Fully Stretchable Textile TriboelectricNanogenerator with Knitted Fabric StructuresSung Soo Kwak,†,⊥ Han Kim,†,⊥ Wanchul Seung,† Jihye Kim,† Ronan Hinchet,†

and Sang-Woo Kim*,†,‡

†School of Advanced Materials Science and Engineering and ‡SKKU Advanced Institute of Nanotechnology (SAINT),Sungkyunkwan University (SKKU), Suwon 440-746, Republic of Korea

*S Supporting Information

ABSTRACT: Harvesting human-motion energy for power-integrated wearable electronics could be a promising way toextend the battery-operation time of small low-power-consumption electronics such as various sensors. For thispurpose, a fully stretchable triboelectric nanogenerator (S-TENG) that has been fabricated with knitted fabrics and hasbeen integrated with the directly available materials andtechniques of the textile industry is introduced. This devicehas been adapted to cloth movement and can generateelectricity under compression and stretching. We inves-tigated plain-, double-, and rib-fabric structures andanalyzed their potentials for textile-based energy harvesting. The superior stretchable property of the rib-knitted fabriccontributed to a dramatic enhancement of the triboelectric power-generation performance owing to the increased contactsurface. The present study shows that, under stretching motions of up to 30%, the S-TENG generates a maximum voltageand a current of 23.50 V and 1.05 μA, respectively, depending on the fabric structures. Under compressions at 3.3 Hz, theS-TENG generated a constant average root-mean square power of up to 60 μW. The results of this work show the feasibilityof a cloth-integrated and industrial-ready TENG for the harvesting of energy from human biomechanical movements incloth and garments.

KEYWORDS: knitted fabric, triboelectric nanogenerator, stretchable power source, self-powered system, wearable electronics

The rapid progress of wearable electronics has initiated aparadigm shift in the lives of humans. The demand forwearable energy sources is increasing, and this is greatly

relevant to the Internet of Things (IoT) that enables access toenvironmental data in real time.1 However, due to thelightweight nature, flexibility, and integration requirements ofwearable electronics, conventional lithium-ion batteries are notvery suitable because of their limited capacity, lesser flexibility,lesser stretchability, and lesser biocompatibility. Consequently,the development of sustainable and self-powered systems ishighly desirable for the wearable electronics of the future.Biomechanical energy is one of the most common andabundant forms of mechanical energy that can be generatedon a daily basis by humans. Scavenging the wastedbiomechanical energy from human motions could solve thewearable-electronics power-supply issue, and this has resultedin the great amount of current corresponding attention.2,3

The development of triboelectric nanogenerators (TENGs),based on the coupling effects between triboelectrification andelectrostatic induction through contact separation or relativesliding motions, has provided a highly feasible, affordable, andpractical approach for the realization of mechanical-movement-based self-powered systems.4−9 Until now, TENGs have been

investigated for the conversion of external mechanical energyinto electrical-power sources and have been developed toenhance power-generation performances through processes likeincreases of the contact area, surface-morphology modification,and surface-treatment functionalization.10,11 These processes,however, have not fully satisfied the wearable-electronics energyrequirements, and until now, the TENGs designed to harvestbiomechanical movements are still suffering from lackedstretchability and durability.12,13

Lastly, the recent progress in mechanical-energy harvestinghas led to the investigations of the TENGs that are based onfabric structures for harvesting energy from various kinds ofhuman motions. In particular, a fabric-type TENG is an idealdesign for the harvesting of the biomechanical energy fromcloth deformations and frictions.14−16 Still, a greatly improvedstretchability, optimized device structure, and enhanced outputperformance are required to increase the biomechanical-energyharvesting of the TENG and its practical applications. It is well-

Received: July 23, 2017Accepted: October 2, 2017Published: October 2, 2017

Artic

lewww.acsnano.org

© XXXX American Chemical Society A DOI: 10.1021/acsnano.7b05203ACS Nano XXXX, XXX, XXX−XXX

known that a knitting technique as one of the mostrepresentative clothing technologies can lead to highstretchability and the aesthetic property of textiles dependingon the knitting patterns.17−19 From the overall knittingproperties, the knitted-fabric characteristics are determined bythe fiber selection and the fabric-surface properties that can becontrolled by the choice of the basic loop of the fabric and thegeometrical configuration of the yarn.20 As for the power-output performance of TENGs, to increase the amount of theinduced triboelectric charges, materials with a larger electron-affinity difference can be selected or the material contact areacan be enlarged.21−23

The present paper delivers a report on a fully stretchableTENG (S-TENG) that has been fabricated with knitted fabricsand that shows a high stretchable property even without theintrinsic stretchability of the constitutive fibers and the largecontact area that are due to microscaled knitted yarns. Toaddress both the durability and industrial-compatibility issues ofprevious studies that are because of the complex and fragilenanostructures,3,24,25 a much simpler constitutive fiber structurethat is composed of triboelectric materials is considered here.The materials are strong and already available in the textileindustry, while the fibers are more robust. As a result, theproposed highly durable devices were created using theavailable industrial machinery. For the selection of the bestdevice structure, the relationship between the stretchability andthe contact area of the diverse knitted structures was firstanalyzed, and the best fabric structure that optimizes the S-TENG power output was selected. The knitted structuresshowed a high stretchability that enables the S-TENG device toundergo cloth stretching up to a 30% strain in the transversedirection. The electrical-output performance of the S-TENG

was systematically characterized through an investigation of thecontact-area effects depending on the stretching conditions andthe knitted structural designs. The measured alternating-current(ac) voltage and current peaks reached 23.50 V and 1.05 μA,respectively, for a 10 × 10 cm2 device under a 30% strain at 1.7Hz. These performances correspond to a generated root-meansquare (RMS) voltage and current of 5.30 V and 29 μA/m2,respectively, thereby representing the power-supply ability ofthe proposed S-TENG device.

RESULTS AND DISCUSSION

For the investigation and fabrication of a robust S-TENG that iscompatible with the current textile industry, the materialselections were carefully considered, and they are based notonly on the available materials in the textile industry, but alsoon the triboelectric series (Figure S1) that indicates whether amaterial is triboelectrically positive or negative. The electricalconductivity of silver (Ag) is high, and it is neutral in thetriboelectric series. In addition, polytetrafluoroethylene (PTFE)is not only widely used in the textile industries, but it is alsovery negative in the triboelectric series. For these reasons, Agand PTFE were selected as the electrode and negativetriboelectric materials, respectively.26−31 The S-TENG isconstructed as a double-arc-shape device of a 10 × 10 cm2

size, and it consists of the knitted PTFE and Ag fabrics on thetop and bottom layers with an Ag electrode in the middle, asshown in Figure 1a and Figure S2. Both yarns, the PTFE (198denier) and the Ag (280 denier) yarns, were knitted into fabricswith a gauge of 12 (needle/inch) using raw materials(produced by the Swicofil AG and X-Silver industrialcompanies, respectively).

Figure 1. Knitted-fabric-based fully stretchable TENGs with different knitted structures: (a) 3D image of the S-TENG structure; (b)nomenclature of the knitted fabrics; (c) FE-SEM image of the fabric loops. Photo images of (d) plain-, (e) double-, and (f) rib-knitted fabricswith insets of the magnified images.

ACS Nano Article

DOI: 10.1021/acsnano.7b05203ACS Nano XXXX, XXX, XXX−XXX

B

Hereafter, such technical knitting-technique nomenclaturewill be used for characterizing the S-TENG device, as shown inFigure 1b. To investigate the exact loop dimensions of theknitted fabrics, field emission scanning electron microscopy(FE-SEM) measurements were carried out, as shown in Figure1c. The lateral dimensions of the unstretched loop of the PTFEplain-knitted fabric pattern show a wale width and a courseheight of approximately 1520 and 2660 μm, respectively, and ayarn diameter of 300 μm. To show the effect of the knittedstructure on the fabric’s mechanical behavior and to determinethe suitable knitting design that maximizes the stretchabilityand the contact area, three different types of knitted structurewere prepared. The following three structures that are widelyused in the commercial textile industries were chosen: plainknit, double knit, and rib knit. Then, fabrics with differences inthe surface properties such as the morphology, roughness,density, and stretchability were fabricated. Tilted and top-viewimages of the plain-, double-, and rib-knitted fabrics are shownin Figure 1d−f. In the experiments, all of the loop dimensions

for the three different knitted structures are varied due to thedifferent loop connections (see Table S1). The yarn integrityenables the loops to deform with interdependent relations, andthis allows the knitted fabrics to express excellent stretchability.It is important to know the stretching and recovery

characteristics of the fabrics depending on the knittingstructures in a determination and attainment of the maximumstretching of the S-TENG device, and consequently, the futurepractical applications are extended.32 Photo images of eachknitted structure under stretching tests are shown in Figure S3.All of the knitted fabrics were fabricated using the same gauge(12-gauge) to characterize both the stretching and surfaceproperties. First, the variation of the stretching property of eachknitted structure is different. The PTFE plain-, double-, and rib-knitted fabrics could recover their original shape afterapproximately 10, 20, and 30% strains in the transversedirection, respectively. The results demonstrated that the rib-knitted fabric distinguishes itself for the greatest stretchability inthe transverse direction. Simple geometrical models of the

Figure 2. Stretching characteristics of the S-TENGs depending on the strain. Schematic images of the stretched structures of the (a) plain, (b)double, and (c) rib fabrics at the maximum elastic deformation. (d) Schematic of the elemental pattern of the rib fabric. (e) Enlarged-arearatio and (f) contact surfaces of the plain, double, and rib fabrics depending on the applied transversal strain.

ACS Nano Article

DOI: 10.1021/acsnano.7b05203ACS Nano XXXX, XXX, XXX−XXX

C

plain-, double-, and rib-knitted fabrics are depicted in Figure 2to show a comparison of their structures during the stretchingmotions.In the case of the plain-knitted fabric, the loop head contacts

the legs of the adjacent loops. Although it is possible to extendthe plain fabric in both the wale and course directions,33 itsextensibility is lower than those of the double and ribstructures, and this is because the contacts between each loopare maintained even during the stretching motions. Therefore,compared with its original state, its potential for increasing thecontact surface during the stretching is lower, leading to almostthe same poststretching contact surface. The experiment resultsshowed that the PTFE plain-knitted fabric can endureapproximately 10% of strain in the transverse direction andrecover itself with no plastic deformation, as shown in Figure2a. Meanwhile, the double-knitted fabric is much morestretchable, and it is worth noting that its high stretchabilityis a product of its double-layered structure, which is composed

of purl- and knit-side layers. The region between both layers,called the middle region, exists for the purpose of not only theconnecting of the loops but also the enhancement of thestretchability. Indeed, the middle region is free from the otherloops at rest, while the contacts are tied up with each other inthe case of the plain-knitted fabrics. In the results, the PTFEdouble-knitted fabric showed approximately 20% of stretch-ability in the transverse direction, which is larger than that ofthe plain-knitted fabric, as shown in Figure 2b. The schematicimage of the stretched rib-knitted fabric is depicted in Figure 2(c). The rib-knitted fabric is also composed of purl- and knit-side layers and includes a middle region. But, in comparisonwith the double-knitted fabric of Figure 2b, the rib fabric iscomposed of two loops on each side, and both sides overlapwhile the middle region is folded and hidden at rest. As aconsequence, the rib-fabric stretchability is even larger, and thepotential in terms of the increasing of the contact surfaceduring the stretching is higher. As a result, the rib-knitted fabric

Figure 3. Schematic images of the working principle of the different S-TENGs. Triboelectric-charge generation process during the stretching(a) plain-, (b) double-, and (c) rib-fabric-based S-TENGs. (d) Output voltage of the rib S-TENG under a 30% strain in the transversedirection with a load resistance of 40 MΩ. (e) Single-period graphic of the output voltage generated in the different stretching and releasingsteps, and the corresponding RMS output voltage.

ACS Nano Article

DOI: 10.1021/acsnano.7b05203ACS Nano XXXX, XXX, XXX−XXX

D

stretched up to approximately 30% of the strain in thetransverse direction, and this represents the most stretchableknit structure in the samples of the present study.For an improved understanding of the different fabric

properties regarding the triboelectric-charge generation, calcu-lations of the contact area of the knitted fabrics depending onthe morphology and the stretching property were performed.Here, the proposed concept of an enlarged-area ratio representsthe ratio of the stretched elemental loop area to the unstretchedplanar yarn area using the cross-sectional view. For theestimation of the enlarged-area ratio, a simplified schematicimage of the elemental loop of the rib-knitted fabric is depictedin Figure 2d. The contact and noncontact regions weredistinguished from each other, and the dimensions and shapeswere considered to obtain the quantitative values. Then, thecalculations of the enlarged-area ratio were carried out byintegrating the elemental loop areas and the planar yarn areasover the whole fabric, as shown in Figure S4 and Table S2. Thevalues of the enlarged-area ratio depending on the strain areplotted in Figure 2e. An increasing trend, where the contactareas of all of the experimented fabrics were increased with theapplied strain until a different elastic-deformation limit wasreached, became evident. In addition, greater enlarged-arearatios are evident in the sequential order of the rib-, double-,and plain-knitted structures with the values of 2.39, 1.77, and1.27 at the maximum stretched state, respectively. Notably,since the properties of the fabrics such as the yarn density andthe structural dimensions differ between the knitted structureseven though all of the knitted samples were knitted using thesame gauge, the enlarged-area ratio varied with each knittedstructure.In another approach, the contact area that is shown in Figure

2f and is based on the top-view photo images of the stretchedknitted fabrics (Figure S3) through the differentiation of the

black and white pixels representing the empty parts, and thePTFE, respectively, was extracted. The contact area wascalculated by multiplying the total stretched area of the fabricby the PTFE pixel density for various strains (see Table S3).Although the PTFE-pixel density was decreased in the unit areaduring the stretching, the total contact area was increased dueto the increased total surface (Figure S5). In the results, thetrends of the calculated total contact areas as a function of thestrain increased for all of the plain-, double-, and rib-knittedfabrics, which is similar to the trends of the enlarged-area ratios.At the maximum stretched states, the contact area reached themaximum values of 108, 163, and 180 cm2 for the plain, double,and ribbed fabrics, respectively. It became evident that themechanical and surface properties of the rib-knitted fabric arenot only superior compared with the plain and double fabrics,but it is also implied that the potential of the rib-knitted fabricfor the generation of triboelectric charges is greater.The working principle of the S-TENG is based on the

coupling of the triboelectric effect and the electrostaticinduction.34−36 Because the PTFE-fabric electron affinity ishigh, it becomes negatively charged after it contacts theconducting lower electron affinity of the Ag fabric. Note thatthe amount of generated charges varies with the knittedstructures due to their different surface properties, such as themorphology and the contact area. The variation of the gapbetween the PTFE and the Ag fabrics under the driving of theexternal forces induces an electrical-potential differencebetween the electrodes, which drives the electrons to flowthrough the external load. Therefore, the repeated contact andseparation of the PTFE and the Ag fabrics generates ACsignals.31,37

Figure 3 shows the fabrics under the different stretching-strain levels 10, 20, and 30% in the transverse direction. Uponthe commencement of the stretching of the fabric, the top,

Figure 4. Simulation- and experimental-voltage trends depending on the structure of the S-TENG. (a) OC electrical-potential distribution thatwas generated by a plain S-TENG under a 10% transversal strain, a double S-TENG under a 20% transversal strain, and a rib S-TENG under a30% transversal strain. (b) Simulation results of the OC electrode-potential difference and (c) experiment results of the electrode-potentialdifference (40 MΩ load) generated by the plain, double, and rib S-TENGs under various transversal strains from 5 to 30%.

ACS Nano Article

DOI: 10.1021/acsnano.7b05203ACS Nano XXXX, XXX, XXX−XXX

E

middle, and bottom knitted layers became closer until contactwas made, as shown in Figure 3a−c, generating the firstelectrical pulse, as shown in Figure 3d,e, or two small pulses ifthe top and bottom layers were not exactly synchronized. Here,the contact surface of the rib-fabric-structured S-TENG ishigher because of its particular double-zigzag knitted structurethat generates a higher total triboelectric-charge amount, and soa higher voltage peak is reached compared with the otherfabrics. Then, with further stretching, the contact area betweeneach part was increased to a greater extent, generating moretriboelectric charges and a second and broader voltage peak.Again, the rib fabric here can extend more than the otherfabrics, enhancing its charge generation, whereby a highervoltage peak can be generated. When releasing the force, thereverse process occurs similarly. Because of the different knittedstructures, the fabric contact areas are different and showdifferent electrical peaks at the point of contact. In addition,because of their different stretching abilities, a discrepancy isshown during the stretching, as shown in Figure S7, leading todifferent overall performances. As a result of this mechanism,the proposed S-TENG can operate in the following threemodes: first, by contacting each layer; second, by stretching thelayers; and finally, a contact and stretching through thecombining of these two modes. Thus, this design is very wellintegrated and adapted for the harvesting of all textilemovements.To investigate the surface effect of the knitted fabrics,11 the

surface-charge density was first calculated using analyticalequations, and this was followed by the performance of finite

element method (FEM) simulations of the open-circuit (OC)-potential distribution for the three different S-TENGs using theCOMSOL Multiphysics software (Comsol). Under the OCconditions, the electrical-potential difference (open-circuitvoltage, VOC) that is generated by the vertical contact andseparation of the S-TENG layers is defined as follows38

σε

= xVOC

0 (1)

where σ is the surface-charge density, x is the gap distancebetween the middle electrode and the triboelectric layers, andε0 is the vacuum permittivity. Since the VOC is proportional tothe surface-charge density, it was expected that the electricalperformance could be enhanced under sufficient friction by theenlarging of the effective contact area.39,40 By facilitatingcontact between a PTFE film and the Ag using eq 1, thesurface-charge density of the triboelectric PTFE−textile layer atapproximately 220 μC/cm2 was estimated. Then, using thisvalue, the triboelectric effects among the three different S-TENGs were simulated and compared. Figure 4a presents thecross-sectional views of the OC electrical-potential distributionacross the plain, double, and rib S-TENGs after the contact andseparation. The simulated electrical-potential differencesbetween the electrodes of the three S-TENGs were plotted asa function of the strain in Figure 4b. The potential differencesthat were generated by the plain, double, and rib S-TENGs are41.2, 45.6, and 50.6 V, respectively, when they were stretchedup to their maximum rates of 10, 20, and 30%. During thestretching beyond 10%, the electrical potential that was

Figure 5. RMS electrical characteristics of the different S-TENGs under a transversal strain and a vertical compression. (a) RMS outputvoltage at a 40 MΩ load and (b) RMS output current at a 100 Ω load generated by the plain, double, and rib S-TENGs under varioustransversal strains from 5 to 30%. (c) RMS voltage and the current and (d) the average power generated by the rib S-TENG in a verticalcontact mode at 3.3 Hz depending on the external-load resistance.

ACS Nano Article

DOI: 10.1021/acsnano.7b05203ACS Nano XXXX, XXX, XXX−XXX

F

generated by the double S-TENG is higher than that of theplain device because of the enlarged contact area. Furthermore,according to the same principle, the electrical potential of theribbed S-TENG reached 50.6 V at the 30% strain, which issuperior to those of the other S-TENGs. Such an increase ofthe rib-fabric electrical potential is due to the increased totaltriboelectric charge that was generated, which is itself due to thecombined enhanced surface at rest and the increased contactarea under stretching. Both these attributes were derived fromthe particular knitted structure of the ribbed fabric given itssuperior performances.The measured output voltage (with a load of 40 MΩ) and

current (with a load of 100 Ω) of the S-TENGs under thestretching conditions are plotted in Figure 4c and Figure S6.The results show that the plain and double S-TENGs producedan output voltage and a current peak of 2.00 V and 0.09 μA and12.25 V and 0.53 μA, respectively. Interestingly, the use of therib structure led to a much higher increase of the output up to23.50 V and 1.05 μA compared with the double fabric. Theseexperiment results show a trend that is in sound agreementwith the previous simulated data. Therefore, the selection of theknitted structure of the fabric highly boosted the generatedtotal triboelectric charges and the output voltage byapproximately 1170%, which is critical for the achievement ofthe design of a high-performance wearable and stretchableTENG.Until now, most of the previous research studies on TENGs

represented the electrical-output performances as the maximumvoltage and current peaks.2,25,41 However, it is difficult toquantify the effective power generation of TENGs because itvaries over time and in terms of the mechanical stimulus. Here,the RMS value was used to describe the average constant powerthat was generated by our devices. The calculated RMS outputvoltage and current with respect to the knitted structures andstrain are plotted in Figures 5a and 5b. All of the experimentedknitted fabrics showed an increasing RMS output voltage andcurrent as the strain was increased, which is in sound agreementwith the increasing trend of the peak-to-peak outputs (FigureS7). The plain- and double-knitted-fabric S-TENG devicesgenerated an RMS output voltage and a current up to 0.90 Vand 0.07 μA and 2.40 V and 0.20 μA, respectively. Concerningthe rib-knitted-fabric S-TENG device, the RMS-output voltageand the current are higher, with respective values of 5.30 V and0.29 μA, confirming that the rib S-TENG is an excellentcandidate for an efficient knitting structure for wearablemechanical-energy harvesting systems.Human motion occurs in various ways through combining,

stretching, and compressions; therefore, apart from thecharacterization of the S-TENG device under the stretchingmode in the transverse direction, additional measurements ofthe electrical performances were carried out in the verticalcontact and separation modes (Figure 5c and Figure S8). Toinvestigate the reliance of the output power on the externalload, the RMS output voltage and the current of the rib S-TENG were systematically measured under different resistanceloads from 104 Ω to 109 Ω. The optimum impedance thatmaximized the power that was generated by the S-TENG isapproximately 5 MΩ, and a maximum average RMS power ofapproximately 60 μW was measured in the contact−separationmode, as shown in Figure 5d. Moreover, to demonstrate thefeasibility of the fabricated S-TENG as a self-powered energysource, commercial light-emitting diodes (LEDs) wereintegrated with sportswear in the form of a power-supply

cloth. Here, 15 green LEDs were instantly illuminated in aseries by a 10 × 10 cm2 device during running, therebyillustrating the potential of human-motion energy harvesting forpowering wearable electronics (Figure S9 and video S1).

CONCLUSIONIn the present study, a highly stretchable and wearable knitted-fabric-based S-TENG has been demonstrated using industrialknitting techniques for the harvesting of biomechanical energyfrom human-body movements. Double-knitted fabric and rib-knitted fabric showed a much superior stretchability, with thelatter showing an even greater superiority, and they can endurehigher transversal strains up to 30%. The selection of theknitted structure of the fabric highly boosted the generatedtotal triboelectric charges and the output voltage byapproximately 1170%, which is critical for realizing a high-performance wearable and stretchable TENG. Under a 1.7-Hzstretching movement, the S-TENG produced an RMS voltageof 5.30 V and an RMS current of 0.29 μA. Under compression,a constant average power of 60 μW could be generated,confirming that the rib S-TENG is an excellent candidate forwearable mechanical-energy harvesting systems. In the nearfuture, it is expected that an integration of the proposed devicewith a full cloth will increase the generated power, leading to arealization of the powering of cloth-equipped wearableelectronic devices through a utilization of continuous human-body movements.

METHODSFabrication. The PTFE (dtex 220, Swicofil Co.) and Ag (280

denier, X-Silver Co.) were knitted into plain-, double-, and rib-patterned fabrics with a gauge of 12 (needle/inch) on an industrialknitting machine. The structure of the S-TENG is a double-arc shapeof a 10 × 10 cm size. It consists of knitted PTFE fabrics for the topand bottom triboelectric layers, and knitted Ag fabrics are used for theelectrode in the middle and on the back of the top and bottomtriboelectric layers. Each PTFE knitted fabric and Ag knitted fabric oftop/bottom layers is attached by sewing at regular intervals in adirection perpendicular to the stretching direction in order not tointerfere with stretching deformation. The width of middle layer isslightly shorter than that of the top and bottom layers, and both endsof the three layers are fixed by sewing so that top and bottom layersare slightly bent to have an arc shape.

Characterization. The Dead Pixel Test MFC Application software(version 1.0, Michael Salzlechner) was used to measure the density ofthe PTFE yarns from photo images. The voltage signals weremeasured and recorded using a Tektronix DPO3052 (Tektronix)oscilloscope with a Tektronix P5100A (Tektronix) voltage probe witha 40-MΩ input impedance. The current signals were measured usingthe SR570 low-noise current amplifier (Stanford Research) with aninput impedance of 100 Ω at a caliber of 1 μA/V.

ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information is available free of charge on theACS Publications website at DOI: 10.1021/acsnano.7b05203.

Additional details include the triboelectric series of thecommon materials in the textile industry and the typicalmetals; photo images of the S-TENG fabrics duringstretching from 0 to 30% strains in the transversedirection; a schematic of the enlarged-yarn contact-surface calculation; the contact area of the S-TENGfabrics depending on the strain; the S-TENG maximumcurrent peak when it is stretched; the increasing output

ACS Nano Article

DOI: 10.1021/acsnano.7b05203ACS Nano XXXX, XXX, XXX−XXX

G

tendency of the rib-fabric S-TENG versus the strain; theoutput electrical performances of the S-TENG under thevertical contact and the separation actuations; and asnapshot of the 15 lit-up commercial green LEDs of thesportswear-integrated S-TENG on a runner (PDF)Movie showing power generation from a running motion(AVI)

AUTHOR INFORMATIONCorresponding Author*E-mail: kimsw1@skku.edu.

ORCIDSung Soo Kwak: 0000-0002-2625-2471Han Kim: 0000-0003-1947-0398Wanchul Seung: 0000-0002-7411-0865Jihye Kim: 0000-0001-8035-8640Ronan Hinchet: 0000-0003-3356-8930Sang-Woo Kim: 0000-0002-0079-5806Author Contributions⊥S.S.K. and H.K. contributed equally to this work.

NotesThe authors declare no competing financial interest.

ACKNOWLEDGMENTSThis research was financially supported by the Center forAdvanced Soft-Electronics as the Global Frontier Project(2013M3A6A5073177) and the Basic Science ResearchProgram (2015R1A2A1A05001851) through the NationalResearch Foundation (NRF) of Korea Grant funded by theMinistry of Science, ICT & Future Planning, and the “HumanResources Program in Energy Technology” of the KoreaInstitute of Energy Technology Evaluation and Planning(KETEP) and granted financial resources from the Ministryof Trade, Industry & Energy, Republic of Korea (No.20174030201800).

REFERENCES(1) Wu, H.; Huang, Y.; Xu, F.; Duan, Y.; Yin, Z. Energy Harvestersfor Wearable and Stretchable Electronics: From Flexibility toStretchability. Adv. Mater. 2016, 28, 9881−9919.(2) Yang, Y.; Zhang, H.; Lin, Z. H.; Zhou, Y. S.; Jing, Q.; Su, Y.; Yang,J.; Chen, J.; Hu, C.; Wang, Z. L. Human Skin Based TriboelectricNanogenerators for Harvesting Biomechanical Energy and as Self-Powered Active Tactile Sensor System. ACS Nano 2013, 7, 9213−9222.(3) Seung, W.; Gupta, M. K.; Lee, K. Y.; Shin, K.-S.; Lee, J.-H.; Kim,T. Y.; Kim, S.; Lin, J.; Kim, J. H.; Kim, S.-W. Nanopatterned Textile-Based Wearable Triboelectric Nanogenerator. ACS Nano 2015, 9,3501−3509.(4) Fan, F.-R.; Tian, Z.-Q.; Lin Wang, Z. Flexible TriboelectricGenerator. Nano Energy 2012, 1, 328−334.(5) Kim, S.; Gupta, M. K.; Lee, K. Y.; Sohn, A.; Kim, T. Y.; Shin, K.-S.; Kim, D.; Kim, S. K.; Lee, K. H.; Shin, H.-J.; Kim, D.-W.; Kim, S.-W.Transparent Flexible Graphene Triboelectric Nanogenerators. Adv.Mater. 2014, 26, 3918−3925.(6) Seung, W.; Yoon, H.-J.; Kim, T. Y.; Ryu, H.; Kim, J.; Lee, J.-H.;Lee, J. H.; Kim, S.; Park, Y. K.; Park, Y. J.; Kim, S.-W. Boosting Power-Generating Performance of Triboelectric Nanogenerators via ArtificialControl of Ferroelectric Polarization and Dielectric Properties. Adv.Energy Mater. 2017, 7, 1600988.(7) Wang, S.; Lin, L.; Wang, Z. L. Triboelectric Nanogenerators asSelf-Powered Active Sensors. Nano Energy 2015, 11, 436−462.

(8) Wang, Z. L.; Chen, J.; Lin, L. Progress in TriboelectricNanogenerators as a New Energy Technology and Self-PoweredSensors. Energy Environ. Sci. 2015, 8, 2250−2282.(9) Hinchet, R.; Seung, W.; Kim, S.-W. Recent Progress on FlexibleTriboelectric Nanogenerators for Selfpowered Electronics. ChemSu-sChem 2015, 8, 2327−2344.(10) Wang, S.; Zi, Y.; Zhou, Y. S.; Li, S.; Fan, F.; Lin, L.; Wang, Z. L.Molecular Surface Functionalization to Enhance the Power Output ofTriboelectric Nanogenerators. J. Mater. Chem. A 2016, 4, 3728−3734.(11) Choi, H. J.; Lee, J. H.; Jun, J.; Kim, T. Y.; Kim, S. W.; Lee, H.High-Performance Triboelectric Nanogenerators with Artificially Well-Tailored Interlocked Interfaces. Nano Energy 2016, 27, 595−601.(12) Rogers, J. A.; Someya, T.; Huang, Y. Materials and Mechanicsfor Stretchable Electronics. Science (Washington, DC, U. S.) 2010, 327,1603−1607.(13) Lee, P.; Lee, J.; Lee, H.; Yeo, J.; Hong, S.; Nam, K. H.; Lee, D.;Lee, S. S.; Ko, S. H. Highly Stretchable and Highly Conductive MetalElectrode by Very Long Metal Nanowire Percolation Network. Adv.Mater. 2012, 24, 3326−3332.(14) Zhong, J.; Zhang, Y.; Zhong, Q.; Hu, Q.; Hu, B.; Wang, Z. L.;Zhou, J. Fiber-Based Generator for Wearable Electronics and MobileMedication. ACS Nano 2014, 8, 6273−6280.(15) Zhou, T.; Zhang, C.; Han, C. B.; Fan, F. R.; Tang, W.; Wang, Z.L. Woven Structured Triboelectric Nanogenerator for WearableDevices. ACS Appl. Mater. Interfaces 2014, 6, 14695−14701.(16) Pu, X.; Li, L.; Song, H.; Du, C.; Zhao, Z.; Jiang, C.; Cao, G.; Hu,W.; Wang, Z. L. A Self-Charging Power Unit by Integration of aTextile Triboelectric Nanogenerator and a Flexible Lithium-IonBattery for Wearable Electronics. Adv. Mater. 2015, 27, 2472.(17) MATKOVIC, V. M. P. The Power of Fashion: The Influence ofKnitting Design on the Development of Knitting Technology. Text. J.Cloth Cult. 2010, 8, 122−147.(18) Li, L.; Au, W. M.; Li, Y.; Wan, K. M.; Chung, W. Y.; Wong, K. S.A Novel Design Method for an Intelligent Clothing Based on GarmentDesign and Knitting Technology. Text. Res. J. 2009, 79, 1670−1679.(19) Pritchard, M.; Sarsby, R. W.; Anand, S. C. Textiles in CivilEngineering. Part 2 − Natural Fibre Geotextiles. In Handbook ofTechnical Textiles; Elsevier, 2000; pp 372−406.(20) Cohen, A. E. Carbon Nanotubes Provide a Charge. Science(Washington, DC, U. S.) 2003, 300, 1235−1236.(21) Lin, Z.-H.; Cheng, G.; Lin, L.; Lee, S.; Wang, Z. L. Water-SolidSurface Contact Electrification and Its Use for Harvesting Liquid-Wave Energy. Angew. Chem. 2013, 125, 12777−12781.(22) Lee, K. H.; Lee, J. Y.; Kwon, Y. H.; Kang, T. W.; You, J. H.; Lee,D. U.; Kim, T. Effects of Defects on the Morphologies of GaNNanorods Grown on Si (111) Substrates. J. Mater. Res. 2009, 24,3032−3037.(23) Yu, Y.; Wang, X. Chemical Modification of Polymer Surfaces forAdvanced Triboelectric Nanogenerator Development. Extrem. Mech.Lett. 2016, 9, 514−530.(24) He, X.; Zi, Y.; Guo, H.; Zheng, H.; Xi, Y.; Wu, C.; Wang, J.;Zhang, W.; Lu, C.; Wang, Z. L. A Higly Stretchable Fiber-BasedTriboelectric Nanogenerator for Self-Powered Wearable Electronics.Adv. Funct. Mater. 2017, 27, 1604378.(25) Wang, J.; Li, S.; Yi, F.; Zi, Y.; Lin, J.; Wang, X.; Xu, Y.; Wang, Z.L. Sustainably Powering Wearable Electronics Solely by BiomechanicalEnergy. Nat. Commun. 2016, 7, 12744.(26) Henniker, J. Triboelectricity in Polymers. Nature 1962, 196,474−474.(27) Elsdon, R.; Mitchell, F. R. G. Contact Electrification ofPolymers. J. Phys. D: Appl. Phys. 1976, 9, 1445−1460.(28) Lin, Z. H.; Cheng, G.; Lee, S.; Pradel, K. C.; Wang, Z. L.Harvesting Water Drop Energy by a Sequential Contact-Electrificationand Electrostatic-Induction Process. Adv. Mater. 2014, 26, 4690−4696.(29) Kwak, S. S.; Lin, S.; Lee, J. H.; Ryu, H.; Kim, T. Y.; Zhong, H.;Chen, H.; Kim, S.-W. Triboelectrification-Induced Large ElectricPower Generation from a Single Moving Droplet on Graphene/Polytetrafluoroethylene. ACS Nano 2016, 10, 7297−7302.

ACS Nano Article

DOI: 10.1021/acsnano.7b05203ACS Nano XXXX, XXX, XXX−XXX

H

(30) Liang, Q.; Yan, X.; Gu, Y.; Zhang, K.; Liang, M.; Lu, S.; Zheng,X.; Zhang, Y. Highly Transparent Triboelectric Nanogenerator forHarvesting Water-Related Energy Reinforced by AntireflectionCoating. Sci. Rep. 2015, 5, 9080.(31) Kang, H.; Kim, H.; Kim, S.; Shin, H. J.; Cheon, S.; Huh, J.-H.;Lee, D. Y.; Lee, S.; Kim, S.-W.; Cho, J. H. Mechanically Robust SilverNanowires Network for Triboelectric Nanogenerators. Adv. Funct.Mater. 2016, 26, 7717−7724.(32) Semnani, D. Mechanical Properties of Weft Knitted Fabrics inFully Stretched Status along Courses Direction: Geometrical ModelAspect. Univers. J. Mech. Eng. 2013, 1, 62−67.(33) Kaldor, J. M.; James, D. L.; Marschner, S. Simulating KnittedCloth at the Yarn Level. In ACM SIGGRAPH 2008 papers on -SIGGRAPH ’08; ACM Press: New York, 2008; p 1.(34) Zhu, G.; Chen, J.; Zhang, T.; Jing, Q.; Wang, Z. L. Radial-Arrayed Rotary Electrification for High Performance TriboelectricGenerator. Nat. Commun. 2014, 5, 3426.(35) Baytekin, H. T.; Patashinski, A. Z.; Branicki, M.; Baytekin, B.;Soh, S.; Grzybowski, B. A. The Mosaic of Surface Charge in ContactElectrification. Science (Washington, DC, U. S.) 2011, 333, 308−312.(36) Zhou, Y. S.; Liu, Y.; Zhu, G.; Lin, Z.-H.; Pan, C.; Jing, Q.; Wang,Z. L. In Situ Quantitative Study of Nanoscale Triboelectrification andPatterning. Nano Lett. 2013, 13, 2771−2776.(37) Wang, S.; Lin, L.; Wang, Z. L. Nanoscale Triboelectric-Effect-Enabled Energy Conversion for Sustainably Powering PortableElectronics. Nano Lett. 2012, 12, 6339−6346.(38) Niu, S.; Wang, S.; Lin, L.; Liu, Y.; Zhou, Y. S.; Hu, Y.; Wang, Z.L. Theoretical Study of Contact-Mode Triboelectric Nanogeneratorsas an Effective Power Source. Energy Environ. Sci. 2013, 6, 3576.(39) Lee, K. Y.; Chun, J.; Lee, J.-H.; Kim, K. N.; Kang, N.-R.; Kim, J.-Y.; Kim, M. H.; Shin, K.-S.; Gupta, M. K.; Baik, J. M.; Kim, S.-W.Hydrophobic Sponge Structure-Based Triboelectric Nanogenerator.Adv. Mater. 2014, 26, 5037−5042.(40) Zhu, G.; Bai, P.; Chen, J.; Lin Wang, Z. Power-Generating ShoeInsole Based on Triboelectric Nanogenerators for Self-PoweredConsumer Electronics. Nano Energy 2013, 2, 688−692.(41) Lai, Y.-C.; Deng, J.; Niu, S.; Peng, W.; Wu, C.; Liu, R.; Wen, Z.;Wang, Z. L. Electric Eel-Skin-Inspired Mechanically Durable andSuper-Stretchable Nanogenerator for Deformable Power Source andFully Autonomous Conformable Electronic-Skin Applications. Adv.Mater. 2016, 28, 10024−10032.

ACS Nano Article

DOI: 10.1021/acsnano.7b05203ACS Nano XXXX, XXX, XXX−XXX

I