Journal ofMaterials Chemistry A

PAPER

Publ

ishe

d on

28

July

201

8. D

ownl

oade

d by

Kor

ea A

dvan

ced

Inst

itute

of

Scie

nce

& T

echn

olog

y / K

AIS

T o

n 2/

20/2

019

4:36

:21

AM

.

View Article OnlineView Journal | View Issue

A flexible comb e

aBeijing Institute of Nanoenergy & Nano

National Center for Nanoscience and Tech

China. E-mail: danverzhu@gmail.combSchool of Mechanical Engineering, Kore

Technology (KAIST), 291 Daehak-ro, YuseoncDepartment of Industrial and Manufactu

Missouri, Columbia, Missouri 65211, USAdState Key Laboratory of Electronic Thin Fi

Electronic Science and Technology, ChengdueSchool of Materials Science and Engin

Technology, Huainan, 232001, P. R. China.fDepartment of Chemistry, Capital Normal

E-mail: songwx@cnu.edu.cn

† Electronic supplementary information (ES

Cite this: J. Mater. Chem. A, 2018, 6,16548

Received 13th May 2018Accepted 27th July 2018

DOI: 10.1039/c8ta04443k

rsc.li/materials-a

16548 | J. Mater. Chem. A, 2018, 6, 16

lectrode triboelectric–electretnanogenerator with separated microfibers fora self-powered position, motion direction andacceleration tracking sensor†

Jianxiong Zhu, *ab Xiaoyu Guo,c Dehuan Meng,d Minkyu Cho,b Inkyu Park,b

Run Huang *e and Weixing Song*f

In this paper, we report a flexible comb electrode triboelectric–electret coupling nanogenerator using

a separated friction microfiber object for self-powered position, motion direction and acceleration tracking

sensing and its energy harvesting. The power was generated from the coupling of the electrostatic and

triboelectric effects among a separated triboelectric object, a polytetrafluoroethylene (PTFE) film and

interdigital electrodes. Under an acceleration of 1 m s�2, we found that the corona charged PTFE film with

a sliding motion reached �3 times more short-circuit current (Isc) and �6 times more open-circuit voltage

(Voc) than the PTFE film without corona charges, respectively. The device can be a good self-powered

acceleration tracking sensor, where the reasons were a stability voltage out during the process of separated

components frictions and a numerical relationship of Isc with increased accelerations. Moreover, potential

applications using several different separated materials such as a bulk of carbon microfibers, a finger sliding

with a nylon glove, and water microdrops were used to show the energy harvesting of various friction

materials and their effective contact area. It was found that a much larger contact area “finger” sliding

showed a Voc of �45 V, whereas a smaller contact area “microdrop” sliding on the prototype presented

a maximum peak Voc of �1.8 V.

1. Introduction

Nowadays, motion tracking is increasingly needed for a widerange of elds, such as automatic control, robotics, sports,entertainment and the Internet of Things (IoT). Traditionalmotion sensors which usually rely on optics, magnetics,microwaves, or acoustic sound have greatly boosted the devel-opment of motion sensing technologies.1–4 Nevertheless, thelimitations of the above-mentioned sensors are due to therequirement of an external power source and the generation ofa small electrical response signal. Sensors with a good power

systems, Chinese Academy of Sciences,

nology (NCNST), Beijing 100083, P. R.

a Advanced Institute of Science and

g-gu, Daejeon 34141, Republic of Korea

ring Systems Engineering, University of

lms and Integrated Devices, University of

, 610054, P. R. China

eering, Anhui University of Science &

E-mail: huang.run@stu.xjtu.edu.cn

University, Beijing 100048, P. R. China.

I) available. See DOI: 10.1039/c8ta04443k

548–16555

harvesting ability and a quick electrical response are drawingthe attention of scientists for self-powered and autonomoussystems.5–9 In addition, energy harvesting using electromag-netic, piezoelectric, electrostatic and triboelectric mechanismshas been demonstrated effectively to convert mechanical energyinto electricity.10–16 Among the above-mentioned various energyharvesting methods, triboelectric is one of the most efficientwith a quick electrical response and it is an interesting topicbecause of the dissimilar material contact resulting in highenergy conversion and quick electron transfer. Though themechanism of the triboelectric effect was under debate, tribo-electric nanogenerators (TENGs) based on the triboelectriceffect have been demonstrated to be cost-effective, reliable, andextremely efficient devices to convert mechanical energy toelectricity. Till now, three fundamental operating principles ofTENGs have been developed, including contact-separationmode, single-electrode mode, and in-plane sliding mode.17–20

Besides, electret polytetrauorethylene (PTFE) material hasbeen largely adopted for actual applications due to attractiveproperties, such as low friction coefficient, high thermalstability, low dielectric constant, good mechanical strength,strong electron-attracting ability and excellent exibility.

Researchers have tried numerous ways to improve the elec-trical performance of energy harvesting and to broaden the

This journal is © The Royal Society of Chemistry 2018

Paper Journal of Materials Chemistry A

Publ

ishe

d on

28

July

201

8. D

ownl

oade

d by

Kor

ea A

dvan

ced

Inst

itute

of

Scie

nce

& T

echn

olog

y / K

AIS

T o

n 2/

20/2

019

4:36

:21

AM

. View Article Online

application of electret-based triboelectric materials.21–44 Xi et al.reported an interdigital electrode triboelectric generator forrotating energy harvesting.22 A micro-grating sliding nano-generator which can increase the energy efficiency by 50% wasreported by Wang et al. around the same time.24 Liang et al.introduced a single electrode triboelectric nanogenerator withsponge-like porous PTFE for energy harvesting.25 However, noneof the research considered carbon microbers as a separatedobject for any energy harvesting. As a tracking sensor, Wang et al.reported tracking detection in a tube system using a single elec-trode triboelectric generator.32 Aer that, their research groupcontinued to introduce a motion tracking array system based ona single electrode.33,34 However, it was found that the harvestedcurrent in their design was small with a single electrode. To avoidusing a single electrode, Zhu et al. designed a linear-gratingtriboelectric system for energy harvesting and sensing applica-tion.44 Themovable electrode in their design with disturbed signaloutputs was unavoidable which constrained its potential indus-trial application. This paper reports a exible triboelectric–electretnanogenerator with a separated carbon microber triboelectricobject for a self-powered position, motion direction and acceler-ation tracking sensor and its energy harvesting. We found that thecorona charged PTFE for sliding energy harvesting can harvest�3 times more short-circuit current (Isc) and �6 times moreopen-circuit voltage (Voc) than that without corona charge.Furthermore, we found that the device can be used as a good self-powered position, motion direction and acceleration trackingsensor due to the numerical relationship of Isc with increasedaccelerations. Moreover, to broaden the real potential application,we found that a much larger contact area “nger” sliding anda much smaller contact area “microdrop” sliding showed a Vocof �45 V and �1.8 V, respectively.

Fig. 1 (a and b) 3D schematics and photograph of a flexible comb trigenerated power by motions of the triboelectric material carbon microfi

This journal is © The Royal Society of Chemistry 2018

2. Experimental section

Fig. 1a and b depict the 3D schematic of a exible triboelectric–electret nanogenerator. The device contained a PTFE lm (4 cm� 18 cm � 80 mm) and a comb electrode layer (at the bottomside of PTFE) which was coated with copper using magnetronsputtering (PVD 75 Pro Line Kurt J. Lesker Company) for 20 minonto the above-mentioned PTFE electret lm. Fig. S1 in the ESI†shows the dimensional comb electrode and its numberedpositions on the triboelectric–electret nanogenerator lm;meanwhile Fig. S2 in the ESI† shows the positive (forward) orthe negative (backward) motion direction along the X axis. Thenumber of nger interdigitals was 8 in total in our design. Thepower was generated due to the coupling of the triboelectriceffect and the electrostatic induction among the separatedtriboelectric object, the PTFE lm and the interdigital electrode.Lead wires were used to connect the comb electrodes for datameasurement. Note that the PTFE lm was charged througha high voltage discharge system (high voltage polarizationapparatus, Model ET-2673A, Entai Company, Nanjing, China) atroom temperature for 5min with a voltage of�6.15 kV as shownin Fig. S3a (ESI†) and the surface voltage Vs with time on thePTFE lm is shown in Fig. S3b (ESI†). The surface potentials ofthe charged samples were measured by using a potentiometer(Monroe Electronics, Isoprobe Electrostatic Voltmeter, Model279). Fig. 1c and d depict the prototype with a separated carbonmicrober object which generated power in the comb electrode(on the bottom side of PTFE lm) by the forward and backwardmotion of the triboelectric material carbon microber objectalong the top surface of the PTFE lm. The reason for choosingthe microber material was that the carbon microbers canresult in a better friction contact with the PTFE lm due to the

boelectric–electret nanogenerator, and (c and d) a prototype whichber object.

J. Mater. Chem. A, 2018, 6, 16548–16555 | 16549

Journal of Materials Chemistry A Paper

Publ

ishe

d on

28

July

201

8. D

ownl

oade

d by

Kor

ea A

dvan

ced

Inst

itute

of

Scie

nce

& T

echn

olog

y / K

AIS

T o

n 2/

20/2

019

4:36

:21

AM

. View Article Online

relatively larger contact surface area compared to materialswithout any microbers.10,45

The surface morphology of the PTFE electret nanogeneratorlm and the carbon microber separated object was charac-terized by using a scanning electron microscope (SEM HitachiSU-8020, Hitachi Company, Tokyo, Japan) as shown in Fig. 2.Fig. 2a and b reveal the morphology of the PTFE lm and thecarbon microber separated object, respectively. It was obviousto conclude that the carbon microber separated objectcontributed a much larger surface area to the PTFE lmresulting in a high power through the triboelectric effect. Inaddition, Fig. S4 in the ESI† shows the atomic force microscopy(AFM) image of the PTFE lm surface, and it was found that thesurface of the PTFE lm was about 70 nm in height.

3. Results and discussion

The PTFE electret lm was characterized by X-ray diffraction(XRD, X'pert3 powder, Panalytical company, Almelo, Nether-lands) and Raman spectroscopy (HORIBA Jobin Yvon LabRAMHR, France), respectively. The XRD pattern of the PTFE electretnanogenerator lm showed an intense peak centered at 2q ¼18.06� (Fig. 3a), which indicated a long-range order in the (100)lattice plane. The Raman spectra of the PTFE electret lmshowed three prominent peaks around 1400 cm�1 which areattributed to the triboelectric material PTFE (Fig. 3b). Fig. 4

Fig. 2 Characterization of the PTFE film by SEM analysis, (a) the morphomicrofibers were arranged in order.

Fig. 3 (a) Wide angle X-ray diffraction pattern and (b) Raman spectra of

16550 | J. Mater. Chem. A, 2018, 6, 16548–16555

schematically displays a triboelectric–electret nanogeneratorwith the separated triboelectric microber object for slidingenergy harvesting. When the separated carbon microberobject was slid on the comb designed nanogenerator, chargeswould be generated due to the triboelectric effect between thecarbon microbers and the PTFE lm (Fig. 4a). The carbonmicrober object was instantaneously positively charged,whereas the contacted PTFE lm was simultaneously negativelycharged. The comb electrode on the other side of the PTFE lmwas induced with opposite charges due to electrostatic induc-tion.36 As shown in Fig. 4b and c, we obtained regular electricaloutput signals which can be used to indicate the trackinginformation during each sliding process cycle. The charges onthe PTFE electret surface cannot be conducted away orneutralized. The potential voltage in the process of the carbonmicrober object (see Fig. 5a and b) was obtained from a FiniteElement Analysis (FEA) simulation performed with COMSOLMultiphysics®. Under ideal conditions, we found that when themovable object was slid onto the nger interdigitals, power wasgenerated by mechanical–electrical energy conversion due toelectrostatic and triboelectric effects. When the PTFE electretwas subjected to a high voltage discharge on its top surface, thecharge density s on its surface can be calculated as follows,46,47

s ¼ Vs

3r30

d(1)

logy of PTFE, and (b) the morphology of triboelectric material carbon

the PTFE electret film.

This journal is © The Royal Society of Chemistry 2018

Fig. 4 Working principle of a flexible triboelectric nanogenerator with a separated object for energy harvesting: (a) carbon microfiber object isslid on the PTFE comb designed nanogenerator. (b) short-circuit current (Isc); themarked number indicates the specific position on the PTFE film.(c) Open-circuit voltage (Voc) during the sliding.

Fig. 5 COMSOL simulation results of a flexible triboelectric–electret nanogenerator with a separated object: (a) separated object moving ontoa finger interdigital of the comb electrode and (b) moving onto a gap of the comb electrode.

Paper Journal of Materials Chemistry A

Publ

ishe

d on

28

July

201

8. D

ownl

oade

d by

Kor

ea A

dvan

ced

Inst

itute

of

Scie

nce

& T

echn

olog

y / K

AIS

T o

n 2/

20/2

019

4:36

:21

AM

. View Article Online

where 3r and 30 are the dielectric constant of the PTFE electretand air, respectively, Vs is the surface potential of the PTFEelectret, and d is the distance between the needle tip and thePTFE lm surface. When a sliding contact occurred between thetop surface of PTFE and the carbon microber separated object,the induction output voltage from the comb electrode was givenas22,46,47

Voc ¼ Qoc

Ct

(2)

Ct is the total capacitance among the comb electrodes andcarbon microber separated object, which can be expressed as31

Ct ¼ C1 þ C2 ¼3r30

XdL� dW

hþ n

3r30Lfhf

Lg

(3)

This journal is © The Royal Society of Chemistry 2018

where dL and dW are the differential length and width of thePTFE lm, respectively, h is the thickness of the PTFE lm, Lf, hfand Lg are the interdigital length, thickness and gap distance ofthe two nger interdigitals, respectively, and n is the number ofinterdigital electrodes. From eqn (1)–(3), we obtained thefollowing equation,

Voc ¼ Vs

LW

d

XdL� dW

hþ n

Lfhf

Lg

! (4)

It was observed that the dimensional structures of the comband the surface voltage were strongly related to the outputvoltage of the PTFE lm.

J. Mater. Chem. A, 2018, 6, 16548–16555 | 16551

Journal of Materials Chemistry A Paper

Publ

ishe

d on

28

July

201

8. D

ownl

oade

d by

Kor

ea A

dvan

ced

Inst

itute

of

Scie

nce

& T

echn

olog

y / K

AIS

T o

n 2/

20/2

019

4:36

:21

AM

. View Article Online

To analyze the inuence with and without corona charge ina triboelectric nanogenerator, a linear motor (LinMot USA, Inc.,Elkhorn, WI, USA) was adopted to provide periodical back-and-forth motion to the carbon microber separated triboelectricobject through periodic reciprocating motion as shown inFig. S5 (ESI†). The time parameter “T1, T2, T3, T4, T5 and T6”were determined from the acceleration of the linear motorsystem. Note that an electrometer Keithley 6514 (KeithleyInstruments, Inc, Ohio, USA) connected to a computer wasused to measure the short-circuit current (Isc) and the open-circuit voltage (Voc). The data were recorded through a Lab-VIEW® program in a real-time manner. Fig. S6 in the ESI†presents the measured Isc and Voc for the PTFE lm withoutand with corona charge. It was found that the corona chargedPTFE electret lm can exhibit �6 times more Voc and �3 timesmore Isc than the untreated PTFE lm. The results obviouslyindicated that corona charging was an efficient method forenhancing the effective Isc and Voc.37,38

To further investigate the electrical performance of thenanogenerator with increased accelerations, the above-mentioned linear motor was used for an increased accelerationtest (Fig. 6a–c). The velocity with respect to time at differentaccelerations is shown in Fig. S7 (ESI†). The electrometerKeithley 6514 was used to measure the Voc, Isc and Qsc, respec-tively. We found that with a 5 m s�2 acceleration on the proto-type, the peak value of Isc increased by �600% compared to 0.1m s�2. It was interesting to nd that both Voc and Qsc presentedalmost similar peak values with increased accelerations. Themaximum Isc, Voc and Qsc can reach up to �0.6 mA, �11.2 V and�4.8 nQ with an acceleration of 5 m s�2, respectively. Fig. 7demonstrates the position sensing and the moving direction of

Fig. 6 Measurement results for a triboelectric–electret nanogenerator uwith increased accelerations, (b) Voc with increased accelerations, and (c)Isc, respectively.

16552 | J. Mater. Chem. A, 2018, 6, 16548–16555

the triboelectric–electret nanogenerator. Fig. 7c and d show thatthe system still works at totally different accelerations, andFig. 7e and f depict the motion direction which was dened bythe output voltage peak value. The specic positions werecalculated from the number of current peaks in Fig. S8ESI†.44,48,49 To explain the chosen 4 pairs of nger interdigitalelectrodes on the above-mentioned PTFE electret lm, we canconclude that the more the number of nger interdigital elec-trodes on the above-mentioned PTFE electret lm, the higherthe position resolution of the device for the sensing that can beidentied. However, with fewer interdigital electrodes, a largesensing range can be obtained due to a relatively large frictionarea between the interdigital electrode and the movableobjects.42,44,45,50 Table S1 in the ESI† shows the comparison ofoutput voltage and short current in our study with the currentstate-of-the-art results. The current triboelectric-based objecttracking sensors are based on a single electrode which easilyresults in an unstable state. It means that one electrode directlyinteracts with the triboelectric electrode layer, and the otherreference electrode acts as a large conductor or the ground. Theleover charges on the ground or the large conductor might bedetrimental to an accurate measurement.51–54 Compared toa triboelectric-based energy harvester based on block objectfriction, our work presented �6 times higher charges than thereference studies.32,34 This concluded that our study had a muchwider acceleration range (0.1–5 m s�2) and higher open-circuitvoltage and short-current compared to the results of previouslypublished work during the real measurement. To demonstratethe reliability and durability of the PTFE lm triboelectric–electret nanogenerator, the open-circuit voltage was measuredfor more than 2400 seconds (�1000 cycles) as shown in Fig. S9

sing separated carbon microfibers with increased accelerations: (a) IscQsc with increased accelerations. (d) Accelerations in respect to Voc and

This journal is © The Royal Society of Chemistry 2018

Fig. 7 Position, motion direction and acceleration tracking sensing: schematic diagram (a) with forward motion and (b) with backward motion.(c) The measured current with an acceleration of 0.1 m s�2 and (d) the measured current with an acceleration of 5 m s�2. (e) Moving directionbased on the voltage peak value and (f) the motion along the X direction was identified from the voltage peak value.

Fig. 8 Water microdrop flowing application of a flexible comb designed nanogenerator: (a) a microdrop was on the top surface of the PTFEnanogenerator, a copper electrode was at the back side of the PTFE film, and (b) a microdrop is located at the gap. (c) The real-time Isc value ofthe flowing microdrop and (d) the real-time Voc value of the flowing microdrop.

This journal is © The Royal Society of Chemistry 2018 J. Mater. Chem. A, 2018, 6, 16548–16555 | 16553

Paper Journal of Materials Chemistry A

Publ

ishe

d on

28

July

201

8. D

ownl

oade

d by

Kor

ea A

dvan

ced

Inst

itute

of

Scie

nce

& T

echn

olog

y / K

AIS

T o

n 2/

20/2

019

4:36

:21

AM

. View Article Online

Journal of Materials Chemistry A Paper

Publ

ishe

d on

28

July

201

8. D

ownl

oade

d by

Kor

ea A

dvan

ced

Inst

itute

of

Scie

nce

& T

echn

olog

y / K

AIS

T o

n 2/

20/2

019

4:36

:21

AM

. View Article Online

(ESI†). The measurements demonstrating the good exibility ofthe triboelectric generator with different bending states areshown in Fig. S10 (ESI†). Furthermore, Fig. S11 and Video S1 inthe ESI† show more measurements which were used todemonstrate the exibility and durability of our device.

Fig. 8a and b depict a water microdrop owing applicationusing a much smaller contact area compared to the carbonmicrober object. When a microdrop was slid on the prototype,the generated Isc and Voc showed a maximum peak of �58 nAand �1.8 V, respectively (see Fig. 8c and d). To nd out theinuence of the contact friction area, the measurement result ofa nger sliding with a glove on the prototype is shown inFig. S12 (ESI†). When a nger was slid across the top side of thePTFE nanogenerator, the generated Isc and Voc from the combelectrode (back side of the PTFE lm) can be recorded as shownin Fig. S12.† The generated maximum peak of Isc and Voc wasobtained to be �0.21 mA and �45 V, respectively. Moreover,a comparison of the above-mentioned separated triboelectricmaterials is shown in Fig. S13 ESI.† It was found that a largecontact area of the triboelectric materials can result in a higherVoc of the designed comb nanogenerator.

4. Conclusions

In conclusion, we presented a comb electrode triboelectric–electret coupling nanogenerator lm with a separated micro-ber object for an enhanced self-powered position, motiondirection and acceleration tracking sensor and its energyharvesting. It was found that the corona charged PTFE lm forsliding energy harvesting can harvest �3 times more Isc and�6 times more Voc than the PTFE lm without corona chargeat an acceleration of 1 m s�2. It was found that a ercer motioncontact caused few contributions to the charges but presenteda �5 times higher Isc at 5 m s�2 than that at 0.1 m s�2.Furthermore, we found that the device can be used as a goodself-powered position, motion direction and accelerationtracking sensor due to the numerical relationship of Isc withincreased accelerations. When a much smaller contact area“microdrop” was slid on the prototype, the generated Voc andIsc presented a maximum peak of �1.8 V and �58 nA,respectively. The triboelectric–electret motion sensor hassignicant advantages such as no external power supplies,facile fabrication, high efficiency, low cost, real-time moni-toring, simple signal processing, and high reliability. More-over, this work opens up new possibilities for self-poweredmotion detection and new prospects for the practical appli-cations of comb electrode TENGs with microber friction. Thiswork not only demonstrates the great application potential ofthe self-powered position and acceleration tracking sensor,but also would be very useful for touch screen displays,handheld tracking devices, domestic security, traffic moni-toring, etc.

Conflicts of interest

There are no conicts to declare.

16554 | J. Mater. Chem. A, 2018, 6, 16548–16555

Acknowledgements

This study was funded through No. GRCK2016082914424352and 20170719075917475 by the Shenzhen Science and Tech-nology Innovation Committee, the authors also acknowledgedthe nancial support of the National Science Foundation ofChina (No. 21773009 and 61404035), the authors nally thankthe National Research Foundation of Korea (NRF) (No.2017M3C1B2085318).

References

1 S. Raja and G. Ilangovan, International Journal of AdvancedResearch in Computer Engineering & Technology, 2013, 2,2882–2884.

2 G. Padmavathi, D. Shanmugapriya and M. Kalaivani,Wireless Sensor Network, 2010, 2, 173–185.

3 M. Omidbeike, Y. R. Teo, Y. K. Yong and A. J. Fleming, IFAC-PapersOnLine, 2017, vol. 50, pp. 10913–10917.

4 U. Tana, W. T. Latt, F. Widjaja, C. Y. Shee, C. N. Riviere andW. T. Ang, Sens. Actuators, A, 2009, 150, 116–123.

5 S. P. Beeby, M. Tudor and N. M. White, Meas. Sci. Technol.,2006, 17, R175–R195.

6 D. Lingam, A. R. Parikh, J. Huang, A. Jain and M. Minary-Jolandan, Int. J. Smart Nano Mater., 2013, 4, 229–245.

7 R. Hinchet, W. Seung and S. W. Kim, ChemSusChem, 2015, 8,2327.

8 C. Dagdeviren, B. D. Yang, Y. Su, P. L. Tran, P. Joe,E. Anderson and J. A. Rogers, Proc. Natl. Acad. Sci. U. S. A.,2014, 111, 1927–1932.

9 S. Li, Q. Zhong, J. Zhong, X. Cheng, B. Wang, B. Hu andJ. Zhou, ACS Appl. Mater. Interfaces, 2015, 7, 14912.

10 Q. Liao, Z. Zhang, X. Zhang, M. Mohr, Y. Zhang andH. J. Fecht, Nano Res., 2014, 7, 917–928.

11 W. J. Xu, M. Kranz, S. H. Kim and M. Allen, J. Micromech.Microeng., 2010, 20, 104003.

12 W. Zhang, R. Zhu, V. Nguyen and R. Yang, Sens. Actuators, A,2014, 205, 164–169.

13 J. Cao and J. Wu, Mater. Sci. Eng., R, 2011, 71, 35–52.14 Q. J. Liang, X. Yan, Y. Gu, K. Zhang, M. Liang, S. Lu and

Y. Zhang, Sci. Rep., 2015, 5, 9080.15 M. S. Sarwar, Y. Dobashi, C. Preston, J. K. M. Wyss,

S. Mirabbasi and J. D. W. Madden, Sci. Adv., 2017, 3,1602200.

16 J. Yang, J. Chen, Y. Liu, W. Yang, Y. Su and Z. L. Wang, ACSNano, 2014, 8, 2649–2657.

17 Y. Chiu and Y. C. Lee, J. Micromech. Microeng., 2013, 23,015012.

18 J. Zhong, Q. Zhong, G. Chen, B. Hu, S. Zhao, X. Li andJ. Zhou, Energy Environ. Sci., 2016, 9, 3085.

19 F. R. Fan, W. Tang and Z. L. Wang, Adv. Mater., 2016, 28,4283–4305.

20 B. U. Hwang, J. H. Lee, T. Q. Trung, E. Roh, D. I. Kim,S. W. Kim and N. E. Lee, ACS Nano, 2015, 9, 8801.

21 N. Kaur, J. Bahadur, V. Panwar, P. Singh, K. Rathi and K. Pal,Sci. Rep., 2016, 6, 38835.

This journal is © The Royal Society of Chemistry 2018

Paper Journal of Materials Chemistry A

Publ

ishe

d on

28

July

201

8. D

ownl

oade

d by

Kor

ea A

dvan

ced

Inst

itute

of

Scie

nce

& T

echn

olog

y / K

AIS

T o

n 2/

20/2

019

4:36

:21

AM

. View Article Online

22 Q. Leng, H. Guo, X. He, G. Liu, Y. Kang, C. Hu and Y. Xi,J. Mater. Chem. A, 2014, 2, 19427–19434.

23 Y. Mao, D. Geng, E. Liang and X. Wang, Nano Energy, 2015,15, 227–234.

24 G. Zhu, Y. S. Zhou, P. Bai, X. S. Meng, Q. Jing, J. Chen andZ. L. Wang, Adv. Mater., 2014, 26, 3788.

25 M. Wang, N. Zhang, Y. Tang, H. Zhang, C. Ning, L. Tian andE. Liang, J. Mater. Chem. A, 2017, 5, 12252.

26 Z. Wang, L. Cheng, Y. Zheng, Y. Qin and Z. L. Wang, NanoEnergy, 2014, 10, 37–43.

27 X. Y. Wei, G. Zhu and Z. L. Wang, Nano Energy, 2014, 10, 83–89.

28 L. E. Helseth and X. D. Guo, Langmuir, 2015, 31, 3269–3276.29 B. W. Jeong, M. Kim, J. Lee, Y. K. Eun, J. W. Choi and J. Kim,

Micromachines, 2017, 8, 51.30 Y. Yang, H. Zhang, X. D. Zhong, F. Yi, R. M. Yu, Y. Zhang and

Z. L. Wang, ACS Appl. Mater. Interfaces, 2014, 6, 3680–3688.31 Z. C. Yang, E. Halvorsen and T. Dong, Appl. Phys. Lett., 2012,

100, 213905.32 F. Yi, L. Lin, S. M. Niu, J. Yang, W. Z. Wu, S. H. Wang,

Q. L. Liao, Y. Zhang and Z. L. Wang, Adv. Funct. Mater.,2014, 24, 7488–7494.

33 C. B. Han, C. Zhang, X. H. Li, L. Zhang, T. Zhou, W. G. Huand Z. L. Wang, Nano Energy, 2014, 9, 325–333.

34 Y. J. Su, G. Zhu, W. Q. Yang, J. Yang, J. Chen, Q. S. Jing,Z. M. Wu, Y. D. Jiang and Z. L. Wang, ACS Nano, 2014, 8,3843–3850.

35 J. K. Chen, P. Ding, R. Z. Pan, W. P. Xuan, D. P. Guo, Z. Ye,W. L. Yin, H. Jin, X. Z. Wang, S. Dong and J. K. Luo, NanoEnergy, 2017, 34, 442–448.

36 F. Hu, Q. Cai, F. Liao, M. Shao and S. T. Lee, Small, 2015, 11,5611–5628.

37 Z. Tao, L. M. Zhang, F. Xue, W. Tang, C. Zhang andZ. L. Wang, Nano Res., 2016, 9, 1442–1451.

38 N. Wu, H. L. Jiang, W. B. Li, S. Z. Lin, J. W. Zhong, F. Yuan,L. Huang, B. Hu and J. Zhou, J. Mater. Chem. A, 2017, 5,12787–12792.

This journal is © The Royal Society of Chemistry 2018

39 Z. G. Fang, K. H. Chan, X. Lu, C. F. Tan and G. W. Ho, J.Mater. Chem. A, 2018, 6, 52–57.

40 X. X. Chen, Y. Song, H. T. Chen, J. X. Zhang and H. X. Zhang,J. Mater. Chem. A, 2017, 5, 12361–12368.

41 Y. Cheng, R. Wang, K. H. Chan, X. Lu, J. Sun and G. W. Ho,ACS Nano, 2018, 12, 3898–3907.

42 G. Zhu, W. Q. Yang, T. J. Zhang, Q. S. Jing, J. Chen,Y. S. Zhou, P. Bai and Z. L. Wang, Nano Lett., 2014, 14,3208–3213.

43 C. H. Liow, X. Lu, C. Fu, T. Kwok, H. Chan, K. Y. Zeng, S. Z. Liand G. W. Ho, Small, 2018, 14, 1702268.

44 G. Zhu, J. Chen, Y. Liu, P. Bai, Y. S. Zhou, Q. Jing, C. F. Panand Z. L. Wang, Nano Lett., 2013, 13, 2282–2289.

45 X. H. Li, Z. H. Lin, G. Cheng, X. N. Wen, Y. Liu, S. M. Niu andZ. L. Wang, ACS Nano, 2014, 8, 10674–10681.

46 S. M. Niu, S. H. Wang, L. Lin, Y. Liu, Y. S. Zhou, Y. F. Hu andZ. L. Wang, Energy Environ. Sci., 2013, 6, 3576.

47 S. M. Niu, Y. Liu, X. Y. Chen, S. H. Wang, Y. S. Zhou, L. Lin,Y. N. Xie and Z. L. Wang, Nano Energy, 2015, 12, 760–774.

48 X. H. Li, C. B. Han, L. M. Zhang and Z. L. Wang, Nano Res.,2015, 8, 3197–3204.

49 M. X. Chen, X. Y. Li, L. Lin, W. M. Du, X. Han, J. Zhu,C. F. Pan and Z. L. Wang, Adv. Funct. Mater., 2014, 24,5059–5066.

50 Z. Lin, G. Cheng, X. H. Li, P. Yang, X. N. Wen and Z. L. Wang,Nano Energy, 2015, 15, 256–265.

51 Z. L. Wang, Faraday Discuss., 2014, 176, 447–458.52 S. H. Wang, L. Lin and Z. L. Wang, Nano Energy, 2015, 11,

436–462.53 T. T. Zhang, C. Z. Lai, M. A. Fierke, A. Stein and

P. Buhlmann, Anal. Chem., 2012, 84, 7771–7778.54 J. Bragard, A. Simic, D. Laroze and J. Elorza, Phys. Rev. E,

2015, 92, 062919.

J. Mater. Chem. A, 2018, 6, 16548–16555 | 16555