-
2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim3918
www.advmat.dewww.MaterialsViews.com
wileyonlinelibrary.com
COMMUNICATION Transparent Flexible Graphene Triboelectric
Nanogenerators
Seongsu Kim , Manoj Kumar Gupta , Keun Young Lee , Ahrum Sohn ,
Tae Yun Kim , Kyung-Sik Shin , Dohwan Kim , Sung Kyun Kim , Kang
Hyuck Lee , Hyeon-Jin Shin , Dong-Wook Kim , and Sang-Woo Kim *
S. Kim, Dr. M. K. Gupta, K. Y. Lee, K.-S. Shin, S. K. Kim, K. H.
Lee, Prof. S.-W. Kim School of Advanced Materials Science and
Engineering Sungkyunkwan University (SKKU) Suwon 440746 , Republic
of Korea E-mail: [email protected] A. Sohn, Prof. D.-W. Kim School of
Department of Physics Ewha Womens University Seoul 120750 ,
Republic of Korea T. Y. Kim, D. Kim, Prof. S.-W. Kim SKKU Advanced
Institute of Nanotechnology (SAINT) Center for Human Interface
Nanotechnology (HINT) SKKU-Samsung Graphene Center Sungkyunkwan
University (SKKU) Suwon 440746 , Republic of Korea Dr. H.-J. Shin
Samsung Advanced Institute of Technology Yongin 446712 , Republic
of Korea
DOI: 10.1002/adma.201400172
the triboelectric effect have been proven to be a
cost-effective, powerful, and robust tool for harvesting mechanical
energy and their application as various self-powered nanosensors
including for pressure, magnetic, vibration, and mercury-ion
detection has been recently demonstrated. [ 19,2124 ] The
triboelectric effect depends on various factors, such as the
electron affi nity, work function, friction, chemical structure,
pressure, surface rough-ness, and humidity. [ 2528 ] A number of
theoretical studies on the electrostatic behavior of graphene have
been reported, and it has been concluded that graphene can store an
electric charge for a period of time, which adds to its suitability
for triboelectric nanogenerators (TNGs). [ 2932 ] In addition,
although graphene is assumed to be fl at its natural shape exhibits
many ripples because of inhomogeneous interactions with the
substrate, which causes stress-induced deformations in the form of
rip-ples along the stiff directions of the graphene lattice,
enhancing its roughness and friction. [ 33,34 ] Therefore, large
surface charges can be created through contact electrifi
cation/triboelectric effect for highly effi cient power generation.
[ 17,18 ]
In this study, we demonstrate electrical energy harvesting from
graphene by mechanical stressing. We fabricated gra-phene-based
TNGs (GTNGs) using large-scale graphene grown by chemical vapor
deposition (CVD) on copper (Cu) and nickel (Ni) foils. We designed
and fabricated fl exible transparent GTNGs by using monolayer (1L),
bilayer (2L), trilayer (3L), and quad-layer (4L) graphene using a
layer-by-layer transfer technique of 1L graphene grown on Cu foils.
Additionally, few-layer graphene samples with Bernal stacking
(rhombohe-dral stacking) grown on Ni foils were also utilized to
fabricate GTNGs. The dependence of the power output performance of
the GTNGs on the number of graphene layers is also discussed in
detail in terms of the work function and friction, which arises due
to different electronic relations between randomly and regularly
stacked graphene layers. This study provides a simple and
cost-effective means of harvesting electrical energy from various
types of mechanical energy sources in nature using GTNGs.
A polyethylene terephthalate (PET) polymer [ 35 ] was selected
for the development of a transparent fl exible 1L-GTNG because of
its high strength, high transparency, and light weight. To achieve
a high triboelectric effect, the TNGs should be fabri-cated using
two materials that have distinctly different tribo-electric
characteristics; one must readily lose electrons, whereas the other
must readily gain electrons. [ 17,27 ] Furthermore, to fully
utilize the other well-known properties of graphene, a 1L of
graphene was transferred onto the PET polymer, thus serving
Graphene, a two-dimensional (2D) aromatic monolayer of carbon
atoms arranged in a hexagonal and honeycomb lattice with an sp 2
atomic confi guration, has demonstrated exceptional physical
properties, including ultra-high electron mobility (as high as 26
000 cm 2 V 1 s 1 ), an excellent optical transparency of
approximately 97%, mechanical fl exibility, high mechanical
elasticity (with an elastic modulus of approximately 1 TPa), high
thermal stability, chemical inertness, and ballistic charge-carrier
transport. [ 13 ] Owing to its unique and exceptional properties,
graphene is considered to have high potential for technological
applications in many areas. The multifunctional properties of
graphene, such as its high transparency, con-ductivity, elasticity,
and impermeability, enable it to be used in fl exible electronics,
transparent protective coatings, and bar-rier fi lms. [ 47 ] These
fascinating properties make graphene an ideal material for
transparent, fl exible electrodes in solar cells, photodetectors,
nanogenerators, and light-emitting diodes (LEDs). [ 813 ] Although
the attention focused on graphene in recent years has been
accompanied by an increasing interest in 2D next-generation
electronics, its application has been limited to transparent
electrodes and catalysts. [ 1316 ] To the best of our knowledge,
graphene has not been used as an active material in
energy-harvesting devices and systems.
Recently, a new type of power-generating device that con-verts
mechanical energy into electricity using triboelectricity was
intensively studied. [ 1720 ] Further, nanogenerators based on
Adv. Mater. 2014, 26, 39183925
-
3919
www.advmat.dewww.MaterialsViews.com
wileyonlinelibrary.com 2014 WILEY-VCH Verlag GmbH & Co.
KGaA, Weinheim
COMMUNICATIO
N
as the top electrode for the GTNG. In the present case,
gra-phene plays a dual role for the GTNG; it functions as the top
electrode as well as the triboelectric layer at the bottom of the
GTNG. Before assembling the device, the graphene samples grown on
the Cu and Ni foils were characterized using Raman spectroscopy
(Details of the growth procedure of graphene and the Raman spectra
(Figures S1, S2) are given in the Supporting Information (Part 1
and 2)). The CVD-grown graphene sam-ples have wrinkles and ripples,
which make these graphene structures more suitable for high-output
voltage applications because of the signifi cant amounts of
friction and surface charge created by the triboelectric
effect.
Figure 1 a-d provides a schematic illustration of the
experi-mental procedures used to fabricate the transparent fl
exible 1L-GTNG. At fi rst, large-scale 1L graphene was grown on a
Cu foil using CVD (Figure 1 a). Next, the 1L graphene was
transferred onto a fl exible PET substrate using the well-known wet
transfer method (Figure 1 b). The two substrates, i.e., the PET/1L
graphene (bottom side) substrate and the PET/gra-phene (top
electrode) substrate, were then connected using a plastic spacer,
leaving a narrow 0.8 mm space between the 1L graphene and top PET
layers (Figure 1 c-d). The use of a spacer in the GTNG signifi
cantly improves the capacitance of the system in the deformation
process because of the presence of air voids between the PET
polymer and the graphene, which increases the strength of the
dipole moments formed during mechanical deformation. A schematic of
the fi nal structure of the GTNG device is shown in Figure 1 d. The
entire device fab-rication process is quite simple and novel, which
leads to an easy understanding of the charge-generation mechanism
and allows for a low-cost device fabrication process that is needed
for possible future commercialization. The well-known features of
graphene, such as its fl exibility, stretchability, and
compat-ibility with arbitrary substrates, are also shown in Figure
1 e-g.
To investigate the performance of the GTNGs, we car-ried out a
detailed electrical characterization of the device. This unique
structure allowed for the generation of an output voltage and
output current density from 1L-GTNG of 5 V and 0.5 A cm 2 ,
respectively, when a vertical compressive force of
1 kgf (1 kgf = 9.880665 N) was applied, as shown in Figures 2 a
and 2 b. The 1L-GTNG exhibited a very stable output voltage and
current under a cyclic compressive force. Switching polarity tests
were also carried out to confi rm that the measured output signals
were generated from the GTNG rather than from the measuring system.
As we reverse the polarity of the voltage and current meters, the
output signals are reversed, as shown in Figures S3 and S4 in the
Supporting Information.
To further examine the effect of the number of graphene layers
on the output performance, we fabricated 2L-GTNGs, 3L-GTNGs, and
4L-GTNGs ( non -AA/AB/ABC/AAA, stacking, i.e., random turbostratic
stacking). We obtained 2L-, 3L-, and 4L-graphene samples by
stacking 1L graphenes on PET sub-strates by using a wet transfer
technique that was subsequently integrated with the PET/graphene
for the fabrication of the TNG. The electrical power output signals
were measured under identical compressive forces for the 2L-, 3L-,
and 4L-GTNGs, and their corresponding data are shown in Figure
2a,b. The output voltage and output current were found to decrease
with an increasing number of graphene layers. Average output
voltage values of 3.0, 2.0, and 1.2 V and average output current
density values of 250, 160, and 100 nA cm 2 were observed for the
2L-, 3L-, and 4L-GTNGs, respectively. These studies confi rmed that
1L graphene is a good candidate for high-per-formance GTNGs and
that randomly stacked graphene layers exhibit a decreased output
performance.
Because multiple graphene layers were prepared on the PET
substrate using a wet transfer method, there are weak interlayer
interactions and random turbostratic stacking between the gra-phene
layers. Thus, regularly stacked (such as AA/AB, ABC, and ABA)
multilayer graphene grown on Ni foils by a CVD method was also
utilized for the fabrication of GTNGs. The electrical power outputs
of few-layer-based GTNGs were measured under identical mechanical
stress, and the output data are shown in Figure 2c,d.
Interestingly, the output voltage and output cur-rent density
dramatically increased to 9 V and 1.2 A cm 2 , respectively, under
a vertical mechanical force of 1 kgf. The observed output voltage
is nearly 1.8 times larger than that of the 1L-GTNG and nearly 7.5
times larger than the randomly
Figure 1. Schematic diagrams of device fabrication and
compatibility of graphene with an arbitrary substrate. a) Cu-foil
1L graphene grown by the CVD method is used. b) The 1L graphene is
transferred to the PET substrate. c) A plastic spacer is connected
to create an air gap. d) The spacer-incorporated 1L graphene is
integrated with the PET/graphene (top electrode) to fabricate the
GTNG. eg) The fl exibility, stretchability, and adjustability of
graphene with the crumpled substrate are shown.
Adv. Mater. 2014, 26, 39183925
-
3920
www.advmat.dewww.MaterialsViews.com
wileyonlinelibrary.com 2014 WILEY-VCH Verlag GmbH & Co.
KGaA, Weinheim
COMMUNICATION
stacked 4L-GTNG. A similar trend was also observed in the output
current, which was nearly twelve times larger than that of the
4L-GTNG (Figure 2 b,d). Further, to investigate the effective
electric power of GTNG, the output performance of the 1L-GTNG was
systematically studied at different loads. As shown in Figure S5a
in the Supporting Information the max-imum current decreases with
increasing resistance, whereas the output voltage shows the
opposite trend. The power density of 1L-GTNG was also plotted as a
function of external resist-ance and is shown in Figure S5b
(Supporting Information). The output power density increases at a
low resistance region and then decreases at a higher resistance
region. The maximum value of the power density of around 2.5 W cm 2
occurs at about 10 M. In addition, a durability test (over 1000
cycles) was also conducted to confi rm the mechanical durability of
the GTNG (Figure S6, Supporting Information).
The operating principle of the GTNG can be described using the
coupling of contact charging and the electrostatic effect under a
cycled compressive force. A COMSOL simulation was also carried out
to understand the working process of the GTNG (Supporting
Information, Part 3). The corresponding surface-charge distribution
and electric potential are shown in Figure 3 . According to the
work function values of the PET (Figure S7, Supporting Information)
and graphene, electrons are injected from the PET to the graphene,
resulting in the build-up of a net negative charge on the graphene
surface and a net positive charge on the PET surface. Furthermore,
because of the spacer placed between the graphene and the PET
surface, air voids are created, which result in the formation of
dipole moments. Therefore, an electric potential difference is
developed between
the two electrodes, that is, between the bottom graphene (the
active triboelectric material as well as the electrode) and the top
graphene (electrode), which results in an electric signal
gener-ated across the electrode.
Figure 3 demonstrates the working mechanism of the GTNGs at each
stage of the cyclic deformation. Initially, the device is neutral
in the absence of any pressure/force, and no charge is generated on
the surface of the PET and graphene; therefore, no electric
potential difference is established between the two electrodes
(Figure 3 a), and no output signal is observed. On the contrary,
when a vertical compressive force is applied to the top surface of
the device, the PET and 1L graphene layer are rubbed together.
Thus, triboelectric charges with opposite signs are generated
because of electron injection in the graphene by induced thermal
energy during the contact between the PET and graphene. These
charges are distributed on the contact sur-faces of PET and
graphene. As we discussed in the above sec-tion, because of the
differences in the work function of the PET and graphene, the
electrons are injected from PET to graphene, resulting in the
generation of negative charges on the graphene surface and positive
charges on the PET surface. At this stage, the generated surface
charges with opposite signs nearly coin-cide on the same plane,
generating an insignifi cant electrostatic potential difference
between PET and graphene (Figure 3 b). Therefore, no electrical
signal was detected at this stage.
When the pressure is released again, the PET fi lm reverts back
to its original position because of its own elasticity and fl
exibility. Once the PET and graphene surfaces are separated from
each other, the dipole moment becomes stronger, and a very strong
electric potential difference is created between the
Adv. Mater. 2014, 26, 39183925
Figure 2. Output performance of 1L, randomly stacked, and
regularly stacked GTNGs. a,b) Output voltage and current density
from a Cu foil-grown 1L GTNG and randomly stacked 2L-, 3L-, and
4L-GTNGs under a vertical compressive force of 1 kgf. c,d) Output
voltage and current density from regularly stacked, few-layer GTNGs
under a vertical compressive force of 1 kgf.
-
3921
www.advmat.dewww.MaterialsViews.com
wileyonlinelibrary.com 2014 WILEY-VCH Verlag GmbH & Co.
KGaA, Weinheim
COMMUNICATIO
N
two graphene electrodes. Therefore, to achieve equilibrium,
electrons start to fl ow from the negative potential side (bottom
graphene) to the positive potential side (top graphene), leading to
an accumulation of electrostatically induced charges on the
electrodes, resulting in a positive electrical signal (Figure 3 c).
No electrical signal is observed at equilibrium (Figure 3 d).
Furthermore, when an instantaneous vertical compression is applied
to the GTNG, the PET and graphene come into con-tact and short each
other out. The dipole moment subsequently disappears or decreases
in magnitude, and the electrostatic potential difference starts to
diminish. Therefore, the reduced electric potential difference
generates a fl ow of electrons from the top electrode side to the
bottom electrode side that causes the accumulated charges to
vanish, resulting in a negative elec-trical potential across the
electrodes (Figure 3 e). This negative electrical potential causes
electrons to be pumped back and forth between the two electrodes
because of contact charging. Therefore, the continuous application
and removal of a vertical compression on the GTNG drives a fl ow of
electrons between the top and bottom electrodes across the external
load via the triboelectric charge, which provides an alternating
current signal from the GTNG. To further confi rm our mechanism, we
fabricated a 1L-GTNG without a spacer and measured the electrical
output signal (Figure S8, Supporting Information). No signifi cant
output voltage was produced using any vertical mechanical strain.
These results demonstrate that our model is fairly valid in
explaining the working principle behind the GNTGs. The COMSOL
simulation results also confi rm the pro-posed mechanism.
The above study revealed that 1L graphene and regularly stacked
few-layer graphene are the best candidates for high-per-formance
GTNGs. We observed that the output performance decreases with an
increasing number of randomly stacked graphene layers, and the
performance increases when regu-larly stacked few-layer graphene is
used. Such enhancements in the output voltage and current observed
in regularly stacked few-layer graphene-based TNG over 1L- and
randomly stacked 2L-, 3L-, and 4L-based GTNGs are attributed to the
increased work function of few-layer graphene with Bernal stacking
and a strong electronic relation between regularly stacked graphene
layers. Raman spectra of the graphenes grown on the Cu and Ni foils
were taken to examine their stacking order (Figures S1 and S2,
Supporting Information).
Many researchers have proposed that the work function of
graphene varies with the number of graphene layers following an
increasing trend with the number of regularly stacked graphene
layers (i.e., graphene work function = 4.3, 4.4, 4.5, and 4.6 eV
for n = 1, 2, 3, and , respectively, where n is the number of
graphene layers). [ 29,30,36 ] Therefore, the difference in
work-function values can signifi cantly change the surface-charge
density on few-layer graphene due to the triboelectric effect when
rubbed with PET, which further increases the output voltage and
current in few-layer-based GTNGs relative to 1L graphene.
Mathematically, the contact potential difference (V) is given
as
V ep g ~ ( )/ , (1)
Adv. Mater. 2014, 26, 39183925
Figure 3. Power generation mechanism of GTNGs. a) Schematic
diagram for the initial state of the GTNG; the device was neutral
when no force was applied. be) Potential distribution of the GTNG
simulated by the COMSOL multi-physics software.
-
3922
www.advmat.dewww.MaterialsViews.com
wileyonlinelibrary.com 2014 WILEY-VCH Verlag GmbH & Co.
KGaA, Weinheim
COMMUNICATION
where p is the effective work function of the PET polymer, g is
the work function of the graphene, and e is the elementary charge.
[ 26,27,31 ] Therefore, an increasing work-function differ-ence
leads to an enhanced GTNG output power.
To calculate the precise value of the work function for 1L-,
2L-, 3L-, 4L graphene (Cu foil-grown), and few-layer graphene (Ni
foil-grown), a Kelvin probe force microscopy (KPFM) tech-nique was
used, as shown in Figure S7 in the Supporting Infor-mation. The
output results are shown in Figure 4 a; the work functions of
graphene were determined to be 4.92, 4.96, 5.04, 5.11, and 5.08 eV
for the Cu foil-grown 1L, 2L, 3L, 4L graphene samples and the Ni
foil-grown, few-layer graphene samples, respectively. These results
coincide with previous reports. [ 36,37 ] The above results confi
rm that the work function plays an important role in the output
performance of GTNGs. However, the observed discrepancy in the
variation of the output voltage/current density between 1L-GTNGs
and randomly oriented 2L-, 3L-, and 4L-GTNGs is discussed
below.
The linear reduction of the output power from randomly stacked
graphene-based TNGs caused by increasing the number of graphene
layers can be explained by the friction depending on the number of
graphene layers used, which arises because of the puckering effect
and electronphonon coupling effect. [ 38,39 ] The surface charge
and output potential is strongly related to the friction generated
between rubbed mate-rials (i.e., graphene and PET in the present
case). When 1L gra-phene grown on a Cu foil using CVD is
transferred onto a cer-tain template, it preserves its corrugated
surface leading to the appearance/enhancement of friction when
rubbed against other materials. Furthermore, the friction in
graphene is decreased
with an increasing number of graphene layers (i.e., the 1L
gra-phene reveals an approximately 20% higher amount of friction
than 2L graphene due to the puckering effect in graphene). [ 40 ]
The puckering is less prominent with an increasing number of
graphene layers because of the larger bending stiffness of the
graphene sheet, and therefore, the friction decreases for the 2L,
3L, and 4L graphene compared to 1L graphene. [ 3840 ]
Further, such variations in the amount of friction between 1L
graphene and 2L, 3L, and 4L graphene can also be attrib-uted to the
strong electronphonon coupling in the single-layer epitaxial
graphene and to the susceptibility of the graphene to out-of-plane
elastic deformation. [ 3843 ] Therefore, the amount of friction in
1L graphene is larger than that in 2L, 3L, and 4L gra-phene, which
results in a larger electrical power output from 1L-GTNGs. This
variation in the amount of friction in the gra-phene is also
observed in regularly stacked graphene layers [ 44 ] but to a
lesser extent. Therefore, the large output voltage and output
current density from the regularly stacked Ni-catalyzed, few-layer
GTNG are mainly related to its large work function.
Regardless, we still carefully investigated the output
vari-ation of the randomly stacked 4L-GTNG to visualize the
sur-face features of the PET to understand their effect on the
output performance. We assumed that in the case of ran-domly
stacked GTNG, a portion of the graphene is attached to the opposite
PET surface because of the weak adhesion/interaction between
randomly stacked graphene layers, which results in a low output
voltage/current density. Therefore, to observe the surface
morphology/features of the rubbed PET for the 4L-GTNG, we obtained
optical images and Raman spectra of the PET polymer before and
after the application
Adv. Mater. 2014, 26, 39183925
Figure 4. Work-function measurement and electric outputs. a)
Measured work function of 1L graphene and randomly and regularly
stacked 2L, 3L, and 4L graphene. b) Work function of pristine and
BV-doped graphene, and ZnO thin fi lm. c,d) Output voltage and
current density measured from the 1L pristine graphene-ZnO TNG and
the BV-doped graphene-ZnO TNG under a vertical compressive force of
1 kgf.
-
3923
www.advmat.dewww.MaterialsViews.com
wileyonlinelibrary.com 2014 WILEY-VCH Verlag GmbH & Co.
KGaA, Weinheim
COMMUNICATIO
N
of the mechanical stress on the randomly stacked 4L-GTNGs
(Figure S9a,b, Supporting Information). Raman peaks cor-responding
to graphene are also observed along with Raman peaks corresponding
to the PET in the case of the randomly stacked 4L-graphene.
Moreover, we have also taken friction force microscope (FFM) images
of pristine and rubbed PET (Figure S9c,d, Supporting Information).
The analysis revealed that because of weak interlayer interactions
between the ran-domly stacked graphene layers, the mechanically
transferred graphene is easily attached to the opposite PET
surface, which results in a signifi cantly decreased output
potential because of the reduced work-function difference between
the rubbed PET and the 4L graphene.
In addition, two controlled experiments were also carried out to
further confi rm the effect of the work function of the two rubbed
materials on the polarity of the induced charge and total
triboelectric output voltage/current. We fabricated two additional
TNGs based on pristine and 1,1-dibenzyl-4,4-bipyridinium dichloride
doped (BV-doped) 1L graphenes, and we utilized a ZnO thin fi lm
instead of PET because of the high work function of ZnO relative to
PET. The work func-tion of pristine graphene decreases signifi
cantly after doping it with BV. [ 45 ] Hall measurement data for
pristine and BV-doped 1L graphenes are given in Table S1 in the
Supporting Information. The work function values of pristine
graphene (4.92 eV), BV-doped graphene (4.59 eV), and the ZnO thin
fi lm (4.85 eV) were also measured by using KPFM measure-ments, as
shown in Figure 4 b. The output voltage and output current density
from pristine graphene/BV-doped graphene-ZnO-based TNGs are signifi
cantly lower than those of the graphene-PET-based TNG when
identical vertical compressive forces are applied (Figure 4 c,d).
The signal polarity is reversed for the BV-doped graphene-ZnO-based
TNG relative to the pristine graphene-ZnO-based TNG because of the
lower work function of BV-doped graphene relative to the ZnO thin
fi lm. These results further prove the importance of the work
function of graphene for high-performance GTNGs. Again, switching
polarity tests were also conducted to confi rm that the measured
output signals were generated from the graphene-ZnO-based TNG
rather than from the measuring system. The output signals were
reversed when we reversed the polarity of the voltage and current
meters. This result, along with the device structure, is shown in
Figure S10 in the Supporting Information.
To study a practical application of the GTNG, we drove small
electronic devices, such as a liquid crystal display (LCD), LED,
and electroluminescence (EL) display unit, using solely the output
power from a fl exible, few-layer graphene-based TNG. Initially, an
LCD screen with the Sungkyunkwan University logo was used for the
test, and it was directly connected to the output of the GTNGs
without any capacitor. A rectifi cation circuit was used to convert
the AC signal into a DC signal to power the LCD. Figure 5 presents
the images of the photos taken before and after the GTNGs were
activated (Figure 5 a). The LCD screen was activated when the
output power gen-erated by the GTNG exceeded the threshold voltage
of the LCD screen. The LCD screen turned on when the GTNG was
stressed vertically. We also manage to power white, blue, and green
LEDs by the GTNG equipped with a rectifi cation
circuit and capacitor (2.2 F). Commercial LEDs with white, blue,
and green emissions were used, as shown in Figure 5 b. When a
periodic mechanical force was applied vertically to the GTNG, the
total rectifi ed output power generated from the GTNG was suffi
cient to simultaneously activate all three LEDs. These results were
recorded, and their corresponding videos are shown in the
Supporting Videos 1 and 2 in the Supporting Information. Moreover,
we directly operated an EL display unit using the power generated
by the GTNG using a rectifi cation circuit, capacitor (22 F), and
inverter. Figure 5 c illustrates that the EL display activates when
a vertical compressive stress is applied to the GTNG. This is the
fi rst demonstration of a novel energy-harvesting application of
graphene using the triboelec-tric effect (See Supporting Video 3,
Supporting Information). The schematic circuit diagrams used for
operating the LCD, LED, and EL displays are shown in the Supporting
Information (Figure S11).
In conclusion, we have successfully demonstrated the
appli-cation of CVD-grown graphene as a transparent, fl exible TNG.
GTNGs based on 1L, 2L, 3L, 4L, and few-layer graphene grown on Cu
and Ni foils using CVD have been fabricated, and their output
voltage and output current density were measured under mechanical
strains. The 1L-GTNG exhibited a high
Adv. Mater. 2014, 26, 39183925
Figure 5. Driving a commercial LCD, LED, and EL displays using
the GTNG. The left panel presents the OFF state, and the right
panel presents the ON state of the LCD, LED, and EL display units.
a) A snapshot of the LCD, which was lit up and displaying
Sungkyunkwan University and the university's logo using the GTNG
under a periodic vertical compressive force. b) A captured image
showing the three LED arrays simultaneously lit up by the power
output generated from the GTNG. c) Commercial EL display unit
containing Sungkyunkwan University and the university's logo was
activated using the GTNG under a periodic vertical force.
-
3924
www.advmat.dewww.MaterialsViews.com
wileyonlinelibrary.com 2014 WILEY-VCH Verlag GmbH & Co.
KGaA, Weinheim
COMMUNICATION output voltage and output current density of 5 V
and 500 nA
cm 2 , respectively. Additionally, the output voltage and output
current density increased to 9 V and 1.2 A cm 2 , respectively, for
the regularly stacked few-layer GTNG. The variations in the
electrical power output of randomly stacked 1L-, 2L-, 3L-, and
4L-GTNGs and regularly stacked few-layer GTNGs were explained in
terms of their work function and friction. We were able to power an
LCD, LEDs, and an EL display using the elec-trical power output of
the GTNG without any other external energy source. This study
provides a simple and novel method for harvesting mechanical energy
using transparent and fl ex-ible GTNGs for powering low-power
portable devices and self-powered electronic systems.
Experimental Section Fabrication of GTNGs based on 1L graphene,
randomly and regularly
stacked graphene : For the device fabrication, 1L graphene was
coated with poly(methyl methacrylate) (PMMA) and immersed in an
etchant (Transene, type 1) to etch away the Cu foil. When the Cu
was completely etched away, the 1L graphene with PMMA was rinsed in
deionized water three times to wash away the etchant residues. The
large-scale monolayer graphene grown on the Cu fi lm by the thermal
CVD method was transferred onto the hard-coated PET (Higashiyama
Film Co., Ltd) substrate using the well-known wet transfer method.
Furthermore, to fabricate randomly stacked double, triple, and
quadruple layers of graphene, one, two and three monolayers of
graphene, each synthesized in an identical manner, were placed onto
the 1L graphene/PET substrate by the wet transfer method,
respectively. Next, the other PET/graphene (electrode) substrate
was connected by a thin plastic spacer to the PET/graphene (active
triboelectric material), leaving a narrow spacing between the
graphene surface and the top PET surface. The spacer was made of an
insulating polymer fi lm with double-sided adhesive with a
thickness of 0.8 mm and the area of each spacer was 2 mm 2 cm.
Graphene plays a dual role in the device: it works as the top
electrode and as the triboelectric material on the bottom and top
sides of the device. Regularly stacked GTNG fabrication was
conducted in an identical fashion.
Characterization and Measurements : Raman spectra and optical
images of the Cu-grown 1L graphene and Ni foil-grown, few-layer
graphene were examined using Raman spectroscopy (Renishaw, RM-1000
Invia, 514 nm, Ar+ ion laser). The friction force image between the
PET and graphene layers were measured using the friction force
microscopy mode (Park system, XE-100). The KPFMs were performed to
precisely determine the work function of the 1L and randomly and
regularly stacked few-layer graphene (Park system, XE-100). A
picoammeter (Keithley 6485) and oscilloscope (Tektronix DPO 3052)
were used to measure the low-noise output currents and voltages
generated by the device using a force stimulator (ZTEC ZPS
100).
Supporting Information Supporting Information is available from
the Wiley Online Library or from the author.
Acknowledgements S. Kim and M. K. Gupta contributed equally to
this work. This work was supported by the Global Frontier Research
Center for Advanced Soft Electronics (2013M3A6A5073177) and the
Basic Research Program (2012R1A2A1A01002787, 20090083540) of the
National Research
Foundation of Korea (NRF) grant funded by the Ministry of
Science, ICT & Future Planning (MSIP).
[1] A. A. Balandin , S. Ghosh , W. Bao , I. Calizo , D.
Teweldebrhan , F. Miao , C. N. Lau , Nano Lett. 2008 , 8 , 902
.
[2] V. Singh , D. Joung , L. Zhai , S. Das , S. I. Khondaker ,
S. Seal , Prog. Mater. Sci. 2011 , 56 , 1178 .
[3] A. V. Rozhkov , G. Giavaras , Y. P. Bliokh , V. Freilikher ,
F. Nori , Phys. Rep. 2011 , 503 , 77 .
[4] J. O. Hwang , J. S. Park , D. S. Choi , J. Y. Kim , S. H.
Lee , K. E. Lee , Y.-H. Kim , M. H. Song , S. Yoo , S. O. Kim , ACS
Nano 2012 , 6 , 159 .
[5] X. Wang , L. Zhi , K. Mullen , Nano Lett. 2008 , 8 , 323 .
[6] D. Choi , M.-Y. Choi , W. M. Choi , H.-J. Shin , H.-K. Park ,
J.-S. Seo ,
J. Park , S.-M. Yoon , S. J. Chae , Y. H. Lee , S.-W. Kim ,
J.-Y. Choi , S. Y. Lee , J. M. Kim , Adv. Mater. 2010 , 22 , 2187
.
[7] Y. Zhao , Y. Xie , Y. Y. Hui , L. Tang , W. Jie , Y. Jiang ,
L. Xu , S. P. Lau , Y. Chai , J. Mater. Chem. 2013 , 1 , 4956 .
[8] J. Wu , H. A. Becerril , Z. Bao , Z. Liu , Y. Chen , P.
Peumans , Appl. Phys. Lett. 2008 , 92 , 263 302 .
[9] K. Chung , C.-H. Lee , G.-C. Yi , Science 2010 , 330 , 655 .
[10] B. Kumar , K. Y. Lee , H.-K. Park , S. J. Chae , Y. H. Lee ,
S.-W. Kim , ACS
Nano 2011 , 5 , 4197 . [11] B. Y. Zhang , T. Liu , B. Meng , X.
Li , G. Liang , X. Hu , Q. J. Wang , Nat.
Commun. 2013 , 4 , 1811 . [12] G. Jo , M. Choe , C.-Y. Cho , J.
H. Kim , W. Park , S. Lee , W.-K. Hong ,
T.-W. Kim , S.-J. Park , B. H. Hong , Y. H. Kahng , T. Lee ,
Nanotechnology 2010 , 21 , 175 201 .
[13] J.-H. Lee , K. Y. Lee , B. Kumar , N. T. Tien , N.-E. Lee ,
S.-W. Kim , Energy Environ. Sci. 2013 , 6 , 169 .
[14] K. S. Kim , Y. Zhao , H. Jang , S. Y. Lee , J. M. Kim , K.
S. Kim , J.-H. Ahn , P. Kim , J.-Y. Choi , B. H. Hong , Nature 2009
, 457 , 706 .
[15] C. Huang , C. Li , G. Shi , Energy Environ. Sci. 2012 , 5 ,
8848 . [16] J. D. Roy-Mayhew , D. J. Bozym , C. Punckt , I. A.
Aksay , ACS Nano
2010 , 4 , 6203 . [17] G. Zhu , C. Pan , W. Guo , C.-Y. Chen ,
Y. Zhou , R. Yu , Z. L. Wang ,
Nano Lett. 2012 , 12 , 4960 . [18] S. Wang , L. Lin , Z. L. Wang
, Nano Lett. 2012 , 12 , 6339 . [19] F.-R. Fan , L. Lin , G. Zhu ,
W. Wu , R. Zhang , Z. L. Wang , Nano Lett.
2012 , 12 , 3109 . [20] P. Bai , G. Zhu , Z.-H. Lin , Q. Jing ,
J. Chen , G. Zhang , J. Ma ,
Z. L. Wang , ACS Nano 2013 , 7 , 2808 . [21] F.-R. Fan , Z.-Q.
Tian , Z. L. Wang , Nano Energy 2012 , 1 , 328 . [22] Y. Yang , L.
Lin , Y. Zhang , Q. Jing , T.-C. Hou , Z. L. Wang , ACS Nano
2012 , 6 , 10 378 . [23] J. Chen , G. Zhu , W. Yang , Q. Jing ,
P. Bai , Y. Yang , T.-C. Hou ,
Z. L. Wang , Adv. Mater. 2013 , 25 , 6094 . [24] Z.-H. Lin , G.
Zhu , Y. S. Zhou , Y. Yang , P. Bai , J. Chen , Z. L. Wang ,
Angew. Chem. Int. Ed. 2013 , 52 , 5065 . [25] A. F. Diaz , R. M.
Felix-Navarro , J. Electrostat. 2004 , 62 , 277 . [26] S. Matsusaka
, H. Maruyama , T. Matsuyama , M. Ghadiri , Chem.
Eng. Sci. 2010 , 65 , 5781 . [27] A. R. Akande , J. A. Adedoyin
, J. Electrostat. 2001 , 51 , 105 . [28] D. K. Davies , J. Appl.
Phys. D 1969 , 2 , 1533 . [29] V. Panchal , R. Pearce , R. Yakimova
, A. Tzalenchuk , O. Kazakova , Sci.
Rep. 2013 , 3 , 2597 . [30] J. Cazaux , Appl. Phys. Lett. 2011 ,
98 , 013 109 . [31] Z. Wang , R. Scharstein , Chem. Phys. Lett.
2010 , 489 , 229 . [32] C. Liu , Z. Yu , D. Neff , A. Zhamu , B. Z.
Jang , Nano Lett. 2010 , 10 ,
4863 .
Received: January 12, 2014 Revised: February 27, 2014
Published online: March 28, 2014
Adv. Mater. 2014, 26, 39183925
-
3925
www.advmat.dewww.MaterialsViews.com
wileyonlinelibrary.com 2014 WILEY-VCH Verlag GmbH & Co.
KGaA, Weinheim
COMMUNICATIO
N
[33] A. Fasolino , J. H. Los , M. I. Katsnelson , Nat. Mater.
2007 , 6 , 858 . [34] F. d. Juan , A. Cortijo , M. A. H. Vozmediano
, A. Cano , Nat. Phys.
2011 , 7 , 810 . [35] S. Fakirov , M. Evstatiev , Polymer 1990 ,
31 , 431 . [36] H. Hibino , H. Kageshima , M. Kotsugi , F. Maeda ,
F.-Z. Guo ,
Y. Watanabe , Phys. Rev. B 2009 , 79 , 125 437 . [37] T. Ohta ,
A. Bostwick , J. L. McChesney , T. Seyller , K. Horm ,
E. Rotenberg , Phys. Rev. Lett. 2007 , 98 , 206 802 . [38] S.
Kwon , J.-H. Ko , K.-J. Jeon , Y.-H. Kim , J. Y. Park , Nano Lett.
2012 ,
12 , 6043 . [39] T. Filleter , J. L. McChesney , A. Bostwick ,
E. Rotenberg , K. V. Emtsev ,
Th. Seyller , K. Horn , R. Bennewitz , Phys. Rev. Lett. 2009 ,
102 , 086 102 .
[40] C. Lee , Q. Li , W. Kalb , X.-Z. Liu , H. Berger , R. W.
Carpick , J. Hone , Science 2010 , 328 , 76 .
[41] Q. Li , C. Lee , R. W. Carpick , J. Horn , Phys. Status.
Solidi. B 2010 , 247 , 2909 .
[42] H. Washizu , S. Kajita , M. Tohyama , T. Ohmori , N.
Nishino , H. Teranishi , A. Suzuki , Faraday Discuss. 2012 , 156 ,
279 .
[43] H. Lee , N. Lee , Y. Seo , J. Eom , S. W. Lee ,
Nanotechnology 2009 , 20 , 325 701 .
[44] K.-S. Kim , H.-J. Lee , C. Lee , S.-K. Lee , H. Jang ,
J.-H. Ahn , J.-H. Kim , H.-J. Lee , ACS Nano 2011 , 5 , 5107 .
[45] H.-J. Shin , W. M. Choi , D. Choi , G. H. Han , S.-M. Yoon
, H.-K. Park , S.-W. Kim , Y. W. Jin , S. Y. Lee , J. M. Kim ,
J.-Y. Choi , Y. H. Lee , J. Am. Chem. Soc. 2010 , 132 , 15 603
.
Adv. Mater. 2014, 26, 39183925