Ho

WD

a

ARRAA

KZNSGP

1

tds(htsuNpeatiinms

n

0h

Sensors and Actuators A 205 (2014) 164– 169

Contents lists available at ScienceDirect

Sensors and Actuators A: Physical

jo u r n al homep age: www.elsev ier .com/ locate /sna

ighly sensitive and flexible strain sensors based on vertical zincxide nanowire arrays

engui Zhang, Ren Zhu, Vu Nguyen, Rusen Yang ∗

epartment of Mechanical Engineering, University of Minnesota, Minneapolis, MN 55455, USA

r t i c l e i n f o

rticle history:eceived 11 July 2013eceived in revised form 1 October 2013ccepted 2 November 2013vailable online 9 November 2013

eywords:

a b s t r a c t

In this paper, a highly sensitive strain sensor with vertically aligned zinc oxide (ZnO) nanowire arrayson polyethylene terephthalate (PET) film was reported. The device fabrication includes conventionalphotolithography, metallization, and ZnO nanowire growth through a hydrothermal method. I–V char-acteristics of the device were highly nonlinear due to the Schottky contact between the nanowire andthe gold (Au) electrode. The conductivity of the device is significantly tuned by the change of ZnO/AuSchottky barrier that reflects the strain-induced piezoelectric potential. A gauge factor up to 1813 was

nOanowire arraytrain sensorauge factoriezotronic effect

obtained from this strain senor, which is higher than the previously reported device based on a lateralZnO microwire. Theoretical analysis of the piezotronic effect shows that the working nanowire with thelargest conductivity change dominates the performance of the device. The non-working nanowire haslimited adverse effect on the performance, which explains the robust performance of this novel strainsensor. The stability and fast response of the sensor were also investigated. The sensitive and robust strainsensor is expected to find applications in civil, medical, and other fields.

. Introduction

Flexible or stretchable strain sensors based on nanowire, nano-ube or polymer composite have been widely studied in the lastecade for potential applications in portable and wearable per-onal devices [1–24]. Many nanomaterials like carbon nanotubeCNT) [1–12], graphene [13–18] and ZnO nanostructure [19–24]ave been investigated for the design of strain sensors. Comparedo conventional rigid strain sensors using metal or silicon, flexibletrain sensors based on nanomaterials exhibit high strain tolerance,ltra-fast response, high sensitivity, and low power consumption.anowires with piezoelectric and semiconducting properties arearticularly suitable for this purpose due to the recently discov-red piezotronic effect, in which the piezoelectric potential from

strained wire can dramatically tune the current flow throughhe nanowire by the change of Schottky barrier [23–27]. Flex-ble piezotronic strain sensor has been demonstrated based onndividual ZnO microwire, with a gauge factor of 1250 [22]. Theanowire-based sensor is expected to provide even better perfor-ance due to the much higher surface-volume ratio and significant

patial confinement.In this paper, a novel strain sensor based on vertical ZnO

anowires grown on PET substrate has been designed and

∗ Corresponding author. Tel.: +1 6126264318.E-mail address: yangr@me.umn.edu (R. Yang).

924-4247/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.sna.2013.11.004

© 2013 Elsevier B.V. All rights reserved.

fabricated, and the improved sensitivity, flexibility, robustness aswell as fast response have been examined. ZnO is an environmentfriendly material with outstanding piezoelectric and semiconduct-ing properties. ZnO nanowires can be achieved in high yield withphysical or chemical approaches [28–30], which facilitates futurescale up and massive production. It has been revealed that theI–V characteristics of the strain sensor are dominated by Schottkycontact at ZnO/Au interfaces and are thus highly nonlinear. TheSchottky barrier height is tuned by piezoelectric potential fromstressed ZnO nanowires when the strain sensor is deformed. Thestrain sensor is extremely sensitive to the local strain and a gaugefactor up to 1813 was observed. This gauge factor is higher thanthat of previous reported strain sensors with single ZnO microwirein lateral configuration [22]. In the Section 2 of this paper, the fab-rication procedure of the device and the measurement system areintroduced. In Section 3, the performance of the strain sensor istested. Finally, the performance as well as the piezotronic workingprinciple of the device is discussed.

2. Fabrication procedure and measurement setup

The fabrication procedure of the strain sensor and the measure-ment setup are shown in Fig. 1. A piece of PET substrate with a

length of ∼3 cm, a width of ∼1 cm, and a thickness of ∼200 �m wasrinsed with acetone, isopropyl alcohol, and deionized water andthen blow-dried with nitrogen. The dry and clean PET substratewas placed in a RF/DC sputtering system (AJA-Sputter, 200 W), in

W. Zhang et al. / Sensors and Actuators A 205 (2014) 164– 169 165

F a) Det ) Spint of me

w(loegforwtwstsP(tttFoos

sarmrdaaa

ig. 1. Schematic fabrication process of the strain sensor and measuring system. (hrough photolithography. (c) Grow ZnO nanowire with hydrothermal solution. (dips. (e) Deposit Au as top electrode. (f) Optical image of the final device. (g) Sketch

hich a bonding layer of Cr (∼40 nm) and a seed layer of ZnO∼200 nm) were deposited in sequence, as shown in Fig. 1(a). Ship-ey S1813 photoresist was spin-coated (∼2 �m, 3000 rpm for 30 s)n the substrate and a pattern of 2 × 2 windows (50 �m × 50 �mach) was opened in Fig. 1(b) through photolithography for therowth of ZnO nanowire arrays in Fig. 1(c). In order to obtain uni-orm, long and high quality of ZnO nanowires, the growth solutionf 16 mmol/L hexamethylenetetramine and 16 mmol/L zinc chlo-ide was mixed with 4% volume concentration of ammonia (30%/w NH3). The substrate was floated upside down on the surface of

he solution. After the growth at 95 ◦C for 14 h, ZnO nanowire arraysere obtained in the 2 × 2 windows. A thin layer of polydimethyl-

iloxane (PDMS, ∼20 �m, 3000 rpm for 2 min) was spin-coated onhe surface of the PET substrate and cured at 80 ◦C for 2 h, whicherved as a protection layer for the nanowire arrays. Top part of theDMS layer was then etched away in a reactive-ion etching systemSTS Dry Etcher, with O2:CF4 = 1:3 for 15 min), which exposed theip of nanowires in Fig. 1(d) for electrode connection in Fig. 1(e). Ahin layer of Au (∼120 nm) was deposited by a RF/DC sputtering sys-em to serve as the top electrode and form ZnO/Au Schottky barrier.inally, the strain sensor was packaged by spin-coating a thin layerf poly(methyl methacrylate) (∼2 �m, NANOTM950, MicroChem)n the surface. The final device was compact and highly flexible, ashown in Fig. 1(f).

The setup in Fig. 1(g) was used to test the performance of thetrain sensor. One end of the strain sensor was fixed tightly on

sample holder while the other end was attached to a movingod driven by a programmable linear motor. During the measure-ent, the linear motor was programmed to push/pull the lateral

od with specified distance, acceleration, maximum speed and

eceleration, so that the strain sensor was bent back and forth in

well-controlled manner. A sinusoidal bias voltage was appliedcross the nanowires with a functional generator (Keithley 3390)nd the current was monitored with a current amplifier (Keithley

posit Cr bonding layer and ZnO seedlayer on PET substrate. (b) Open the window-coat PDMS protection layer for ZnO nanowire and dry etch to exposure nanowireasuring system that includes sample holder and linear motor with moving rod.

428). The measurement was carried out in atmosphere at roomtemperature. The strain sensor was located in a home-built Fara-day cage so that the environmental influence and electromagneticnoises were excluded.

The performance of strain sensor largely depends on the qualityof ZnO nanowire arrays. Fig. 2 shows typical ZnO nanowire arraysfrom hydrothermal growth. Fig. 2(a) confirmed that ZnO nanowirearrays grew uniformly within designed pattern with a length of∼10 �m and a diameter of ∼100 nm. The PDMS layer was spin-coated on the substrate and then partially etched to expose thetip of the ZnO nanowire for the electrode deposition, as shownin Fig. 2(b). After etching, ZnO nanowires in Fig. 2(b) were intactand surrounded by PDMS. The distorted and very bright area inFig. 2(b) is due to the common charging effect in the scanningelectron microscopy.

3. Experimental result

The typical I–V characteristics of the strain sensor and itsresponse to the strain are shown in Fig. 3. The nonlinear and rec-tifying behavior in Fig. 3(a) is due to the Schottky contact formedbetween the ZnO nanowire and the Au electrode. Fig. 3(a) also indi-cates that the I–V characteristic is dramatically changed when thedevices is stressed. Because the dimension of the nanowire is muchsmaller than PET and PDMS is much thinner (15 �m) and morecompliant (360–870 kPa) than the PET (200 �m and 2–2.7 GPa), PETdominates the elastic behavior of the device. The longitudinal nor-mal strain at the surface of the sensor was estimated from equation

ε = 3aD

l2(1 − x

l) (1)

Here, a is the half thickness of the PET substrate (∼100 �m), D isthe maximum bending deflection of the device, l is the length from

166 W. Zhang et al. / Sensors and Actuators A 205 (2014) 164– 169

Fig. 2. SEM image of vertical ZnO nanowires from hydrothermal growth. (a) Topv(e

fie

ftdnbsrst

s

wbcf1ria

Ttwrro

Fig. 3. Electromechanical behaviors of the strain sensor. (a) Significant change ofnonlinear I–V curve in response to different strain. (b) Gauge factor at 1.5 V bias. (c)Linear dependence of logrithmic current on the strain at different bias of 2 V, 1.5 V,

electrode connected to the continuous ZnO seed layer. The I–Vcharacteristic of this metal-semiconductor-metal structure underforward bias, V > 3kT/q ∼ 77mV, is given by [26]

iew of the as-grown nanowire. Inset: cross-section view of the nanowire arrays.b) Top view of the nanowire after PDMS protection coating and dry etching toxpose nanowire tips.

xed end to free end, and x is the average distance from the fixednd.

Because of the Poisson effect, the normal strain along the sur-ace of the PET induced the strain along the axial direction ofhe nanowire that grew perpendicular to the surface, and causedramatic current change. The conductivity of the compressedanowire was significantly increased, especially when a forwardias was applied. For example, at the bias of +2 V, the current intrain-free nanowires was about 0.02 �A. As the normal strain εeached 0.4%, the current increased to 0.08 �A. When the strainensor was further pushed forward until the strain reached 0.8%,he current reached 0.26 �A. The performance of the strain sen-

or can be characterized with the strain gauge factor, GF =∣∣∣�I/I0

ε

∣∣∣,here I0 is the strain-free current of the device at a fixed forward

ias, and ε and �I are the strain and the corresponding currenthange at the same bias. With a forward bias of 1.5 V, the gaugeactor of the strain sensor is plotted in Fig. 3(b) and is as high as813 at a normal strain ε = 0.6%. In Fig. 3(c), the logarithmic cur-ent is plot against the strain at different biases. A linear behaviors revealed and the slope is roughly constant for different biases,nd the mechanism is discussed in next section.

The stability and response of the sensor is examined in Fig. 4.he “Stress” state and “Release” state in Fig. 4 correspond tohe state when the device was mechanically bent and the state

hen the device returned to its straight and strain-free condition,

espectively. It is worthwhile pointing out that, the current alwayseached the similar value at “Stress” state and went back to itsriginal current level at “Release” state when the substrate was

and 1 V.

cyclically bent under different frequencies, which demonstratesthe stability of the strain sensor. The instant response to 2.5 Hzexcitation indicates a quick switch of the strain sensor.

The fast response of the strain sensor with ZnO nanowires wasfurther studied when the device was deformed with different strainrates. The current of the strain sensor was under real time monitoras the linear motor pushed the free end of the substrate at differentspeeds from 0.001 m/s to 1 m/s to the strain ε = 0.8%. The observedcurrent fluctuation at the beginning and end of the deformation wasascribed to the instability of the acceleration and the deceleration ofthe linear motor. However, even for the fast speed like v = 1 m/s, thestrain sensor responded very well to the mechanical deformationwith a response time below 100 ms.

4. Discussion and explanations

The high sensitivity of the strain sensor in this study canbe ascribed to the piezotronic effect in the piezoelectric andsemiconducting nanowires [25,26]. To explain the piezotroniceffect, the sensor can be simplified as a one dimensional metal-semiconductor-metal structure. Schottky contact forms at oneterminal between the top gold electrode and the tip of nanowires,while Ohmic contact at the other terminal is realized with Cr

I = SA∗∗T2 exp

(−q�B0

kT

)exp(

q(��E + V)kT

)(2)

W. Zhang et al. / Sensors and Actuators A 205 (2014) 164– 169 167

F or at fia

w3

I

w

tetdifi

gd[

�

wεtscitttcoF

t

GF =∣∣∣∣�I/I0

ε

∣∣∣∣ =∣∣∣∣∣exp(− q��B(ε)

kT ) − 1

ε

∣∣∣∣∣ (5)

ig. 4. Switching between “Stress” (ε = 0.8%) and “Release” states of the strain sensnd (d) 2.5 Hz.

hile I–V characteristic in the reverse direction with∣∣V∣∣>

kT/q∼77mV is

= SA∗∗T2 exp(

−q�B0

kT

)exp

(q√

q�/4��

kT

)(3)

here � =√

2qND�

(V + Vbi − kT

q

), S is the contact area, A** is

he effective Richardson constant, T is the temperature, q is thelectron charge, �B0 is the Schottky barrier height at zero elec-ric field, Vbi is the built-in potential at the barrier, ND is theonor impurity density, � is the permittivity of ZnO and �E

s the lowering of Schottky barrier under an external electriceld.

When ZnO nanowire is compressed, piezoelectric potential isenerated along its length as shown in Fig. 5(a). This potentialecreases the Schottky barrier (Fig. 5(b)) in Eq. (2) in the form of25]

�B =q�piezoW2

piezo

2εs(4)

here Wpiezo is the width of the polarization layer on the interface,s is the permittivity of ZnO, and �piezo is the density of polariza-ion charges (in units of electron charge) that is proportional to thetrain ε along nanowire c-axis. Consequently, the Schottky barrierhange, ��B, is proportional to strain ε. ��B < 0 when the nanowires compressed because of the strain-induced negative charge onhe ZnO/Au interface and ��B > 0 when nanowire is stretched dueo the strain-induced positive charge. Because is proportional tohe strain, it is clear from Eqs. (2) and (4) that the logarithmicurrent is proportional to the strain and the slope is independent

f the applied bias, which agrees well with our observation inig. 3(c).

The change of the Schottky barrier due to the piezoelec-ric potential, ��B, modulates the current through the nanowire

xed bias of +5 V when it is cyclically bent with at (a) 0.125 Hz, (b) 0.25 Hz, (c) 1 Hz,

(Fig. 5(c)) from Eq. (2). As a result, the strain gauge factor, GF, of thestrain sensor can be expressed as

Fig. 5. Explanation of piezotronic working principle. (a) “Original” and “Com-pressed” states of ZnO nanowire. (b) Corresponding energy band diagram at ZnO/Auinterface on two states. (c) Correpsonding I–V characteristics on two states.

1 Actua

Te

hfadroa

G

Hb

fWnttwtpitsmdi

mStsdtnbHpStft

5

ofsTarfdat

A

t

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

68 W. Zhang et al. / Sensors and

he exponential dependence of the current on the Schottky barrierxplains the high gauge factor observed in the strain sensor.

Considering the current of nanowires under stress is muchigher than the stress-free current and the nanowires are very uni-

orm, we assume that the stress-free conductivity are the same forll nanowires. The stress-induced current change is different forifferent nanowires because the Schottky barrier and thus the cur-ent are very sensitive to the local strain that can vary slightly fromne nanowire to another. Therefore, the gauge factor is estimateds

F =

∣∣∣∣∣∣∣

∑�Ii∑Ii0

ε

∣∣∣∣∣∣∣ =∣∣∣∣∣∑

exp(− q��i(ε)kT ) − N

εN

∣∣∣∣∣ (6)

ere, N represents the number of nanowires, ��i (ε) is the Schottkyarrier change in the i-th nanowire.

Because of the variation in the nanowire growth and the deviceabrication, all nanowires had not the same response to the stress.

hen the device in Fig. 1(d) was bent to the left such that theanowire was experiencing compressive strain along the length,he current in some nanowires was significantly increased due tohe Schottky barrier drop, while other nanowires might not workell and did not change much. Exponential term in Eq. (6) indicates

hat the nanowires with dramatic current increase dominated theerformance, which agreed with the high sensitivity we observed

n Fig. 3. In comparison, when the device was bent to the right suchhat the nanowire was elongated, the current of the most respon-ive nanowires drops significantly, but the irresponsive nanowiresaintaining at a relatively higher current level dominated the

evice’s current. Therefore, the total change of current in Eq. (6)s insignificant, resulting in very low sensitivity in this case.

Quality of Schottky barrier significantly influences the perfor-ance of the strain sensor. In this study, Au was selected to form

chottky contact with ZnO, which has a higher work function thanhe silver used in the previously reported single-nanowire strainensor [22,31]. Oxygen plasma clean for the tip of ZnO nanowireuring dry etching process improved the quality of Schottky con-act [32]. Size of the nanowire array determines the number ofanowires. Compared to the exponential term in Eq. (6), the num-er of nanowires N is a minor factor and can usually be neglected.owever, when there are more nanowires in a larger designedattern, the variance of nanowires make it difficult to achievechottky contact on every nanowire. For this reason, the size ofhe ZnO nanowire array as well as growth process are controlledor improved Schottky contact and such optimization is expectedo further enhance the device’s performance.

. Conclusion

In summary, we have developed a novel strain sensor basedn vertical ZnO nanowire arrays grown on a flexible PET substraterom a hydrothermal process. The strain sensor is extremely sen-itive and a strain gauge factor up to 1813 has been demonstrated.he high sensitivity is ascribed to the high quality ZnO nanowirerrays and the piezotronic effect. In addition, the stability and fastesponse of the device have been observed. The excellent per-ormances of the strain sensor promote further applications forelicate and sensitive force, strain and stress measurement in manyreas, such as tissue/cell examination, smart skin design, and struc-ural health monitoring.

cknowledgements

The authors are truly grateful for the financial support fromhe Department of Mechanical Engineering and the College of

[

[

tors A 205 (2014) 164– 169

Science and Engineering of the University of Minnesota. Theauthors would like to thank the National Science Foundation forits financial support (ECCS-1150147). Parts of this work were car-ried out in Characterization Facility, University of Minnesota, whichreceives partial support from NSF through the MRSEC program.

References

[1] N.K. Chang, C.C. Su, S.H. Chang, Fabrication of single-walled carbon nanotubeflexible strain sensors with high sensitivity, Applied Physics Letters 92 (2008).

[2] P. Dharap, Z.L. Li, S. Nagarajaiah, E.V. Barrera, Nanotube film based on single-wall carbon nanotubes for strain sensing, Nanotechnology 15 (2004) 379–382.

[3] C. Hierold, A. Jungen, C. Stampfer, T. Helbling, Nano electromechanical sensorsbased on carbon nanotubes, Sensors and Actuators A – Physical 136 (2007)51–61.

[4] N. Hu, Y. Karube, M. Arai, T. Watanabe, C. Yan, Y. Li, et al., Investigation onsensitivity of a polymer/carbon nanotube composite strain sensor, Carbon 48(2010) 680–687.

[5] N. Hu, Y. Karube, C. Yan, Z. Masuda, H. Fukunaga, Tunneling effect in a poly-mer/carbon nanotube nanocomposite strain sensor, Acta Materialia 56 (2008)2929–2936.

[6] I.P. Kang, M.J. Schulz, J.H. Kim, V. Shanov, D.L. Shi, A carbon nanotube strainsensor for structural health monitoring, Smart Materials & Structures 15 (2006)737–748.

[7] K.J. Loh, J.P. Lynch, B.S. Shim, N.A. Kotov, Tailoring piezoresistive sensitivityof multilayer carbon nanotube composite strain sensors, Journal of IntelligentMaterial Systems and Structures 19 (2008) 747–764.

[8] H. Maune, M. Bockrath, Elastomeric carbon nanotube circuits for local strainsensing, Applied Physics Letters 89 (2006).

[9] M. Park, H. Kim, J.P. Youngblood, Strain-dependent electrical resistance ofmulti-walled carbon nanotube/polymer composite films, Nanotechnology 19(2008).

10] G.T. Pham, Y.B. Park, Z. Liang, C. Zhang, B. Wang, Processing and modeling ofconductive thermoplastic/carbon nanotube films for strain sensing, Compos-ites Part B – Engineering 39 (2008) 209–216.

11] C. Stampfer, T. Helbling, D. Obergfell, B. Schoberle, M.K. Tripp, A. Jungen, et al.,Fabrication of single-walled carbon-nanotube-based pressure sensors, NanoLetters 6 (2006) 233–237.

12] L. Valentini, C. Cantalini, I. Armentano, J.M. Kenny, L. Lozzi, S. Santucci, Highlysensitive and selective sensors based on carbon nanotubes thin films for molec-ular detection, Diamond and Related Materials 13 (2004) 1301–1305.

13] S.H. Bae, Y. Lee, B.K. Sharma, H.J. Lee, J.H. Kim, J.H. Ahn, Graphene-based trans-parent strain sensor, Carbon 51 (2013) 236–242.

14] X. Chen, X.H. Zheng, J.K. Kim, X.X. Li, D.W. Lee, Investigation of graphenepiezoresistors for use as strain gauge sensors, Journal of Vacuum Science &Technology B 29 (2011).

15] V. Eswaraiah, K. Balasubramaniam, S. Ramaprabhu, Functionalized graphenereinforced thermoplastic nanocomposites as strain sensors in structural healthmonitoring, Journal of Materials Chemistry 21 (2011) 12626–12628.

16] X. Li, R.J. Zhang, W.J. Yu, K.L. Wang, J.Q. Wei, D.H. Wu, et al., Stretchableand highly sensitive graphene-on-polymer strain sensors, Scientific Reports2 (2012).

17] A. Sakhaee-Pour, M.T. Ahmadian, A. Vafai, Potential application of single-layered graphene sheet as strain sensor, Solid State Communications 147(2008) 336–340.

18] Y. Wang, R. Yang, Z.W. Shi, L.C. Zhang, D.X. Shi, E. Wang, et al., Super-elasticgraphene ripples for flexible strain sensors, Acs Nano 5 (2011) 3645–3650.

19] X.D. Wang, J. Zhou, J.H. Song, J. Liu, N.S. Xu, Z.L. Wang, Piezoelectric field effecttransistor and nanoforce sensor based on a single ZnO nanowire, Nano Letters6 (2006) 2768–2772.

20] Z.L. Wang, Piezopotential gated nanowire devices: piezotronics and piezo-phototronics, Nano Today 5 (2010) 540–552.

21] X. Xiao, L.Y. Yuan, J.W. Zhong, T.P. Ding, Y. Liu, Z.X. Cai, et al., High-strain sen-sors based on ZnO nanowire/polystyrene hybridized flexible films, AdvancedMaterials 23 (2011), 5440-+.

22] J. Zhou, Y.D. Gu, P. Fei, W.J. Mai, Y.F. Gao, R.S. Yang, et al., Flexible piezotronicstrain sensor, Nano Letters 8 (2008) 3035–3040.

23] W.J. Mai, Z.W. Liang, L. Zhang, X. Yu, P.Y. Liu, H.M. Zhu, et al., Strain sensingmechanism of the fabricated ZnO nanowire-polymer composite strain sensors,Chemical Physics Letters 538 (2012) 99–101.

24] W. Mai, L. Zhang, Y. Gu, S. Huang, Z. Zhang, C. Lao, et al., Mechanical andelectrical characterization of semiconducting ZnO nanorings by direct nano-manipulation, Applied Physics Letters 101 (2012) 081910.

25] Y. Zhang, Y. Liu, Z.L. Wang, Fundamental Theory of Piezotronics, AdvancedMaterials 23 (2011) 3004–3013.

26] Z.Y. Gao, J. Zhou, Y.D. Gu, P. Fei, Y. Hao, G. Bao, et al., Effects of piezoelectricpotential on the transport characteristics of metal-ZnO nanowire-metal field

effect transistor, Journal of Applied Physics 105 (2009).

27] Z.L. Wang, Piezotronic, Piezophototronic effects, Journal of Physical ChemistryLetters 1 (2010) 1388–1393.

28] L. Vayssieres, Growth of arrayed nanorods and nanowires of ZnO from aqueoussolutions, Advanced Materials 15 (2003) 464–466.

Actua

[

[

[

[

B

WoSiii

of Minnesota-Twin Cities as an assistant professor in 2010. He has discov-

W. Zhang et al. / Sensors and

29] M.H. Huang, Y.Y. Wu, H. Feick, N. Tran, E. Weber, P.D. Yang, Catalytic growthof zinc oxide nanowires by vapor transport, Advanced Materials 13 (2001)113–116.

30] S. Xu, N. Adiga, S. Ba, T. Dasgupta, C.F.J. Wu, Z.L. Wang, Optimizing, Improv-ing the growth quality of ZnO nanowire arrays guided by statistical design ofexperiments, Acs Nano 3 (2009) 1803–1812.

31] L.J. Brillson, Y.C. Lu, ZnO Schottky barriers and ohmic contacts, Journal ofApplied Physics 109 (2011).

32] B.J. Coppa, R.F. Davis, R.J. Nemanich, Gold Schottky contacts on oxygenplasma-treated, n-type ZnO(000(1)over-bar), Applied Physics Letters 82 (2003)400–402.

iographies

engui Zhang received his Master degree in Plasma Physics from the University

f Science and Technology of China (USTC), China, in 2010 and Bachelor degree inpace Physics from the University of Science and Technology of China (USTC), China,n 2004. He is currently a PhD candidate in University of Minnesota. His researchnterests are design and fabrication of novel piezotronic micro/nano-scale devices,ncluding mechanical sensors and gas/chemical sensors.

tors A 205 (2014) 164– 169 169

Ren Zhu is currently a PhD candidate in the department of mechanical engi-neering, University of Minnesota. He received his bachelor’s degree in mechanicalengineering and automation at Shanghai Jiao Tong University, China. During hisundergraduate study, he was awarded with National Scholarship and AcademicExcellence Scholarship. His research interests are piezoelectric nanogenerators, anovel technology for energy harvesting

Vu Nguyen received his B.S. degree in Mechanical Engineering from WorcesterPolytechnic Institute, Worcester, Massachusetts in 2012. He is currently pursu-ing Ph.D. degree at the University of Minnesota, Minneapolis, Minnesota. Hisresearch interests are energy harvesting and self-power systems at micro/nanoscale.

Rusen Yang received his PhD degree in Materials Science and Engineer-ing from Georgia Institute of Technology in 2007, where he continued asPost-Doctoral Associate. He joined Mechanical Engineering at the University

ered novel nanostructures, such as ZnO, SnO2, Zn3P2, and investigated theirapplication potentials. His most recent work on energy harvester based on piezo-electric nanomaterials made significant contribution in the field of renewableenergy.