Flexible piezoelectric energy nanogenerator based on ZnO nanotubes hosted in a polycarbonate membrane

Nano Energy (2015) 13, 474–481

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/nanoenergy

RAPID COMMUNICATION

Flexible piezoelectric energy nanogenerator based on ZnO nanotubes hosted in a polycarbonate membrane Stefano Stassia,1, Valentina Caudab,n,1, Carminna Ottonea,b, Angelica Chiodonib, Candido Fabrizio Pirria,b, Giancarlo Canavesea a

Department of Applied Science and Technology, Politecnico di Torino, Corso Duca degli Abruzzi 24, Torino 10129, Italy b Center for Space Human Robotics@PoliTo, Istituto Italiano di Tecnologia, Corso Trento 21, Torino 10129, Italy Received 9 January 2015; received in revised form 13 March 2015; accepted 15 March 2015 Available online 23 March 2015

KEYWORDS

Abstract

ZnO nanotube; Porous templating membrane; Nanogenerators; Nanoconfinement; Oriented crystallization

Highly oriented zinc oxide (ZnO) nanotubes were synthesized in a porous polycarbonate (PC) matrix, leading to a highly flexible ZnO–PC composite able to work as efficient energy nanogenerator. The crystalline direction of the ZnO c-axis is obtained parallel to the membrane surface, thus advantageous for the exploitation of composite under bending stresses. Three different pore sizes of the templating PC membranes were successfully employed, i.e., having nominal pore diameter 30, 50 and 100 nm, thus obtaining three different ZnO one-dimensional nanostructures supported into the PC membrane. The ZnO–PC nanogenerators were successfully tested both under compressive and bending strains, showing an influence of the ZnO nanotube size on the output voltage. Using the 100 nm pore PC membrane, a maximum output voltage of 1.15 V, a maximum current of 100 μA and a maximum output power density of 287.5 mW/cm3 were reached, being these values among the highest reached in zinc oxide-based piezoelectric nanogenerators. Such remarkable results make the nanostructured ZnO–PC composite a promising material for energy harvesting applications. & 2015 Elsevier Ltd. All rights reserved.

Introduction n

Corresponding author. Tel.: +39 011 0903436. E-mail address: [email protected] (V. Cauda). 1 These authors have equally contributed to the work. http://dx.doi.org/10.1016/j.nanoen.2015.03.024 2211-2855/& 2015 Elsevier Ltd. All rights reserved.

Zinc oxide (ZnO) is a prominent example of both piezoelectric and semiconductor material. The enhancement of these properties by exploiting ZnO nanostructuration spread

Flexible piezoelectric energy nanogenerator its application as functional material in a wide variety of different fields, including nanosensors [1–3], thermoelectric devices [4], third-generation solar cells [5–7], energy storage [8], electrochemical water splitting devices [9,10], and field-effect transistors [11,12]. Moreover ZnO nanowires (NWs) recently pioneered a new branch of research and devices called piezoelectric nanogenerators, widely studied in the last period [13–15]. Indeed the combination of the piezoelectric and semiconducting properties of zinc oxide enables to convert mechanical deformations into electric potential with the purpose to harvest energy from the surrounding environment. This concept was firstly introduced by Wang in 2006 [16] measuring the voltage generated deforming a single ZnO NW with a platinum-coated Atomic Force Microscopy (AFM) tip, and further broadly developed on arrays of self-standing ZnO NWs [13,17], ZnO nanocomposites [18,19] and several other ZnO nanostructures. Generally, the amount of converted energy is small, i.e., in the order of mW up to few mW, however enough for supplying MEMS sensors, implantable medical devices or even wearable electronics, up to charging mobile phone batteries. In general zinc oxide nanostructures and in particular nanowires (NWs) and nanotubes (NTs) can be grown using several different techniques, mostly leading to a high crystalline quality and an elevate reproducibility of the material. The most reported ones included chemical-vapor-deposition (CVD) [20,21], physical-vapor-deposition (PVD) [22], electrochemical deposition [23–25], hydrothermal synthesis [13,26– 28], and sol–gel templating approach [29–31]. In particular this last method is generally applied using anodic alumina membranes as hosting matrix [24,29,30,32], with the advantage of being very simple and high throughput. It leads however to the formation of polycrystalline NWs or NTs with low quality in terms of morphology. In addition, the corrosion of the alumina membrane during the chemical synthesis and its intrinsic fragility pose also not negligible problems [1]. To overcome these challenges, it reported the use of polycarbonate (PC) membranes as templating matrix, adopting, as synthesis technique for ZnO material, the electrochemical deposition [23,24,33–35] or physical templating-assisted sol– gel route [36]. However, none of the previously cited works went a step further the synthetic process and tested the functional properties and potentialities of these composite systems, focusing only on the NTs growth mechanisms and varying different synthetic parameters. In the present work, ZnO nanostructures were synthesized by sol–gel route into the pores of polycarbonate membranes (hereafter indicated as ZnO–PC), having three different pore diameters (i.e., 100, 50 and 30 nm), and for the first time it demonstrated their application as nanogenerators for mechanical energy harvesting. The advantages offered by this technique are the rapidity and the high-throughput of the synthetic procedure, forming, in one-single step, arrays of vertically aligned one-dimensional (1D) ZnO nanostructures supported in a flexible and insulating matrix. The obtained nanocomposite materials are thus quite easy to handle and can be straightforward used, without any further chemical or thermal treatments. In particular there is no need of removing the templating matrix, which is in contrast highly desired because it gives high flexibility and compactness to the whole sample, enabling the direct assembly or integration into devices for different applications. Here we present for the first time the piezoelectric

475 characterization and application as energy harvester of the templated 1D ZnO nanostructures, exploiting the potential of having such inorganic nanomaterial vertically oriented in an extremely flexible hosting matrix. The ZnO–PC nanogenerators were successfully tested both under compressive and bending stresses, examining the effect of the PC matrix pore size, which is reflected on the diameter of the templated ZnO, toward the generated voltage. In the case of 100 nm composite nanostructure, we reached a maximum output voltage and current of 1.15 V and 100 μA, respectively, and a maximum output power density of 287.5 mW/cm3, one of the highest reached in zinc oxide-based piezoelectric nanogenerators [15,37].

Experimental section Synthesis of ZnO-templated 1D nanostructures Track-etched porous polycarbonate membranes (Nucleopore from Whatmann, disc of 25 mm in diameter) having thickness of 5 μm and a nominal pore diameter of 100, 50 and 30 nm were immersed for 3 h at 88 1C in an hydrothermal growth bath composed by 25 mM zinc nitrate hexahydrate Zn(NО3)2
6H2O (purity 98%, Sigma-Aldrich), 12.5 mM hexamethylenetetramine C6H12N4 (HMT, purity 98%, Sigma-Aldrich), 5 mM polyethylenimine (PEI, average Mw  800, Sigma-Aldrich), and 320 mM ammonium NH4OH (28%, Sigma-Aldrich) in 100 mL bi-distilled water (from Direct-Q, a Millipore purification system), as reported for the synthesis of vertically aligned ZnO NWs [13]. After growth, the samples were thoroughly rinsed with bidistilled water and dried with nitrogen flow. Gentle cleaning of the PC surface with 1 M HCl was carried out to remove the ZnO material deposited on the top of the surface. This procedure did not affect the nanostructures formed into the PC pores, as confirmed by the measurements reported below. The samples were named ZnO–PC100, ZnO–PC50, and ZnO–PC30 referring to the ZnO 1D nanostructures templated into the PC membranes having 100, 50 and 30 nm pores, respectively.

Material characterization The morphology of the ZnO 1D nanostructures was investigated, after PC membrane dissolution in N-methyl pyrrolidone for 2 min, using both ZEISS Auriga and ZEISS Merlin Field Emission Scanning Electron Microscopes (FESEM). High Resolution Transmission Electron Microscopy (HRTEM) images were collected by an FEI Tecnai F20ST transmission electron microscope, operating at 200 kV. The ZnO–PC crystal structure was evaluated using a Panalytical X'Pert X-ray diffractometer in the Bragg–Brentano configuration. A Cu Kα monochromatic radiation was used as X-ray source, with a characteristic wavelength λ=1.54059 Å.

Nanogenerator preparation and measurement To collect the generated electric charges by the ZnO–PC samples, both surfaces of the membranes were first metalized with 50-nm thick platinum electrodes (by a Q150T sputter system form Quorum Technologies, operating at 50 mA and 1
10  4 bar using a suited circular mask in order to avoid metallization at the membrane edges and

476 thus cause short circuits during characterization) and then sandwiched between bottom and top electrodes of 50-μmthick copper metalized polyimide film. Therefore insulated copper wires were connected to the electrodes. The samples were then subjected to a controlled mechanical strain by means of a mechanical shaker (Tira TV 51110) with current amplifier, used as actuator. A Vibration Research 9500 and an accelerometer were used as controller and sensor for feedback loop, respectively. An harmonic steel cantilever was used as sample holder. The generated current and voltage were acquired by means of a National Instruments (NI PC 6115) and synchronized with the shaker actuation with a LabVIEW interface. The charge generated by the ZnO– PC100 sample under the compressive force of water droplets was converted into a voltage output by using a homemade charge amplifier circuit [13] and measured by using the NI PC 6115 board configured with a LabVIEW software. By considering the C1 and C2 capacitors used in the circuits (schematized in Figure 5a), the generated charge (Q) was calculated from the output voltage of the charge amplifier.

Results and discussion Figure 1a depicts, as representative example, the PC membrane having 100 nm pore diameter as average size and randomly

S. Stassi et al. distributed throughout the whole membrane. The other PC membranes have an average pore size of 50 and 30 nm (data not shown). After impregnating the three organic templates for 3 h in the precursor solution of ZnO, complete filling of the

Figure 2 X-ray diffraction patterns of the three ZnO–PC samples, carried out in the configuration depicted in the inset, where the scheme of ZnO 1D nanostructures templated into the porous PC matrix is shown. The c-axis and thus also the polarization P are oriented parallel to the membrane plane, thus perpendicular to the pore walls.

Figure 1 (a) Top FESEM image of the PC membrane having 100 nm pore size before the ZnO NT synthesis. (b) The ZnO–PC100 sample, showing high flexibility even after the ZnO NTs synthesis. FESEM images of the ZnO nanotubes, after the PC template dissolution, having (c) 100 (d) 50 and (e) 30 nm diameter. (f) Transmission electron microscopy and (g) High-Resolution TEM images of the ZnO– PC100 sample, showing the direction of the crystalline planes. (h) Graphical sketch of a portion of the ZnO–PC composite samples.

Flexible piezoelectric energy nanogenerator templating channels was achieved, resulting in hosted ZnO nanostructures of 5 μm in length and diameters of 100, 50 or 30 nm in average, which matches the dimensions of the respective templating pores (Figure 1c–e). In addition long portions of ZnO 1D nanostructures were observed (up to 4 μm as reported in Figure 1e), meaning that the templating PC pores (having a thickness of 5 μm) were completely filled throughout their whole length by the ZnO nanostructures. No dissolution effect was observed, in contrast with what normally occurring when using an alumina hosting membrane [1] and high flexibility was even maintained, as reported in Figure 1b. The composite samples were bended with mechanical tweezers for 100 times and afterwards no physical damages were observed, neither any degradation of the harvesting response was detected. In order to properly characterize the morphology of the prepared nanostructures, the PC matrices were dissolved in N-methyl pyrrolidone and the ZnO nanostructures imaged with FESEM (Figure 1c–e). The ZnO diameter is quite uniform and constant over the whole length. Unfortunately the dissolution process and dispersion produces aggregation of the nanostructures as visible throughout the TEM holey carbon grid used for the FESEM analysis, (Figure 1d), where either bundles (on the left) or single ZnO NTs (on the right, as indicated by the arrow) are visible. The surface of the

477 ZnO 1D nanostructures is corrugated and irregular in all the three sample cases, probably related to the not uniformity of PC surface channels and to traces of undissolved polymer, as clearly visible in Figure 1d and e. Indeed these PC membranes are fabricated by using high energy particles that first create nuclear track damage and then a suitable chemical etching is performed. They are originally destined for filtration functions, therefore their pore wall quality does not need to be very finely controlled. The Transmission Electron Microscopy (TEM) image reported in Figure 1f of the ZnO–PC100 sample reveals that the ZnO material was synthesized in the form of nanotubes. At the High Resolution TEM, the NT wall (Figure 1g) shows the orientation of the crystalline planes in the direction perpendicular to the long NT axis. In the analyzed portion of the NT, a single crystalline structure is observed; however, considering the corrugated surface of the NTs, a polycrystalline structure is postulated. A scheme of the final nanocomposite material depicting the ZnO–PC100 sample, is reported in Figure 1h. The crystalline structure and orientation of the ZnO nanotubes were further confirmed by X-ray diffraction analysis. The XRD analysis was performed in the Bragg–Brentano configuration on all the ZnO–PC nanocomposites as depicted in the inset of Figure 2. The X-ray patterns of the ZnO–PC

Figure 3 (a) Sketch and image of the mechanical shaker set up for the piezoelectric harvesting measurements. (b) Output voltage from the three ZnO–PC nanogenerator samples under bending stress at a frequency of 75 Hz. (c) Output generated voltage from an empty PC membrane.

478 samples presented in Figure 2 reveal the presence of an intense diffraction peak at 31.691, consistent with the (100) reflection of wurtzitic ZnO. The other typical peaks of ZnO at 34.171 and 36.151, assigned to the (002) and (010) reflections, respectively, are barely visible in all the three samples. It results from both XRD and TEM evaluations that all the crystalline domains constituting the 1D nanostructures walls are all oriented in the same [100] direction, whereas the [001] direction, corresponding to the c-axis, is tilted by 901. As evaluated through the Debye–Scherrer equation, the size of the crystalline domains results of 35.9 nm, 34.4 nm and 11.1 nm for the samples ZnO–PC100, ZnO–PC50 and ZnO– PC30, respectively. These calculated values are compatible with the hypothesis that the 1D nanostructures are nanotubes, as already evidenced by TEM for the ZNO–PC100 sample (Figure 1f), in which the wall thicknesses are between 30 and 40 nm. The small crystalline size of the ZnO–PC30 sample acknowledges on the broadening of the (100) diffraction peak of the related spectrum (Figure 2, in red) with respect to the other two samples. In addition the position of this peak is shifted to slightly higher angles, indicating shrinkage of the crystalline wurtzitic cell with respect to those of the other samples. From these characterizations, a possible formation mechanism of the ZnO crystalline domains can be proposed.

S. Stassi et al. Normally, the nanowires of ZnO grows along the highest energy direction, the [001], starting from the base to the top when the growth begins on a supporting substrate (seeded or not with ZnO nanocrystals) [13,38]. Here a similar growth mechanism can be assumed, since the supporting substrate in this case is the wall of the polycarbonate channels. It was indeed reported that polycarbonate can efficiently interact with the zinc cations during the reaction synthesis [36,39]. Since this polymer derives from a polycondesation reaction of a carbonate, bisphenol A, and carbonyl dichloride, there are several O–C(O)–O groups in the structural unit of polycarbonate. These groups can be easily hydrolyzed at above 60 1C leading to either alcoholic (–COH) and carboxylic (–COOH) groups, thus interacting with the Zn2 + cations coming from the dissociation reaction of zinc nitrate hexahydrate. It comes out that the PC pore walls are easily wetted by the hydrothermal solution (in our case the temperature is set to 88 1C). The OH  anions, coming from the dissociation of HMT and NH4OH, reacts with zinc cations on the PC wall surface, producing nuclei of different intermediate hydroxide species, such as Zn (OH)2
xH2O, ZnðOHÞ24  , ZnOH + (aq), Zn(OH)2(aq), Zn(OH)2(s), Zn(OH)3 (aq) [40]. When the concentration of zinc hydroxide complex reaches the supersaturation, under the thermal conditions used, the reaction of intermediates leads to large þ 2y  2xÞ zinc hydroxide species ZnxOyðOHÞðz , which then z

Figure 4 (a) Output current and (b) output voltage from the ZnO–PC100 nanogenerator under a bending stress at a frequency of 75 Hz. (c) Output current and voltage and (d) electrical power generated by the ZnO–PC100 nanogenerator when connected to different resistive loads.

Flexible piezoelectric energy nanogenerator condense forming the ZnO crystal nuclei on the active sites of the PC walls. Therefore the ZnO crystal structure starts to grow at the surface of the PC channels with an outward–inward direction in the pore channels, and not along bottom–top direction of the membrane. The fastest growth crystalline direction [001], which in the ZnO crystalline structure corresponds to the c-axis, results therefore perpendicular to the channels wall. A similar crystallization mechanism was also reported in the literature by Bechelany et al. [36] in polycarbonate membrane, as well as by Feng et al. [41] in the oriented pores of APA membranes. Therefore regarding the piezoelectric properties of the composite material, the polarization axis P will be oriented parallel to the organic template plane (as depicted with the red arrow in the inset of Figure 2). In order to assemble the nanogenerator device, the ZnO– PC composite samples were sputtered with platinum on both top and bottom sides (each electrode area was 0.8 cm2) and then sandwiched between two coppermetalized polyimide electrodes. The composite samples were then tested under bending strains, evaluating the charge generation and thus the output voltage. The characterization setup was ad-hoc designed for the direct piezoelectric characterizations and mechanical harvesting. It is based on a mechanical shaker controlled with a

479 feedback loop using an accelerometer sensor mounted on the vibrating stage (Figure 3a). The sample under test is mounted on the top surface of a harmonic steel cantilever clamped to the shaker head. In order to test the nanogenerator under bending stress, firstly a frequency sweep of the shaker actuation was performed, obtaining the frequency range where the cantilever, mounted with the device on the top side, had the higher displacement (i.e., resonant frequency). The oscillating frequency was then fixed at the selected value of 75 Hz and open-circuit voltages of 1.15 V, 0.91 V and 0.21 V were measured for the samples ZnO–PC100, ZnO–PC50, ZnO– PC30, respectively (Figure 3b). Therefore, the best sample in terms of highest piezoelectric generation results the ZnO– PC100 one, with a maximum generated short-circuit current of 100 μA (Figure 4a). The theoretically maximum output power that could be generated from the ZnO–PC100 sample is 115 μW, corresponding to an electrical power density of 287.5 mW/cm3 and an area power density of 143.88 μW/ cm2. To confirm that the output of the device is generated only by the piezoelectric ZnO nanotubes, an empty PC membrane was tested in the same bending configuration showing a voltage output comparable with electrical noise (Figure 3c).

Figure 5 (a) Scheme of the measurement set up for the water droplet impact and (b) of the charge amplifier electric circuit. Generated output charge under the impact of (c) a single droplet and (d) several consecutive water droplets.

480 When the nanogenerator is deformed by the bending of the steel bar, a side of the device suffers for compressive stress, while the other for tensile stress. Then, since the ZnO crystalline cell of the piezoelectric nanotubes is distorted, the center of gravity of the negative charges will no longer coincide with the positive one producing an electric dipole. This deformation will globally induce a generation of electric potential which is here maximized since the stress direction is parallel to the ZnO c-axis, the most polar one. As a result, the electrons flow into the external circuit to compensate the piezoelectric potential inside the sample. When the steel bar bends in the other direction, the compressed and tense part of the nanogenerator are inverted as well as the electrical potential and thus the electrons inside the circuit flow in the opposite direction. For these reasons the generated current and the voltage plots present both positive and negative peaks. The favorable configuration of the crystalline orientation of the ZnO 1D nanostructures enhances the harvesting of the nanogenerator since the bending stress, due to the oscillation of the cantilever structure, is parallel to the membrane plane, i.e., parallel to the [001] direction of the nanotube crystalline planes, along which the highest piezoelectric c-axis of the ZnO nanostructures is oriented. Figure 3b evidences that the piezoelectric voltage output decreases together with the nanostructure diameter. Since piezoelectricity is a material property related to the crystalline structure, this trend is mostly ascribable to the decrease of the size of the crystals constituting the ZnO 1D nanostructures, and to the resulting increase of grain boundaries and defects, as revealed by the X-Ray analysis in Figure 2. To test the amount of electrical power that the nanogenerator can produce in practical application, the output of the ZnO–PC100 composite sample was connected to variable resistive loads (Figure 4c). In this way one can evaluate which is the electric impedance that will best couple with the nanogenerator one, thus optimizing its harvesting properties. As expected, by increasing the resistive load in the tested range between 10 kΩ and 10 MΩ, a rise of the generated electrical potential and a consistent decrease of the electrical current, due to ohmic losses, were observed. It was then found that the tested nanogenerator has its best performances for a resistive load of 50 kΩ. The harvesting output power was 4.5 μW, corresponding to an electrical power density of 11.25 mW/cm3 and an area power density of 5.63 μW/cm2 (Figure 4d). To estimate the versatility of our ZnO–PC100 device, which showed the best piezoelectric performances among the three composite samples, and its ability to recover energy also from small compressive deformations, a simple but effective set up was arranged based on an automatic syringe pump (Figure 5a), as already described elsewhere [42,43]. The nanogenarator, packaged within electrodes and insulated by a polyimide layer, was connected to a charge amplifier board (schematized in Figure 5b) and the generation of electric charges was evaluated upon the repetitive fall of water drops, hitting the material surface. Indeed, water drops of about 45 mL (i.e., 45 mg) were dispensed by a syringe pump at a height of 15 mm from the surface of the sample. The very small deformation induced by the splash of the drop on the piezoelectric material was however able to produce an appreciable generation of charges (maximum 6 pC, see Figure 5c and d) detected by the charge amplifier electronic circuits. The result of this simple sensing test confirms the very high sensitivity of the ZnO–PC100 composite system.

S. Stassi et al.

Conclusion In summary, we demonstrated composite flexible nanogenerators with remarkable piezoelectric properties and high sensitivity to the induced deformations. ZnO 1D nanostructures were synthesized by sol–gel templating approach in three different pore sizes of organic track-etched polycarbonate membranes, which ensures an extremely noteworthy flexibility to the whole sample. The three types of nanogenerators showed good performances under bending stress since the ZnO nanotubes presents the polarization axes parallel to the applied strain. In this configuration the ZnO–PC100 nanogenerator showed outstanding performances, providing a remarkable output power density of 287.5 mW/cm3, one of the highest reached in zinc oxide based piezoelectric nanogenerator [15,37] and comparable with the output of the firstgeneration of triboelectric harvesting devices [44]. The significant nanogenerator properties were also tested by the impact of a water drop delivered on its surface, generating significant piezoelectric generated charges. These properties together with the high flexibility and electrical insulation of the polymeric matrix make the ZnO–PC construct a promising and ready-to-use smart material for both energy harvesting and sensing applications. With this regard the presented sol–gel templating approach could be also extended to other piezoelectric materials, particularly those having higher piezoelectric performance than ZnO (Lead zirconium Titanate, i.e., PZT, Barium Titanate,…), envisioning the possibility to obtain highly performing nanocomposite for energy generation.

References [1] V. Cauda, P. Motto, D. Perrone, G. Piccinini, D. Demarchi, Nanoscale Res. Lett. 9 (2014) 1–10. [2] S.M. Al-Hilli, M. Willander, A. Ost, P. Stralfors, J. Appl. Phys. 102 (2007) 014302. [3] S.N. Das, J.P. Kar, J.H. Choi, T. Il Lee, K.J. Moon, J.M. Myoung, J. Phys. Chem. C 114 (2010) 1689–1693. [4] A.I. Hochbaum, P.D. Yang, Chem. Rev. 110 (2010) 527–546. [5] J. Elias, C. Lévy-Clément, M. Bechelany, J. Michler, G.-Y. Wang, Z. Wang, L. Philippe, Adv. Mater. 22 (2010) 1607–1612. [6] A. Lamberti, R. Gazia, A. Sacco, S. Bianco, M. Quaglio, A. Chiodoni, E. Tresso, C.F. Pirri, Prog. Photovolt.: Res. Appl. 22 (2014) 189–197. [7] D. Pugliese, F. Bella, V. Cauda, A. Lamberti, A. Sacco, E. Tresso, S. Bianco, ACS Appl. Mater. Interfaces 5 (2013) 11288–11295. [8] V. Cauda, D. Pugliese, N. Garino, A. Sacco, S. Bianco, F. Bella, A. Lamberti, C. Gerbaldi, Energy 65 (2014) 639–646. [9] S. Hernández, V. Cauda, A. Chiodoni, S. Dallorto, A. Sacco, D. Hidalgo, E. Celasco, C.F. Pirri, ACS Appl. Mater. Interfaces 6 (2014) 12153–12167. [10] X. Yang, A. Wolcott, G. Wang, A. Sobo, R.C. Fitzmorris, F. Qian, J.Z. Zhang, Y. Li, Nano Lett. 9 (2009) 2331–2336. [11] J.R. LaRoche, Y.W. Heo, B.S. Kang, L.C. Tien, Y. Kwon, D.P. Norton, B.P. Gila, F. Ren, S.J. Pearton, J. Electron. Mater. 34 (2005) 404–408. [12] S.N. Cha, J.E. Jang, Y. Choi, G.A.J. Amaratunga, G.W. Ho, M.E. Welland, D.G. Hasko, D.J. Kang, J.M. Kim, Appl. Phys. Lett. 89 (2006) 263102. [13] V.F. Rivera, F. Auras, P. Motto, S. Stassi, G. Canavese, E. Celasco, T. Bein, B. Onida, V. Cauda, Chem. Eur. J. 19 (2013) 14665–14674. [14] Z.L. Wang, Mater. Sci. Eng. R 64 (2009) 33–71. [15] Z.L. Wang, W. Wu, Angew. Chem. Int. Ed. 51 (2012) 11700–11721. [16] Z.L. Wang, J. Song, Science 312 (2006) 242–246.

Flexible piezoelectric energy nanogenerator [17] S. Xu, Y. Wei, J. Liu, R. Yang, Z.L. Wang, Nano Lett. 8 (2008) 4027–4032. [18] Y. Qin, X. Wang, Z.L. Wang, Nature 451 (2008) 809–813. [19] M. Lee, C.Y. Chen, S. Wang, S.N. Cha, Y.J. Park, J.M. Kim, L.J. Chou, Z.L. Wang, Adv. Mater. 24 (2012) 1759–1764. [20] Y.C. Kong, D.P. Yu, B. Zhang, W. Fang, S.Q. Feng, Appl. Phys. Lett. 78 (2001) 407–409. [21] L.V. Podrezova, S. Porro, V. Cauda, M. Fontana, G. Cicero, Appl. Phys. A 113 (2013) 623–632. [22] Z. Fan, J.G. Lu, J. Nanosci. Nanotechnol. 5 (2005) 1561–1573. [23] Y.C. Wang, I.C. Leu, M.H. Hon, J. Mater. Chem. 12 (2002) 2439–2444. [24] M.J. Zheng, L.D. Zhang, G.H. Li, W.Z. Shen, Chem. Phys. Lett. 363 (2002) 123–128. [25] D. Pullini, A. Pruna, S. Zanin, D.B. Mataix, J. Electrochem. Soc. 159 (2011) E45–E51. [26] B. Liu, H.C. Zeng, J. Am. Chem. Soc. 125 (2003) 4430–4431. [27] J. Wang, L. Gao, J. Mater. Chem. 13 (2003) 2551–2554. [28] M. Laurenti, V. Cauda, R. Gazia, M. Fontana, V.F. Rivera, S. Bianco, G. Canavese, Eur. J. Inorg. Chem. 14 (2013) 2520–2527. [29] C. Ottone, K. Bejtka, A. Chiodoni, V. Farıas, I. Roppolo, G. Canavese, S. Stassi, V. Cauda, New J. Chem. 38 (2014) 2058–2065. [30] H. Zhang, X.Y. Ma, J. Xu, J.J. Niu, D.R. Yang, Nanotechnology 14 (2003) 423–426. [31] G.S. Wu, T. Xie, X.Y. Yuan, Y. Li, L. Yang, Y.H. Xiao, L.D. Zhang, Solid State Commun. 134 (2005) 485–489. [32] V. Cauda, S. Stassi, K. Bejtka, G. Canavese, ACS Appl. Mater. Interfaces 5 (2013) 6430–6437. [33] S.K. Sharma, A. Rammohan, A. Sharma, J. Colloid Interface Sci. 344 (2010) 1–9. [34] M. Lai, D.J. Riley, Chem. Mater. 18 (2006) 2233–2237. [35] Y. Leprince-Wang, G.Y. Wang, X.Z. Zhang, D.P. Yu, J. Cryst. Growth 287 (2006) (2006) 89–93. [36] M. Bechelany, A. Amin, A. Brioude, D. Cornu, P. Miele, J. Nanopart. Res. 14 (2012) 980–987. [37] G. Zhu, A.C. Wang, Y. Liu, Y. Zhou, Z.L. Wang, Nano Lett. 12 (2012) 3086–3090. [38] J.-H. Tian, J. Hu, S.-S. Li, F. Zhang, J. Liu, J. Shi, X. Li, Z.Q. Tian, Y. Chen, Nanotechnology 22 (2011) 245601. [39] M. Goyal, R. Nagahata, J. Sugiyama, M. Asai, M. Ueda, K. Tekeuchi, Polymer 41 (2000) 2289–2293. [40] V. Cauda, R. Gazia, S. Porro, S. Stassi, G. Canavese, I. Roppolo, A. Chiolerio, Nanostructured ZnO materials: synthesis, properties and applications, in: D.L.B. Bhushan, S.R. Schricker, W. Sigmund, S. Zauscher (Eds.), Handbook of Nanomaterial Properties, Springer Verlag, Berlin Heidelberg, 2014, pp. 137–177. [41] W. Feng, P. Huang, L. Jiang, Ceram. Int. 40 (2014) 6383–6387. [42] R. Gazia, P. Motto, S. Stassi, A. Sacco, A. Virga, A. Lamberti, G. Canavese, Nano Energy 2 (2013) 1294–1302. [43] G. Canavese, S. Stassi, V. Cauda, A. Verna, P. Motto, A. Chiodoni, S.L. Marasso, D. Demarchi, Sens. J. IEEE 13 (2013) 2237–2244. [44] Z.L. Wang, ACS Nano 7 (2013) 9533–9557.

Stefano Stassi works as a post-doc researcher at the Department of Applied Science and Technology, Politecnico di Torino in Turin, Italy. He received his M.S. degree in nanotechnology engineering and his Ph.D.in physics from Politecnico di Torino and Istituto Italiano di Tecnologia, Torino, Italy, in 2009 and 2013, respectively. His current research interests include Micro Electro-Mechanical Systems (MEMS) resonating sensor fabrication, and integration for biomedical analysis, preparation, and characterization of piezoresistive and piezoelectric materials and metal nanoparticle synthesis.

481 Valentina Cauda works as senior post-doc at the Istituto Italiano di Tecnologia in Turin, Italy. She received her Ph.D. in material science and technology in 2008 from Politecnico di Torino, Italy, and graduated with a degree in chemical engineering in 2004. From 2008 to 2010, she worked as a postdoc at the Faculty of Chemistry, University of Munich, Germany. She is involved in the chemical synthesis and characterization of polymeric and oxide-based nanostructures for piezoelectric and sensing applications. Carminna Ottone received her Master degree from the Politecnico di Torino in 2011. She is currently pursuing Ph.D in Materials Science and Technology at the same university under the supervision of Prof. Barbara Bonelli. She has performed part of her Ph.D thesis at the Istituto Italiano di Tecnologia (IIT), where she has collaborated with Dr. Valentina Cauda. Her research focused on the study of the activity of manganese oxides as water oxidation catalysts and the nanostructuration of ZnO by the template assisted method. Angelica Chiodoni received her Master Degree in Material Science at the Università Degli Studi di Torino, Italy in 2001 and her Ph. D. in Physics at the Politecnico di Torino in 2005. At present she is a researcher of Istituto Italiano di Tecnologia at the Center for Space Human Robotics-Politecnico di Torino. Her scientific activity is devoted to the characterization of materials and devices for bio-inspired energy applications. In particular, her activity is focused on the characterization and processing of materials by means of X-ray diffraction, Field Emission Scanning Electron Microscopy, Focused Ion Beam and Transmission Electron Microscopy. Candido Fabrizio Pirri is Professor of Physics of Matter at Politecnico di Torino in Turin, Italy and since 2011 he is Director of the Center for space Human Robotics of the Istituto Italiano di Tecnologia in Turin. He is responsible of the Materials and Microsystems Laboratory of Politecnico of Torino (Chi-LAB) and coordinator, for the Politecnico of Turin, of the International Master Degree in Micro and Nanotechnologies for Integrated Systems (at POLITO, INP Grenoble, and sEPF Lausanne). He is reviewer for the main international journals in the field of the Physics of Matter and Nanoscience/Nanotechnology. His research activity is reported in more than 150 articles published in international journals. Giancarlo Canavese is a researcher at Department of Applied Science and Technology, Politecnico di Torino in Turin, Italy. He received his M.E. degree in mechanical engineering in 2004 and his Ph.D. degree in biomedical engineering in 2008 from Politecnico di Torino. From 2009 to 2014 he worked as Senior Post-Doc at the Istituto Italiano di Tecnologia in Turin, Italy. In 2013–2014, he also worked for 6 months as a visiting researcher at the Houston Methodist Research Institute, Texas. His areas of interest include piezoresistive composite materials, Micro ElectroMechanical Systems (MEMS) technologies, and the distribution of tactile sensors for robotic applications.