structural integrity and enabled the formation of highly
conductive freestanding films without the need for any polymer binder,
conductive additives, or current collector. In the absence of the f-CNTs, it was not possible to form freestanding electrodes
because of the poor mechanical integrity of the pristine PPyC and
PPyNP.
Figure 3
SEM images of (a) f-CNT, (b) PPyC/f-CNT, and (c) PPyNP/f-CNT composite films.
SEM images of (a) f-CNT, (b) PPyC/f-CNT, and (c) PPyNP/f-CNT composite films.FTIR spectra of the PPyC, PPyNP, f-CNT, and PPyNP/f-CNT composite films are shown
in Figure . Figure S1 presents
the FTIR spectrum of curcumin. The absorption bands at 3510 and 2849
cm–1 may be designated as the O–H vibration
and C–H stretching vibration of the OCH3 unit, respectively.[42] The absorption peak at 1628 cm–1 is due to the stretching vibrations of the C=O and C=C
units.[43] The other characteristic absorption
peaks at 1510, 1430, 1376, and 1283 cm–1 represent
the stretching vibrations of the C=O group,[43] olefinic C–H bending vibration,[42] the CH3 bending vibration,[42] and the aromatic enol C–O stretching,[43] respectively. The absorption peaks of the symmetric
and asymmetric C–O–C vibrations appear at 1154 and 1027
cm–1.[43] The absorption
peaks at 964, 814, and 714 cm–1 represent the benzoate trans-CH vibration,[43] the out-of-plane
C–CH vibrations of the aromatic rings,[42] and cis-CH vibrations of the aromatic rings, respectively.[43] The absorption peaks of curcumin were absent
from the FTIR spectrum of PPyNP, but it displayed the characteristic
signals of pure PPy, similar to those in the spectrum of PPyC. In
the spectra of PPyC and PPyNP, a broad band centered around 3118 cm–1 is designated as the N–H stretching vibration
of the pyrrole ring.[10,44] The small absorption band at
2890 cm–1 may be assigned to the symmetric vibrations
of the CH2 units.[45] The band
at 1555 cm–1 is due to the combination of C=C
and C–C stretching vibrations of PPy.[46] The signal appearing at 1462 cm–1 is ascribed
to C–N stretching vibrations, while the band at 1314 cm–1 represents the C–H and C–N in-plane
deformation modes.[47] The absorption bands
near 1188 and 924 cm–1 indicate the presence of
a doped PPy state.[46,48] The absorption band at 1046 cm–1 is designated as the C–H in-plane deformation
vibration of the PPy ring,[46] while the
signal at 790 cm–1 is due to the C–H out-of-plane
ring deformations.[47] The characteristic
absorption bands of PPy underwent various redshifts in the FTIR spectrum
of the PPyNP/f-CNT nanocomposite, suggesting strong
interactions between PPyNP and f-CNT in this composite.[49] The broad band near 3449 cm–1 represents the O–H stretching vibration in f-CNT.
Figure 4
FTIR spectra of f-CNT, PPyC, PPyNP, and the PPyNP/f-CNT freestanding composite film.
FTIR spectra of f-CNT, PPyC, PPyNP, and the PPyNP/f-CNT freestanding composite film.Figure displays
the XRD patterns of the prepared samples. For f-CNTs,
one broad peak assigned to the (002) lattices of hexagonal graphite
has appeared at 2θ = 25.6°; a low-intensity peak appeared
at 43.4°, corresponding to diffractions of the graphitic planes
(110) and (100) collectively. In the XRD patterns of PPyC and PPyNP,
the characteristic broad diffraction peak of PPy was centered at about
2θ = 25°, revealing an amorphous nature of the PPy.[4,45] The characteristic high-intensity diffraction peaks of curcumin
were located at values of 2θ of 8.9, 12.2, 14.5, 15.7, 17.3,
19.5, 21.3, 23.8, 26.1, 28.9, and 31.6° in the XRD pattern, indicating
its well-developed crystalline nature.[43,50] The diffraction
peaks of PPyNP were almost identical to those of PPyC, indicating
the formation of pristine PPy nanoparticles without the introduction
of additional peaks resulting from the presence of curcumin. In the
XRD pattern of the PPyNP/f-CNT nanocomposite, the
broad peak of the PPy polymer backbone merged with the f-CNT peak at 2θ = 25°. In addition, the low-intensity
peak of the f-CNTs near 43° also appeared in
the XRD pattern of the PPyNP/f-CNT nanocomposite.
Figure 5
XRD patterns
of the f-CNTs, PPyC, PPyNP, and the
PPyNP/f-CNT freestanding composite film.
XRD patterns
of the f-CNTs, PPyC, PPyNP, and the
PPyNP/f-CNT freestanding composite film.An electrode material having a large effective surface area
increases
the electrolyte accessibility and thereby allows complete utilization
of the active materials during the electrochemical processes. Thus,
their surface area and pore distribution in the electrode both play
critical roles in determining the electrochemical properties. Figure and Figure S2 demonstrate the N2 sorption
isotherms of the samples. The composite films provided type-IV isotherms,
indicative of porous structures that were mainly composed of mesopores.
The BET surface area of the f-CNTs was 270.6 m2 g–1, and the average pore diameter was
21.42 nm. The specific surface area of PPyNP (32.5 m2 g–1) was much larger compared with that of PPyC (12.4
m2 g–1). PPyNP and PPyC displayed average
pore diameters of 18.2 and 23.0 nm, respectively, confirming their
mesoporous structures. We attribute the low specific surface area
of PPyC to the agglomeration of the PPy nanoparticles resulting in
a compact morphology, which was clearly observed in the TEM and SEM
images. The specific surface areas of the PPyNP/f-CNT and the PPyC/f-CNT composites were 85.6 m2 g–1 and 31.03 m2 g–1, respectively. The high specific surface areas of the PPyNP/f-CNT and PPyC/f-CNT composites, compared
with those of pure PPyNP and PPyC, might result from the incorporation
of the f-CNTs with a larger surface area. Apart from
that, the average pore diameters of the PPyNP/f-CNT
and the PPyC/f-CNT composites, calculated from the
BJH pore size distribution curves, were 17.24 and 30.49 nm, respectively,
confirming that the porous structure was dominated by mesopores. The
PPyNP/f-CNT freestanding composite with a higher
specific surface area, coupled with suitable pore distributions, would
provide a high electrolyte contact area that might potentially facilitate
rapid ion diffusion.
Figure 6
Nitrogen sorption isotherms of (a) f-CNTs,
(b)
PPyC, (c) PPyNP, and (d) PPyNP/f-CNT freestanding
film. Inset: pore size distribution curves.
Nitrogen sorption isotherms of (a) f-CNTs,
(b)
PPyC, (c) PPyNP, and (d) PPyNP/f-CNT freestanding
film. Inset: pore size distribution curves.The wettability of f-CNTs, PPyC/f-CNT, and PPyNP/f-CNT freestanding films was measured
using water contact angle analysis (Figure ). The wetting behavior of a porous film
is governed mainly by the chemical composition, the geometrical structure
of solid surfaces, and the capillary forces induced by strong intermolecular
interactions between the liquid and the substance.[51,52] The water contact angle of PPyNP/f-CNT film (39.35°)
is slightly smaller than that of f-CNT (44.58°)
and PPyC/f-CNT (44.62°) films, indicating its
good wettability. The better wetting behavior of the PPyNP/f-CNT composite film would suggest ready penetration of
the electrolyte into the porous freestanding electrode matrix, potentially
enabling maximum use of the active materials.
Figure 7
Contact angle measurements.
Photographs of water droplets on (a) f-CNT, (b) PPyC/f-CNT, and (c) PPyNP/f-CNT freestanding
films.
Contact angle measurements.
Photographs of water droplets on (a) f-CNT, (b) PPyC/f-CNT, and (c) PPyNP/f-CNT freestanding
films.The electrochemical behaviors
of the f-CNT-, PPyC/f-CNT-, and
PPyNP/f-CNT-based composites
were evaluated in a three-electrode setup. Figure a presents the CV (cyclic voltammetry) plots
measured at a 5 mV s–1 scan rate. The f-CNT electrode displayed a CV curve that is close to a
rectangular shape, indicative of an almost ideal electrical double-layer
capacitance. PPyNP/f-CNT and PPyC/f-CNT electrodes exhibited quasi-rectangular curves, indicative of
the capacitance being contributed by the PPy (pseudocapacitance) and
the f-CNTs (double-layer capacitance). The PPyNP/f-CNT-based electrode displayed a much larger integrated
area inside the CV curve than f-CNT- and PPyC/f-CNT-based electrodes, revealing the former’s superior
capacitance. Figure b provides the GCD (galvanostatic charge/discharge) curves recorded
at a 2 mA cm–2 current density. The f-CNT-based electrode with a highly symmetric triangular-shaped
charge/discharge curve confirmed its typical double-layer capacitance
behavior with good electrochemical reversibility. The existence of
pseudocapacitive behavior in both the PPyNP/f-CNT-
and PPyC/f-CNT-based composites was confirmed by
their slightly distorted triangular-shaped GCD curves. The PPyNP/f-CNT electrode had the longest discharge time and the highest
areal capacitance (4585 mF cm–2) compared with the f-CNT electrode (563 mF cm–2) and the
PPyC/f-CNT based electrode (3611 mF cm–2). Besides, the PPyNP/f-CNT-based electrode attained
corresponding gravimetric and volumetric capacitances of 327.5 F g–1 and 176.35 F cm–3, respectively.
The areal, gravimetric, and volumetric capacitances obtained for the
PPyNP/f-CNT-based electrode were larger than those
reported previously for PPy-based composite electrodes (Table S1). We attribute the higher capacitance
of the PPyNP/f-CNT electrode to the greater use of
the inner active redox sites in its PPyNP structure, resulting from
the larger active surface area and greater wettability of this composite
electrode and its porous architecture that facilitated rapid electrolyte
ion diffusion. The electrolyte ions could not access the interior
of the PPyC in the case of the PPyC/f-CNT freestanding
electrode due to its more compact structure that hindered the electrolyte
diffusion, leading to an inferior capacitance.
Figure 8
Electrochemical behavior
of the f-CNT, PPyC/f-CNT, and PPyNP/f-CNT electrodes: (a)
CV curves and (b) GCD curves measured at a 5 mV s–1 scan rate and a 2 mA cm–2 current density, respectively.
Electrochemical behavior
of the f-CNT, PPyC/f-CNT, and PPyNP/f-CNT electrodes: (a)
CV curves and (b) GCD curves measured at a 5 mV s–1 scan rate and a 2 mA cm–2 current density, respectively.Figure displays
the CV plots of the PPyC/f-CNT-, PPyNP/f-CNT-, and f-CNT-based electrodes measured at different
scan rates. As revealed in Figure a–c, at all of the tested scan rates, the current
responses of the PPyNP/f-CNT-based electrode were
much larger compared with the f-CNT- and PPyC/f-CNT-based electrodes, indicating its superior capacitance
properties. The GCD plots of the f-CNT-, PPyC/f-CNT-, and PPyNP/f-CNT-based electrodes
(Figure d–f,
respectively) revealed that the f-CNT-based electrode
provided symmetric triangular-shaped charge–discharge curves
without obvious IR drops at all current densities, suggesting low
intrinsic resistance and rapid charge/discharge kinetics.
Figure 9
(a–c)
CV and (d–f) GCD plots at various current densities
of (a,d) f-CNT, (b,e) PPyC/f-CNT,
and (c,f) PPyNP/f-CNT electrodes.
(a–c)
CV and (d–f) GCD plots at various current densities
of (a,d) f-CNT, (b,e) PPyC/f-CNT,
and (c,f) PPyNP/f-CNT electrodes.Figure a summarizes
the areal capacitances of the prepared electrodes at different current
densities. The capacitances of all freestanding electrodes decreased
as the current density increased because of the limited electrolyte
ion accessibility to the internal redox sites at higher current densities
resulting in insufficient use of the PPy. The capacitance decay at
high current densities was less significant for the f-CNT freestanding electrode than it was for the PPyNP/f-CNT and PPyC/f-CNT electrodes because of its excellent
conductivity and mesoporous structure. Nevertheless, the PPyNP/f-CNT freestanding electrode displayed a high capacitance
retention of 2752 mF cm–2 at a 10 mA cm–2 current density.
Figure 10
(a) Areal capacitances vs. current density plot and (b)
EIS spectra
of the f-CNT-, PPyC/f-CNT-, and
PPyNP/f-CNT-based electrodes.
(a) Areal capacitances vs. current density plot and (b)
EIS spectra
of the f-CNT-, PPyC/f-CNT-, and
PPyNP/f-CNT-based electrodes.EIS tests of the freestanding electrodes were performed to obtain
further information regarding the electrochemical characteristics
(Figure b). The
intercept of the semicircle in the high-frequency region is known
as the equivalent series resistance (ESR), which consists of combined
intrinsic resistances of the electrode materials and the electrolyte.[53,54] The low ESR in the high-frequency region and the lack of a semicircle
for the pure f-CNT-based electrode indicate that
its rapid and reversible charge storage kinetics was due to high conductivity.[55] The Nyquist plots of the composite electrodes
featured semicircles at high frequencies, indicating the existence
of charge transfer resistance across the interface of the electrode
and electrolyte. The vertical lines with slopes at low frequencies
correspond to diffusion of electrolyte in the porous active materials.[51,53,56] The PPyNP/f-CNT
composite with a porous morphology resulted in the low electrolyte
resistance in the pores of the active material.[54] As a result, the ESR of the PPyNP/f-CNT-based
electrode (ca. 2.99 Ω) was lower than the ESR of the PPyC/f-CNT electrode (ca. 3.86 Ω). Figure S3a displays the cycling stability of the PPyNP/f-CNT freestanding electrode recorded at a high current
density of 60 mA cm–2 for 10,000 charge/discharge
cycles. The freestanding electrode exhibited good cycling stability
with the retention of 71.6% of the initial capacitance after 10,000
charge/discharge cycles. The SEM images of the PPyNP/f-CNT freestanding electrode exhibited similar morphologies before
(Figure c) and after
the cycle test (Figure S3b). This suggests
that the volume changes of PPyNP that occurred during ion doping and
dedoping processes did not cause the separation of PPyNP and f-CNT. The electrochemical tests revealed that the PPyNP/f-CNT-based composite has a better performance than the
PPyC/f-CNT electrode. We attribute the excellent
capacitance performance of the PPyNP/f-CNT-based
electrode to (i) the high active-material mass loadings; (ii) the
high active surface area of the freestanding electrode, providing
more redox sites for charge storage; (iii) the higher material use,
resulting from the synergistic effect between the highly conductive f-CNTs and PPy; (iv) the increase in interfacial contact
between the small PPy nanoparticles and the electrolyte; (v) the interconnected
porous structure, which provided electrolyte reservoirs that ensured
efficient infiltration of the electrolyte into the entire porous matrix
of the electrode material; (vi) the high electrical conductivity,
which facilitated rapid charge transport; and (vii) the good wettability
of the electrode.To further study the electrochemical properties
for practical applications,
a symmetric supercapacitor device was fabricated using two identical
PPyNP/f-CNT freestanding electrodes and a PVA/H2SO4 gel electrolyte. The quasi-rectangular-shaped
CV curves indicate a combination of EDL capacitance and pseudocapacitance
toward charge storage (Figure a). The charge/discharge tests were performed to further
evaluate the capacitance properties of the supercapacitor device (Figure b). The nonlinear
shapes of the GCD curves, indicative of Faradaic redox reactions,
are in well agreement with the CV results. The slow diffusion of ions
in the gel electrolyte and the poor interfacial contact between the
gel electrolyte and the electrode material resulted in a large IR
drop in the discharge curve and reduced areal capacitance at higher
specific currents.[57] The areal, gravimetric,
and volumetric capacitances of the assembled supercapacitor were estimated
using eqs , 4, and 7. The supercapacitor
exhibited superior areal (1454.06 mF cm–2), gravimetric
(103.86 F g–1), and volumetric capacitances (17.52
F cm–3) at a current density of 2 mA cm–2. Figure c presents
the impedance plot of the supercapacitor cell. The Nyquist plot featured
a semicircle and a vertical line at high and low frequencies, respectively,
suggesting good capacitance performance. The ESR of the supercapacitor
cell was approximately 11.88 Ω.
Figure 11
Electrochemical behaviors
of the symmetric supercapacitor. (a)
CV and (b) GCD plots at various scan rates and current densities,
respectively. (c) EIS spectrum.
Electrochemical behaviors
of the symmetric supercapacitor. (a)
CV and (b) GCD plots at various scan rates and current densities,
respectively. (c) EIS spectrum.The energy and power densities were estimated based on the GCD
data; Figure a shows
the corresponding Ragone plot. The prepared supercapacitor device
exhibited maximum areal (gravimetric, volumetric) energy densities
of 129.24 μW h cm–2 (9.23 W h kg–1, 1.56 mW h cm–3) with power densities of 0.858
mW cm–2 (61.31 W kg–1, 10.34 mW
cm–3). The device maintained areal (gravimetric,
volumetric) energy densities of 51.06 μW h cm–2 (3.65 W h kg–1, 615.18 μW h cm–3) at maximum power densities of 12.5 mW cm–2 (892.86
W kg–1, 150.6 mW cm–3). The areal/gravimetric/volumetric
capacitances and energy/power densities of the PPyNP/f-CNT electrode-based supercapacitor device were considerably larger
than those of other literature-reported PPy-based supercapacitors,
including a CNFs/rGO/PPy aerogel (720 mF cm–2, 60.4
μW h cm–2 at 0.1 mW cm–2),[58] RGO/PPy/cellulose hybrid papers (510
mF cm–2, 1.18 mW h cm–3), a PPy-coated
air-laid paper (702 mF cm–2, 62.4 μW h cm–2 at 0.42 mW cm–2),[59] a KF@PPy/f-CNT-based electrode (258 mF
cm–2, 22.3 μW h cm–2 at
0.21 mW cm–2),[8] an lCNT-GO/PPy
composite electrode (72.3 mF cm–2, 6.3 μW
h cm–2, 3.7 mW cm–2),[60] and an RGO/PPy CCFs electrode (363 mF cm–2, 0.28 mW h cm–3 at 20.9 mW cm–3).[61] Despite having many
attractive features, supercapacitor electrodes based on PPy have a
major shortcoming of poor cycling charge–discharge stability
due to repeated swelling and contracting of the polymer chains while
undergoing charging and discharging, respectively. Thus, the capacitance
degradation that occurs during prolonged cycling due to the volume
changes caused by the insertion and de-insertion of counterions limits
their practical applications.[10,62,63]Figure b displays
the cycling stability of the supercapacitor tested at a high galvanostatic
charge/discharge current density of 20 mA cm–2.
Capacitance retention with 79.03% of the initial capacitance after
10,000 cycles confirms the superior cycling charge/discharge stability
of the supercapacitor cell. We attribute this remarkable cycle life
to the morphological flexibility of the PPy nanoparticle, thereby
accommodating the swell–shrink behavior of the polymeric backbone
during long-term cycling.[64]
Figure 12
(a) Ragone
plot and (b) cycle life test (10,000 cycles at 20 mA
cm–2) of the symmetric supercapacitor cell.
(a) Ragone
plot and (b) cycle life test (10,000 cycles at 20 mA
cm–2) of the symmetric supercapacitor cell.
Conclusions
We have
prepared PPy nanoparticles by using curcumin, a renewable
biomaterial, as the structure-directing agent. A highly porous film
was obtained having a high surface area and conductivity after blending
the as-prepared PPy nanoparticles with f-CNTs. This
film was then used as a freestanding electrode for supercapacitors.
The PPyNP/f-CNT-based electrode exhibited superior
areal capacitance (4585 mF cm–2) and volumetric
capacitance (176.35 F cm–3) at 2 mA cm–2. The remarkable electrochemical performance of the as-prepared composite
electrode was ascribed to the high active-material mass loadings,
the large effective surface area, the synergy between the highly conductive f-CNTs and the pseudocapacitive PPy, the increased interfacial
contact between the small PPy nanoparticles and the electrolyte, and
the interconnected porous morphology and excellent conductivity of
the electrode. A symmetric all-solid-state cell based on the freestanding
PPyNP/f-CNT-based electrodes exhibited maximum areal
energy and power densities of 64.62 μW h cm–2 and 6.25 mW cm–2, respectively. Besides, the cell
also exhibited superior cycle stability with a retention of 79.03%
capacitance after 10,000 charge/discharge cycles. These results suggest
the potential application of the PPyNP/f-CNT composite
in all-solid-state supercapacitors.
Experimental
Section
Materials
The CNTs were purchased
from C-nano Technology (China). Curcumin [1,7-bis(4-hydroxy-3-methoxyphenyl)-1,6-heptadiene-3,5-dione],
poly(vinyl alcohol) (PVA), sulfuric acid (H2SO4), ferric chloride hexahydrate (FeCl3·6H2O), nitric acid (HNO3), and ethanol (C2H5OH) were procured from Showa Chemical Industry
(Japan). Pyrrole was procured from Alfa Aesar (UK), which was distilled
under reduced pressure prior to use.
PPy Nanoparticles
(PPyNP)
Curcumin
and pyrrole were dissolved in anhydrous ethanol (50 mL). The precipitation
of curcumin was induced by the addition of DI water (200 mL). The
dropwise addition of an aqueous solution of ferric chloride initiated
the polymerization reaction; the molar ratio of pyrrole to ferric
chloride was kept as 1:1.5. After performing the polymerization reaction
for 4 h, the sediments were separated and thoroughly washed with DI
water. The curcumin template was removed completely by soaking in
anhydrous ethanol. The obtained PPy was rewashed with DI water and
vacuum-dried in an oven. The PPy nanoparticle sample prepared using
the curcumin template is denoted herein as “PPyNP”;
the control sample was prepared under the same conditions but without
the curcumin template is denoted as “PPyC”.
PPy Nanoparticle/f-CNT Composite
Film
To prepare the functionalized CNTs, raw CNTs were refluxed
at 90 °C using 3 M HNO3 for 14 h. The treated CNTs
were then sonicated for 1 h, filtered, washed, and dried to obtain
functionalized CNTs (f-CNTs).[65] The as-synthesized PPyNP and f-CNTs were
dispersed separately in ethanol through ultrasonication. The dispersions
were thoroughly mixed by magnetic stirring and ultrasonication before
filtered off to form the PPyNP/f-CNT composite film.
The PPyC/f-CNT composite film was prepared from PPyC
using the same method. The mass ratio of PPyNP and f-CNT was fixed at 8:2. The composite films prepared using PPyNP and
PPyC are denoted herein as “PPyNP/f-CNT”
and “PPyC/f-CNT”, respectively. The
thickness of the PPyNP/f-CNT nanocomposite film was
approximately 260 μm.
Characterization
A PerkinElmer Spectrum
RX1 FTIR spectrophotometer and pressed KBr pellets of the PPyNP/f-CNT composites were used to measure the Fourier transform
infrared (FTIR) spectra. Morphologies of the composite films were
analyzed using scanning electron microscopy (SEM, JSM 7401F). A drop
of the sample dispersed in ethanol was placed on a Cu grid to collect
the TEM images of dried samples using a transmission electron microscope
(TEM, JEOL JEM-1400). In addition, X-ray diffraction (XRD) images
were studied using a Rigaku RINT 2000 diffractometer (Tokyo, Japan).
Contact angle measurements were taken using a contact angle meter
(JC2000D1, Shanghai Zhongchen Digital Technic Apparatus, China). A
Micromeritics surface area and porosity analyzer (ASAP 2010, USA)
was used to record the nitrogen sorption isotherms of the composites.
The Brunauer–Emmett–Teller (BET) and Barret–Joyner–Halenda
(BJH) calculations were used to confirm the specific surface areas
and pore size distributions, respectively, of the composite films.
Electrochemical Characterization of the Freestanding
Composite Electrodes
A CHI6273D electrochemical workstation
(CH Instruments, USA) was used to perform the electrochemical tests
in a three-electrode system. A saturated calomel electrode (SCE),
a piece of the composite film, platinum foil, and aqueous 1 M H2SO4 were employed as the reference electrode, working
electrode, counter electrode, and electrolyte, respectively. CV and
GCD experiments were performed at different scan rates and applied
current densities, respectively, in the voltage range from −0.2
to +0.6 V. EIS tests were performed over the frequency range from
0.01 Hz to 100 kHz with an amplitude of 5 mV.The areal capacitances
(CA, F cm–2) were estimated
using eq from the GCD
curve[8]where ΔV represents the potential window (V)
excluding the IR drop, I is the discharge current
(A), Δt is the discharge time (s), and A is the area of
the freestanding electrode (cm2). The gravimetric capacitance
(Cm, F g–1) is calculated
by replacing A by m, the mass of
the freestanding electrode (g) in eq .The volumetric capacitance (CV, F cm–3) was calculated using eq (8)where d is
the thickness of the freestanding electrode (cm).
Electrochemical Performance Evaluation of
the Symmetric Supercapacitor Device
For the preparation of
the H2SO4/PVA-based gel electrolyte, PVA powder
(1 g) and H2SO4 (1 g) were dissolved in DI water
(10 mL) at 85 °C by magnetic stirring. The symmetric supercapacitor
cell was assembled using a filter paper and soaked in the H2SO4/PVA gel electrolyte and two pieces of PPyNP/f-CNT freestanding electrodes. The thickness of the assembled
device was nearly 0.083 cm. The electrochemical behavior of the symmetric
supercapacitor cell was characterized using a CHI6273D (CH Instruments,
USA) electrochemical work station. The areal (Ca–cell, F cm–2) and gravimetric capacitance
(Cm–cell, F g–1) of the cell was estimated according to eqs and 4(8)The energy (E, μW h cm–2) and power densities
(P, μW cm–2) of the supercapacitor
cell were estimated using eqs and 7(8)The volumetric capacitance (CV–cell, F cm–3) was estimated
using eq where ΔV represents the potential window (V)
excluding the IR drop, I is the discharge current
(A), Δt is the discharge time (s), d is the thickness of
the supercapacitor cell (cm), m is the mass of the
active material loaded on a single electrode, and A is the area of a single electrode (cm2).