In this paper, we demonstrate a hybrid generator, derived from the concurrent adoption of piezoelectric and triboelectric mechanisms in one press-and-release cycle, called a Hybridized Self-Powered sensor (HSPS). A new integration of print circuit board (PCB) technology-based piezoelectric generator (PG) concurrently adopted the direct-write, near-field electrospun polyvinylidene fluoride (PVDF) nano/micro-fibers as piezoelectric source materials. On the other hand, triboelectric nanogenerators have the advantages of a high output performance with a simple structure which is also concurrently combined with the PG. The working mechanism of the HSPS includes the PCB-based substrate mounted with parallel aligned piezoelectric PVDF fibers in planar configuration which first bended and generated the electric potential via the effect of piezoelectricity. In what follows, the deformation of a cylindrical rolled-up piezoelectric structure is exercised, and finally, the triboelectric contact of Cu and PTFE layers is physically rubbed against each other with a separation to induce the triboelectric potential. This hybridized generator with a double domed shape design simultaneously combines piezoelectric output and triboelectric output and offers a built-in spacer with automatically spring back capability, which produces a peak output voltage of 100 V, a current of 4 μA, and a maximum power output of 450 nW. A self-powered smart window system was experimentally driven through finger-induced strain of HSPS, showing the optical properties with reversibly tunable transmittances. This research is a substantial advancement in the field of piezoelectric PVDF fibers integration toward the practical application of the whole self-powered system.

Significantly concentrative research on renewable or green energy has been actively performed to mitigate the ever-increasing energy issues.1–5 One of the missing pieces for the green energy lies in the massively wasted mechanical energy, primarily the bio-mechanical energy ubiquitously available in the modern civilization. Fundamentally there are several basic principles for the mechanical-to-electric energy conversion mechanisms such as electrowetting,6 electromagnetics,7,8 magnetostriction,9 piezoelectricity,10–14 and triboelectricity.15–20 The first piezoelectric nanogenerator (NG) was previously proposed with the one-dimensional zinc oxide (ZnO) nanowires (NWs).21 In particular, polyvinylidene fluoride (PVDF) is a polymeric material with electromechanical coupling of a piezoelectric constant (d33 ∼ 57.6 pm/V).22 A typical method for making the PVDF nano/micro-fibers (NMFs) was reported via the near-field electrospinning technique (NFES).23In situ poled and highly aligned NMFs can be fabricated directly on the flexible substrates without post-poling treatments.24 Due to the advantageous characteristics, PVDF piezoelectric NMFs have been previously demonstrated as human motion sensors.25 On the other hand, a variety of energy harvesting opportunities for human-induced motions including walking step, vocal cord vibration, heart beating, and muscle movement had been widely investigated.26–29 Other forms of energy scavenging mechanism which rekindles intensive interest are adopting the principle of triboelectricity, particularly in the fields of nanotechnology. The so-called triboelectric nanogenerators (TGs) have been tremendously developed in recent years as self-powered, sustainable, and green sources for personal devices.30–32 However, several limitations pose the severe challenges for the self-powered electronics such as considerably low output power, long-time structural endurability, unfavorable massive production, and the adaptability for small mechanical force.33 The recent review of the current progress in the various types of hybrid energy harvesters was extensively summarized that the hybridization of energy sources can be ranging from the piezoelectric, triboelectric, pyroelectric, thermoelectric, and photovoltaic effects.34 

In order to overcome a limitation of the PVDF piezoelectric generator (PG) or TG such as a low output power,35–39 we demonstrate a prototype of the piezoelectric/triboelectric hybridized generator that can produce high output power due to the cooperative operation of piezoelectric and triboelectric mechanisms in a single press-and-release cycle. Both planar and fully rolled-up piezoelectric structures are adopted and integrated with the TG. Moreover, our triboelectric generator has a facile microstructured surface modification via deionied (DI) water etching at elevated temperature on an aluminum electrode. This makes our hybridized generator exhibit a reliable lifetime and easily scale-up fabrication process. Using a full-wave bridge rectifier, the peak output voltage, current, and power of 100 V, 4 μA, and 450 nW can be successfully obtained, respectively. The electrical energy output of the hybridized generator successfully changes the optical transmittance of smart window at a finger-induced strain. On the other hand, the hybridized generator can be fixed to an insole to differentially detect various human motions of treading, walk, run, and jump, respectively.

Figure 1(a) illustrates a schematic diagram of a fabricated Hybridized Self-Powered sensor (HSPS), where the planar print circuit board (PCB)-based piezoelectric generator was fabricated on the top, and two fully rolled-up structures were fabricated in the middle of thermal plastic polymers (TPE, 3DMART) which are 3D-printed and flexible hollow structures. On the other hand, the copper (Cu)–polytetrafluoroethene (PTFE) triboelectric generator consists of a Cu film, a PTFE film, and an Al electrode at the bottom. Figures 1(b) and 1(c) display the working mechanism of the HSPS, showing the sequentially actuated deformations based on the principles of piezoelectricity and triboelectricity. Under a finger-induced strain on the in situ poled PVDF NMFs, the PCB-based piezoelectric generator can produce output voltage/current signals due to the change in polarization intensity, where the “±” signs indicate the polarity of the local piezoelectric potential created in situ inside the NMFs as depicted in Figures 1(b) and 1(c). For the Cu-PTFE triboelectric generator as shown in Figure 1(d), the triboelectric difference in polarities between the surfaces of the Cu film and PTFE film will induce the electrons transferring process, as the separation between the Cu film and the PTFE film occurs and the output current/voltages can be generated. Both the triboelectric charges and the electrostatic induction are contributing to the electron flow process. Furthermore, the structure size of the TPE 3D-printed hollow structure is illustrated in detail in Figure S1 of the supplementary material, and the large-scale manufacturing process and the structure of the PCB technology are illustrated in Figure S2 of the supplementary material. The reason for designing the 3D-printed hollow structure aims to maximize the power output by simultaneously integrating the piezoelectric layers with planar and rolled-up substrates and the triboelectric rubbing layers with self-regulating spacers.

FIG. 1.

(a) Structural design of the Hybridized Self-Powered sensor (HSPS) in the left column, including the cross-sectional view with multiple layer structures. The right column shows the actual fabricated device. The orange dotted inset in the right shows the fabricated HSPS. ((b) and (c)) The scheme of a working mechanism of the PCB-based piezoelectric generator based on nano/micro fibers (NMFs). Multiple layers of piezoelectric NMFs with both planar and fully rolled-up structures (∼500 NMFs for each configuration functionally serve as the spacer). The orange dotted inset in the right shows the PCB-based PG. (d) The scheme of a working mechanism for the Cu-PTFE triboelectric generator. The orange dotted inset in the right shows the Cu-PTFE TG.

FIG. 1.

(a) Structural design of the Hybridized Self-Powered sensor (HSPS) in the left column, including the cross-sectional view with multiple layer structures. The right column shows the actual fabricated device. The orange dotted inset in the right shows the fabricated HSPS. ((b) and (c)) The scheme of a working mechanism of the PCB-based piezoelectric generator based on nano/micro fibers (NMFs). Multiple layers of piezoelectric NMFs with both planar and fully rolled-up structures (∼500 NMFs for each configuration functionally serve as the spacer). The orange dotted inset in the right shows the PCB-based PG. (d) The scheme of a working mechanism for the Cu-PTFE triboelectric generator. The orange dotted inset in the right shows the Cu-PTFE TG.

Close modal

The detail of the fabrication processes for a proposed print circuit board (PCB) based piezoelectric generator is shown in Figure 2(a). Experimentally, the electrospinning process parameters used in this case are 16 wt. % PVDF, solvent (DMF: acetone with 1:1 weight ratio), and 4 wt. % fluoro-surfactant (Capstone FS-66); 3D structural integrity can be ensured by properly selecting the balance between the electrostatic, capillary, and evaporative forces. The general NFES parameters adopted in this paper are the following: the applied voltage at 1.5 kV, a motion speed of 50 mm/s, and an initial spinneret-to-collector distance of 1.5 mm. NFES has been demonstrated previously to controllably deposit very delicate pattern fibers. The final packaging step is utilized PDMS to fully encapsulate and isolate from the environmental disturbances. Integrating PCB technology with the NFES NMFs will be beneficial to the massively produced technique at the very affordable cost. Moreover, the PCB based approach will be inherently suitable to the large-scale application of a piezoelectric generator due to the benign nature of economically affordable, industrially compatible, and structurally robust. The typical physical construction of a piezoelectric generator consists of ∼500 polymeric PVDF NMFs which were directly deposited on a PCB substrate with dimension 40 000 μm × 25 000 μm × 75 μm. The intervals of an Au plated electrode were pre-designed at 0.15 mm for the consideration of the best electrical output. Due to the improved continuous deposition in a massive scale, some extremely long piezoelectric NMFs as long as 4 cm can only be directly deposited, in situ poled, and simultaneously assembled via the NFES technique. Finally, the fully encapsulated PDMS layer is cured for the packaging layer. Due to the use of all polymer layers in our PCB PG, the device could be easily deformed for the generation of electricity. Figures 2(b) and 2(c) are the as-spun PVDF NMFs which have diameters ranging from 500 nm to 3 μm as shown in the OM and SEM photos. Furthermore, in order to verify that the electrical outputs were generated from the piezoelectric properties of PVDF NMFs and exclusion of the triboelectric effect, we performed the validated experiments as shown in Figure S3 of the supplementary material. Another aspect for the integration should include the high electrical output of the triboelectric generator (TG); the surface area can be increased effectively through the wet etching technique. Figure 2(d) shows the process of the Al electrode which can be etched into microstructures (or equivalently, to increase the effective surface area of the Al electrode for the TG) by immersing the Al electrode in hot deionized (DI) water under different temperatures in the range of 27–170 °C for 15 min. Figure 2(d) also presents Ra of the etched Al electrode, which is unevenly covered by microstructures with Ra from 0.181 to 0.336 μm. OM images of the etched Al electrodes and the output voltage of different temperatures etching were presented in Figure S4 of the supplementary material, showing the effectiveness of increased electrical output by the increasing surface area. Furthermore, the effect of operating frequency on the triboelectric generator can be measured and plotted in Figure S5 of the supplementary material, showing that the proportional increase of output voltage can be closely related to the operating frequency.

FIG. 2.

(a) Schematic diagram of the PCB-based piezoelectric generator manufacturing process with both configurations of planar and rolled-up structures. (b) Top-view of an optical microscope (OM) photograph of NFES PVDF NMFs. (c) A scanning electron microscope (SEM) photograph of a single PVDF fiber. (d) Topology of an Al electrode after DI water etching under different temperatures in the range of 27–170 °C. Stylus measured roughness (Ra) as a function of etching temperature.

FIG. 2.

(a) Schematic diagram of the PCB-based piezoelectric generator manufacturing process with both configurations of planar and rolled-up structures. (b) Top-view of an optical microscope (OM) photograph of NFES PVDF NMFs. (c) A scanning electron microscope (SEM) photograph of a single PVDF fiber. (d) Topology of an Al electrode after DI water etching under different temperatures in the range of 27–170 °C. Stylus measured roughness (Ra) as a function of etching temperature.

Close modal

Figure 3(a) depicts the measured output voltage and output current signals of the triboelectric generator under a finger-induced strain, indicating that the output voltage is about 85 V and the output current is about 3 μA at 2.5 Hz actuation with a maximum strain of 0.2. For the triboelectric generator, the output current experimentally decreased with incrementally increasing loading resistance, while the output voltage reacted on the completely opposite trend as depicted in Figure 3(c). The corresponding output power of the triboelectric generator increases in the resistance region from 1 to 20 MΩ and reaches a climax, i.e., it decreases accordingly under a larger loading resistance from 20 to 40 MΩ. The output power was calculated by V2/R, where V is the output voltage and R is the loading resistance. The largest output power of the triboelectric generator is calculated as 312 nW under a loading resistance of 20 MΩ, as displayed in Figure 3(e). Figure 3(b) presents the output voltage and current of the piezoelectric generator under the same finger-induced strain, indicating that the output voltage and current peaks are about 12 V and 1 μA, respectively. As illustrated in Figure 3(d), the largest output power of the piezoelectric generator is about 120 nW under a loading resistance of 10 MΩ as displayed in Figure 3(f).

FIG. 3.

((a) and (b)) Individually measured output voltage and current signals of the triboelectric generator/piezoelectric generator. ((c) and (d)) Resistance matching measurement for the output voltage and current under the different loading resistances from 1 kΩ to 40 MΩ. ((e) and (f)) Generated output powers as a function of loading resistances from 1 kΩ to 40 MΩ.

FIG. 3.

((a) and (b)) Individually measured output voltage and current signals of the triboelectric generator/piezoelectric generator. ((c) and (d)) Resistance matching measurement for the output voltage and current under the different loading resistances from 1 kΩ to 40 MΩ. ((e) and (f)) Generated output powers as a function of loading resistances from 1 kΩ to 40 MΩ.

Close modal

A fully HSPS with the piezoelectric generator (two rolled-up cylindrical layers and one planar layer of piezoelectric NMFs) and one pair of Cu-PTFE triboelectric generator can hybridize and harvest piezoelectric/triboelectric power simultaneously. Figure 4(a) shows the schematic diagram and the working condition of HSPS. When the HSPS is subject to the downward deformation of finger-induced strain, the corresponding piezo-potential fields will be generated along the NMFs in both rolled-up and planar configurations, while the corresponding tribo-potential fields are generated by the physical contact of Cu-PTFE interfaces. Similarly for the resistance matching experiments as shown in Figure 4(b), the corresponding output power of the hybridized generator increases in the resistance region from 1 kΩ to 10 MΩ and then decreases under a larger loading resistance from 10 to 40 MΩ. In addition, as depicted in Figure S6 of the supplementary material, the output current dropped with increasing loading resistance, while the output voltage increased with increasing loading resistance. Figures 4(c) and 4(d) presents the measured output voltage/current signals of ∼100 V/4 μA at 2.5 Hz. Above deformations demonstrated the flexibility of HSPS which could be operated under a finger-induced strain at 2.5 Hz and the mechanical durability as well as stability of the HSPS is confirmed as shown in Figure S7 of the supplementary material. The tested results show that when both the piezoelectric generator and the triboelectric generator are working simultaneously, the HSPS has an improved and superimposed electrical performance better than that of the individually operated piezoelectric generator or triboelectric generator. In summary, a HSPS with fundamentally different energy harvesting functions of piezoelectric and triboelectric effects is demonstrated experimentally. Furthermore, the numerical simulation for HSPS under the progression of compression and deformation is performed using the commercially available large plastic finite element software DEFORM 3D for simulating the large substrate deformation, and the seven steps of strain were presented in Figure S8 of the supplementary material.

FIG. 4.

(a) A fully HSPS with the piezoelectric generator (two rolled-up cylindrical layers and one planar layer of piezoelectric NMFs) and one pair of Cu-PTFE triboelectric generator. (b) Corresponding measured output powers of the HSPS under the different loading resistances from 1 kΩ to 40 MΩ. ((c) and (d)) The output voltage and current were measured as ∼100 V/4 μA for the insulation glove against finger-induced strain.

FIG. 4.

(a) A fully HSPS with the piezoelectric generator (two rolled-up cylindrical layers and one planar layer of piezoelectric NMFs) and one pair of Cu-PTFE triboelectric generator. (b) Corresponding measured output powers of the HSPS under the different loading resistances from 1 kΩ to 40 MΩ. ((c) and (d)) The output voltage and current were measured as ∼100 V/4 μA for the insulation glove against finger-induced strain.

Close modal

To detect and in situ monitor the bio-mechanical human motions in a self-powered manner, the stretchable HSPS can be directly placed under insole in Figure 5(a). As shown in Figure 5(b) for the integration of HSPS, the device could discernibly detect and discriminate the feet tread from different heights (5, 10, 15 cm). Output voltage under various height treadings of 5 cm, 10 cm, and 15 cm was measured as ∼1 V/1.5 V/1.9 V, respectively. Furthermore, in Figure 5(c), the output current of various human motions including walking, running, and jumping could be measured as ∼0.3 μA/0.9 μA/1.9 μA, respectively. Similarly for the output voltage under various heights, human motions were collectively measured and presented in Figure S9 of the supplementary material, indicating the great potential for the proposed self-powered device. In contrast to conventional rotary encoders with restricted direction of motion, the insole-integrated device is repeatable and beneficial for human friendly rehabilitation. In comparison, the most recent publication of the triboelectric generator based on the ultrathin flexible single-electrode design can exhibit an output current of 1 μA simply by tapping by a bare finger. Moreover, the device was also applied to in-sole application and the results showed a stable peak-to-peak current of ∼3 μA.40 The presented HSPS also has a comparable peak-to-peak output current of ∼3 μA by converting the similar bio-mechanical motion of an in-sole device. Another latest and robust design of unearthed single-electrode TENG was proposed and under an acceleration of 3 m s−2 with a linear motor driven external force of 15 N by, the output current reaches a peak-to-peak output current of ∼6–7 μA.41 Though the result is slightly superior to the current device, therefore, the three dimensionally hybridized structures may be integrated into current design in the future. Furthermore, the HSPS-attached shoe insole exhibits excellent flexibility and conformability for the integrated piezoelectric/triboelectric hybridization of self-powered sensors.

FIG. 5.

Photographs of HSPS fixed to (a) insole. Relative changes in current versus time for various bio-mechanical motions of (b) treading and (c) walk, run, and jump, respectively.

FIG. 5.

Photographs of HSPS fixed to (a) insole. Relative changes in current versus time for various bio-mechanical motions of (b) treading and (c) walk, run, and jump, respectively.

Close modal

Figure 6(a) shows the hybrid schematic of both the HSPS and the smart window (Tintable Kibing Co., Ltd.), where a full bridge rectification circuit was connected with the HSPS for an alternating voltage to direct voltage conversion. Furthermore, detail information regarding the smart window specification such as the chemical composition, switching time, and others is summarized in Table I of the supplementary material. In Figure 6(b), we used a different Cu-PTFE friction area to control an output voltage of HSPS. There were three kinds of HSPS, and the blue, orange, and yellow lines indicated three different Cu-PTFE friction areas of 2 × 4 cm2, 3 × 4 cm2, and 4 × 4 cm2,, respectively. The output voltages from three kinds of HSPS were measured as ∼43 V/82 V/100 V under finger-induced strain at 4 Hz. In addition, performance measurement of the HSPS output voltage and current were measured to be about ∼50 V/∼1.5 μA, ∼72 V/∼2.7 μA, and ∼90 V/∼3.5 μA, corresponding to 2 × 4 cm2, 3 × 4 cm2, and 4 × 4 cm2 contact area, respectively, as detailed in Figure S10 of the supplementary material. Figure 6(c) shows the transmittance change of the HSPS driven by a finger-induced strain. Significant transmittance differences can be monitored by Micro Spectrometers-SE1020, and photographs of the smart window in response to different HSPS are also presented as insets in Figure 6(c), showing the color switching of the smart window between gray and black driven by finger-induced mechanical motions, and the smart window transmittance is measured as ∼35.4%, 23.6%, and 18.3%, respectively, for three kinds of HSPS with different frictional areas. Figure 6(d) shows the transmittance change of the smart window at a certain wavelength (550 nm) as driven by the HSPS. For example, the switch time of the change of transmittance from ∼50.2% to 18.3% for the sample of 4 × 4 cm contact area can be measured as 0.45 s. These results provide solid evidence to support that the smart window can be successfully integrated by the proposed HSPS device.

FIG. 6.

(a) Schematic diagram of the equivalent circuit with HSPS and smart window. (b) Use different Cu-PTFE friction areas to control HSPS output voltage. (c) Photographs of the color change of the smart window from transparent to opaque at an applied voltage from three kinds of HSPS. The gray, blue, orange, and yellow lines represent the smart window without HSPS, HSPS with Cu-PTFE contact area 2 × 4 cm2, 3 × 4 cm2, 4 × 4 cm2, respectively. (d) Transmittance change of the smart window in response to different HSPS (blue: Cu-PTFE contact area 2 × 4 cm2, orange: Cu-PTFE contact area 3 × 4 cm2, yellow: Cu-PTFE contact area 4 × 4 cm2) at 550 nm (measuring instrument: Micro Spectrometers-SE1020).

FIG. 6.

(a) Schematic diagram of the equivalent circuit with HSPS and smart window. (b) Use different Cu-PTFE friction areas to control HSPS output voltage. (c) Photographs of the color change of the smart window from transparent to opaque at an applied voltage from three kinds of HSPS. The gray, blue, orange, and yellow lines represent the smart window without HSPS, HSPS with Cu-PTFE contact area 2 × 4 cm2, 3 × 4 cm2, 4 × 4 cm2, respectively. (d) Transmittance change of the smart window in response to different HSPS (blue: Cu-PTFE contact area 2 × 4 cm2, orange: Cu-PTFE contact area 3 × 4 cm2, yellow: Cu-PTFE contact area 4 × 4 cm2) at 550 nm (measuring instrument: Micro Spectrometers-SE1020).

Close modal

The paper proposed a double domed shape structural design which aims to simultaneously harvest the piezoelectric and triboelectric generations at various surfaces while at the same time offers a built-in spacer with automatically spring back capability. The HSPS is composed of piezoelectric structures of a planar and two fully rolled-up PCB-based PG as well as triboelectrically rubbed Cu-PTFE substrates. The PCB-based PG including PVDF NMFs, Au electrode, and PDMS for the packaging layer provides piezoelectricity. On the other hand, the Cu-PTFE TG including Cu tape, PTFE tape, and Al electrode provides triboelectricity. The HSPS is capable of individually/simultaneously converting bio-mechanical energy to electricity. As compared with the PG, the TG has a much larger output voltage and current. When both the PG and the TG are working simultaneously, the HSPS has a much better electrical performance than that of the individual energy harvesting unit (PG or TG). The HSPS can scavenge and monitor bio-mechanical force in various forms such as foot treading and human motions (walking, running, and jumping) to deliver a distinctly different output voltage and current in a self-powered manner. Furthermore, smart window application demonstrated that reversible electrochromic reactions can be driven directly by the HSPS, and a minimum transmittance change of 18.3% at a certain wavelength (550 nm) as well as visible color change can be driven by simply finger-induced strain. Substantial improvement in the practically promising applications of hybridized and self-powered systems is demonstrated and future advancement in self-powered wearable electronics and flexible displays can be expected.

See supplementary material for the details of structure, manufacturing processes (design concept, fabrication steps, etching effect on surface roughness, and deformation simulation using the finite element method), and electrical (resistance matching, long term stability) properties of HSPS as well as the specification of the used smart window.

1.
G. J.
Aubrecht
,
Energy: Physical, Environmental, and Social Impact,
3rd ed. (
Pearson Education
,
London
,
2006
), Chap. I, pp.
2
15
.
2.
Z. L.
Wang
,
Adv. Mater.
24
,
280
285
(
2011
).
3.
J.
Paradiso
and
T.
Starner
,
IEEE Pervasive Comput.
4
,
18
27
(
2005
).
4.
Z. L.
Wang
and
J.
Song
,
Science
312
,
242
246
(
2006
).
5.
S. P.
Beeby
,
R. N.
Torah
,
M. J.
Tudor
,
P.
Glynne-Jones
,
T.
O’Donnell
,
C. R.
Saha
, and
S.
Roy
,
J. Micromech. Microeng.
17
,
1257
1265
(
2007
).
6.
T.
Krupenkin
and
J. A.
Taylor
,
Nat. Commun.
2
,
448
(
2012
).
7.
S.
Roundy
and
E.
Takahashi
,
Sens. Actuators, A
195
,
98
104
(
2013
).
8.
Y.
Chen
,
T. E.
Pollock
, and
A.
Salehian
,
Smart Mater. Struct.
22
,
094027
(
2013
).
9.
T.
Lafont
,
L.
Gimeno
,
J.
Delamare
,
G. A.
Lebedev
,
D. I.
Zakharov
,
B.
Viala
,
O.
Cugat
,
N.
Galopin
,
L.
Garbuio
, and
O.
Geoffroy
,
J. Micromech. Microeng.
22
,
094009
(
2012
).
10.
M. Y.
Choi
,
D.
Choi
,
M. J.
Jin
,
I.
Kim
,
S. H.
Kim
,
J. Y.
Choi
,
S. Y.
Lee
,
J. M.
Kim
, and
S. W.
Kim
,
Adv. Mater.
21
,
2185
(
2009
).
11.
H.
Gullapalli
,
V. S. M.
Vemuru
,
A.
Kumar
,
A.
Botello-Mendez
,
R.
Vajtai
,
M.
Terrones
,
S.
Nagarajaiah
, and
P. M.
Ajayan
,
Small
6
,
1641
(
2010
).
12.
J.
Han
,
F.
Fan
,
C.
Xu
,
S.
Lin
,
M.
Wei
,
X.
Duan
, and
Z. L.
Wang
,
Nanotechnology
21
,
405203
(
2010
).
13.
S.
Xu
,
B. J.
Hansen
, and
Z. L.
Wang
,
Nat. Commun.
1
,
93
(
2010
).
14.
Y.
Qi
,
N. T.
Jafferis
,
K.
Lyons
, Jr.
,
C. M.
Lee
,
H.
Ahmad
, and
M. C.
McAlpine
,
Nano Lett.
10
,
524
(
2010
).
15.
L.
Lin
,
S.
Wang
,
Y.
Xie
,
Q.
Jing
,
S.
Niu
,
Y.
Hu
, and
Z. L.
Wang
,
Nano Lett.
13
,
2916
2923
(
2013
).
16.
B.
Hwang
,
J.
Lee
,
T. Q.
Trung
,
E.
Roh
,
D.
Kim
,
S. W.
Kim
, and
N.
Lee
,
ACS Nano
9
,
8801
8810
(
2015
).
17.
Q.
Liang
,
X.
Yan
,
Y.
Gu
,
K.
Zhang
,
M.
Liang
,
S.
Lu
,
X.
Zheng
, and
Y.
Zhang
,
Sci. Rep.
5
,
9080
(
2015
).
18.
J. H.
Lee
,
K. Y.
Lee
,
M. K.
Gupta
,
T. Y.
Kim
,
D. Y.
Lee
,
J.
Oh
,
C.
Ryu
,
W. J.
Yoo
,
C. Y.
Kang
,
S. J.
Yoon
,
J. B.
Yoo
, and
S. W.
Kim
,
Adv. Mater.
26
,
765
(
2014
).
19.
Y.
Zi
,
L.
Lin
,
J.
Wang
,
S.
Wang
,
J.
Chen
,
X.
Fan
,
P. K.
Yang
,
F.
Yi
, and
Z. L.
Wang
,
Adv. Mater.
27
,
2340
(
2015
).
20.
P.
Bai
,
G.
Zhu
,
Y.
Liu
,
J.
Chen
,
Q.
Jing
,
W.
Yang
,
J.
Ma
,
G.
Zhang
, and
Z. L.
Wang
,
ACS Nano
7
,
6361
6366
(
2013
).
21.
22.
C.
Chang
,
V. H.
Tran
,
J.
Wang
,
Y. K.
Fuh
, and
L.
Lin
,
Nano Lett.
10
,
726
(
2010
).
23.
Y. K.
Fuh
,
P. C.
Chen
,
H. C.
Ho
,
Z. M.
Huang
, and
S. C.
Li
,
RCS Adv.
5
,
67787
67794
(
2015
).
24.
Y. K.
Fuh
,
S. Y.
Chen
, and
J. C.
Ye
,
Appl. Phys. Lett.
103
,
033114
(
2013
).
25.
Y. K.
Fuh
,
J. C.
Ye
,
P. C.
Chen
,
H. C.
Ho
, and
Z. M.
Huang
,
ACS Appl. Mater. Interfaces
7
,
16923
16931
(
2015
).
26.
S.
Sato
,
K.
Yamada
, and
N.
Inagaki
,
Med. Biol. Eng. Comput.
44
,
353
(
2006
).
27.
M.
Kanik
,
M. G.
Say
,
B.
Daglar
,
A. F.
Yavuz
,
M. H.
Dolas
,
M. M.
El-Ashry
, and
M.
Bayindir
,
Adv. Mater.
27
,
2367
(
2015
).
28.
T.
Huang
,
C.
Wang
,
H.
Yu
,
H.
Wang
,
Q.
Zhang
, and
M.
Zhu
,
Nano Energy
14
,
226
235
(
2015
).
29.
Z.
Li
and
Z. L.
Wang
,
Adv. Mater.
23
,
84
(
2010
).
30.
Y.
Yang
,
H.
Zhang
,
J.
Chen
,
Q.
Jing
,
Y. S.
Zhou
,
X.
Wen
, and
Z. L.
Wang
,
ACS Nano
7
,
7342
7351
(
2013
).
31.
Y.
Yang
,
H.
Zhang
,
Z. H.
Lin
,
Y. S.
Zhou
,
Q.
Jing
,
Y.
Su
,
J.
Yang
,
J.
Chen
,
C.
Hu
, and
Z. L.
Wang
,
ACS Nano
7
,
9213
9222
(
2013
).
32.
P. K.
Yang
,
Z. H.
Lin
,
K. C.
Pradel
,
L.
Lin
,
X.
Li
,
X.
Wen
,
J. H.
He
, and
Z. L.
Wang
,
ACS Nano
9
,
901
907
(
2015
).
33.
W. S.
Jung
,
M. G.
Kang
,
H. G.
Moon
,
S. H.
Baek
,
S. J.
Yoon
,
Z. L.
Wang
,
S. W.
Kim
, and
C. Y.
Kang
,
Sci. Rep.
5
,
9309
(
2015
).
34.
J. H.
Lee
,
J.
Kim
,
T. Y.
Kim
,
M. S. A.
Hossain
,
S. W.
Kim
, and
J. H.
Kim
,
J. Mater. Chem. A
4
,
7983
7999
(
2016
).
35.
H.
Yu
,
T.
Huang
,
M.
Lu
,
M.
Mao
,
Q.
Zhang
, and
H.
Wang
,
Nanotechnology
24
,
405401
(
2013
).
36.
S. H.
Bae
,
O.
Kahya
,
B. K.
Sharma
,
J.
Kwon
,
H. J.
Cho
,
B.
Özyilmaz
, and
J. H.
Ahn
,
ACS Nano
7
,
3130
3138
(
2013
).
37.
M. B.
Lee
,
C. Y.
Chen
,
S.
Wang
,
S. N.
Cha
,
Y. J.
Park
,
J. M.
Kim
,
L. J.
Chou
, and
Z. L.
Wang
,
Adv. Mater.
24
,
1759
1764
(
2012
).
38.
C.
Chang
,
V. H.
Tran
,
J.
Wang
,
Y. K.
Fuh
, and
L.
Lin
,
Nano Lett.
10
,
726
731
(
2010
).
39.
Y.
Liu
,
G.
Tian
,
Y.
Wang
,
J.
Lin
,
Q.
Zjang
, and
H. F.
Hohmann
,
J. Intell. Mater. Syst. Struct.
20
,
575
585
(
2009
).
40.
S. W.
Chen
,
X.
Cao
,
N.
Wang
,
L.
Ma
,
H. R.
Zhu
,
M.
Willander
,
Y.
Jie
, and
Z. L.
Wang
,
Adv. Energy Mater.
7
,
1601255
(
2016
).
41.
M.
Zhang
,
Y.
Jie
,
X.
Cao
,
J.
Bian
,
T.
Li
,
N.
Wang
, and
Z. L.
Wang
,
Nano Energy
30
,
155
161
(
2016
).

Supplementary Material