Fabrication of flexible energy harvesting device based on K0.5Na0.5NbO3 nanopowders

Article abstract of DOI:10.1016/j.jallcom.2014.12.239

K_0.5Na_0.5NbO₃ (KNN) powder was synthesized by a novel sol-gel method and used to fabricate an energy harvesting device. The precursor gel and KNN powder were studied by thermogravimetric analysis (TG), differential scanning calorimetry (DSC), X-ray diffraction (XRD), scanning electron microscopy (SEM), and transmission electron microscopy (TEM). The elemental composition of the KNN powder was characterized by energy dispersive X-ray analysis (EDAX). Well crystallized single phase perovskite KNN powder with an average particle size on the order of 450 nm was obtained from the gel after calcining at 750 °C for 2 h. A flexible and implantable energy harvesting device fabricated using the KNN nanopowder exhibited an output power of up to 0.13 μW with a load resistance of 100 MΩ. The Young's modulus of the device was determined to be 10.4 GPa by atomic force microscopy (AFM).

Full text of DOI:10.1016/j.jallcom.2014.12.239

Journal of Alloys and Compounds 629 (2015) 113–117
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: www.elsevier.com/locate/jalcom
Fabrication of flexible energy harvesting device based on K0.5Na0.5NbO3
nanopowders
a,
a
b
b
Yongyong Zhuang a, , Zhuo Xu , Fei Li , Zhipeng Liao , Weihua Liu
⇑
⇑
a Electronic Materials Research Laboratory, Key Laboratory of the Ministry of Education and International Center for Dielectric Research, Xi’an Jiaotong University, Xi’an 710049, China
b Vacuum Microelectronic and Microelectronic Mechanical Institute, Xi’an Jiaotong University, Xi’an 710049, China
a r t i c l e i n f o
a b s t r a c t
Article history:
K0.5Na0.5NbO3 (KNN) powder was synthesized by a novel sol–gel method and used to fabricate an energy
harvesting device. The precursor gel and KNN powder were studied by thermogravimetric analysis (TG),
differential scanning calorimetry (DSC), X-ray diffraction (XRD), scanning electron microscopy (SEM), and
transmission electron microscopy (TEM). The elemental composition of the KNN powder was character-
ized by energy dispersive X-ray analysis (EDAX). Well crystallized single phase perovskite KNN powder
with an average particle size on the order of 450 nm was obtained from the gel after calcining at 750 °C
for 2 h. A flexible and implantable energy harvesting device fabricated using the KNN nanopowder exhib-
Received 9 September 2014
Received in revised form 28 December 2014
Accepted 29 December 2014
Available online 8 January 2015
Keywords:
Nanostructure materials
Sol–gel processes
Piezoelectricity
Transmission electron microscopy, TEM
ited an output power of up to 0.13
lW with a load resistance of 100 MX. The Young’s modulus of the
device was determined to be 10.4 GPa by atomic force microscopy (AFM).
Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction
In this paper, we report the fabrication and characterization of
piezoelectric K0.5Na0.5NbO3 (KNN) based nanomaterials. A flexible
and implantable KNN-based energy harvesting device was devel-
oped. X-ray diffraction (XRD) and energy dispersive spectroscopy
(EDAX) were used to characterize the molar ratio of K/Na in the
KNN samples. The output voltage and current of the fabricated
device was measured, and the output power was calculated to be
High-quality power source, of microscale or even nanoscale size
have been attracting increasing attention, owing to the recent min-
iaturization of various functional electronic devices. These nanode-
vices need to be independent, sustainable, maintain-free, and able to
operate continuously through charging of their power source. Thus,
the development of nanotechnology that can harvest energy from
the environment is urgently required. Various such energy harvest-
ing devices have been developed, such as nanogenerators [1–3],
solar cells [4,5], thermoelectric cells [6], and hydrogen cells [7–9].
Zinc oxide (ZnO) [10,11] is a piezoelectric material reported to be
suitable for harvesting energy to generate electricity from ambient
mechanical energy, including heartbeats, blood flow, muscle
stretching, and body movement. However, the low piezoelectric
coefficient of ZnO material compared with those of other piezoelec-
tric materials, such as lead zirconate titanate (PZT) [12,13], barium
titanate (BT) [14], sodium potassium niobate (KNN) [15,16], and
lead magnesium niobium titanate (PMNT) [17] limits its application.
Among these piezoelectric materials, KNN is an environmentally-
friendly material (no lead content), and has a higher Curie tempera-
ture, which indicates a higher stability at higher temperatures.
Therefore, K1ꢀxNaxNbO3 (KNN)-based materials have become prom-
ising candidates for mechanical energy harvesting.
0.13
l
W with a load resistance of 100 M
X
. This value is slightly
W) reported by Wu
higher than that of the PZT materials (0.12
l
et al. [18]. The low-cost, flexibility, implantability and wearablility
of the present devices demonstrate their potential for application
in the future energy harvesting technologies.
2. Experimental procedures
2.1. Synthesis of K0.5Na0.5NbO3 powders
K0.5Na0.5NbO3 (KNN) nanopowder was synthesized using a sol–gel method. Com-
mercial Na2CO3 (Analytical reagent grade (AR), 99.8%, Tianjin Tianli Chemical Reagent
Co. Ltd., Tianjin, China), Nb2O5 (4 N, 99.99%, Sinopharm Chemical Reagent Co. Ltd.,
China), K2CO3 (AR, 99.0%), citric acid (AR, 99.5%), aqueous ammonia (AR, 25–28%,)
(Tianjin Hedong District Hongyan Chemical Reagent Factory, Tianjin, China), HF (AR,
40.0%, Tianjin Fuyu Fine Chemical Co. Ltd., Tianjin, China), were used as the starting
materials. Firstly, a suitable amount of Nb2O5 powder was dissolved in HF (40%) solu-
tion after heating in water at 90 °C for 12 h. Then, ammonia hydroxide solution was
added dropwise, which lead to the precipitation of Nb5+ as hydroxide (Nb(OH)5) under
basic conditions (ca. pH ꢁ 9). To synthesize KNN, stoichiometric amounts of Na2CO3
and K2CO3 powder were mixedwith the precipitated Nb(OH)5 powderina Na+:K+:Nb5+
molar ratio of 1:1:2. After that, citric acid solution (2 mol/L) was added dropwise to the
mixture until the powder was completely dissolved to form a colorless sol. The main
chemical reactions that produced the niobium precursor can be expressed as follows:
⇑
Corresponding authors. Tel.: +86 15929945189.
E-mail addresses: xiaozhuang235@163.com (Y. Zhuang), xuzhuo@mail.xjtu.edu.
cn (Z. Xu).
http://dx.doi.org/10.1016/j.jallcom.2014.12.239
0925-8388/Ó 2015 Elsevier B.V. All rights reserved.

114
Y. Zhuang et al. / Journal of Alloys and Compounds 629 (2015) 113–117
Nb2O5 þ 2ðn þ 5ÞHF ! 2HnNbF5þn þ 5H2O;
ð1Þ
NbFꢀ6 ꢀ NbF5 þ Fꢀ ꢀ NbFþ4 þ 2Fꢀ ꢀ NbF23þ þ 3Fꢀ ꢀ NbF23þ þ 4Fꢀ
ꢀ NbF4þ þ 5Fꢀ ꢀ Nb5þ þ 6Fꢀ;
ð2Þ
ð3Þ
ð4Þ
2Nb5þ þ 10OHꢀ þ 10NH4þ ! 2NbðOHÞ5 # þ10NH3 " þ5H2 ";
3NbðOHÞ5 þ 5H3C6H5O7 ! 3Nb5þ þ 15H2O þ 5C6H5O3ꢀ
:
7
To obtain a dry gel, the precursor sol solution was heated at 120 °C for 12 h. The dry
gels were finally calcined at 750 °C for 2 h to form KNN powder with perovskite
structure.
2.2. Fabrication of flexible energy harvesting device with the K0.5Na0.5NbO3 nanopowder
The energy harvesting device consisted of three layers as shown in Fig. 1. The
copper clad laminate films (two copper layers and one intermediate Kapton layer,
120 lm total thicknesses) were chosen as the flexible substrate and bottom elec-
trode. A mixture of the KNN nanopowder and poly(dimethylsiloxane) (PDMS) in a
mass ratio of 1:10 were spin-coated on the clean substrate at 3000 rpm (rpm),
and cured for 30 min at 80 °C to obtain the piezoelectric potential layer. A copper
foil (25 lm) was placed on the mixture layer as the top electrode.
Fig. 2. Thermogravimetric analysis (TG) and differential scanning calorimetry (DSC)
curves of the gel precursor powder.
2.3. Characterization
The suitable calcination temperature of the powder was confirmed by thermo-
gravimetric analysis (TG) and differential scanning calorimetry (DSC) (NETZSCH
STA 449C, Germany). The phase composition of the KNN powder was analyzed
using XRD (D/max2200 Rigaku, Japan). The morphology and microstructure of the
powder were characterized by scanning electron microscopy (SEM; Quanta FEG
250 FEI, USA) and transmission electron microscopy (TEM; JEM 2100 JEOL, Japan).
The elemental composition of the powder was characterized by energy dispersive
X-ray analysis (EDAX; Quanta FEG 250 FEI, USA). The Young’s modulus and the
topography of the energy harvesting device under external force were measured
by atomic force microscopy (AFM; INNOVA Veeco, Germany) with a probe radius
of 2 nm. The sensitivity and spring constant of the probe were determined to be
93.2% was observed from room temperature to 690 °C and no fur-
ther mass loss occurred at higher temperatures. Thus, to ensure
complete crystallization of KNN, the gel powder was calcined at
750 °C.
Fig. 3 shows the XRD patterns of KNN powder obtained using
the present sol–gel technique and the traditional solid state reac-
tion method, annealed at 750 °C for 2 h. All the powder showed
perovskite structure and no impurity phases were observed. The
composition of the as-synthesized sol–gel KNN powder was similar
to that of the K0.5Na0.5NbO3 powder obtained by the solid state
reaction method. This confirms the retention of the 1:1:2 M ratios
of the Na+/K+/Nb5+ species in the KNN powder. However, the char-
acteristic peak splitting usually observed at around 45° for pseu-
docubic powder is not obvious in Fig. 3(a). This is attributed to
broadening effects associated with the fine crystalline size of the
KNN nanopowders [20].
Fig. 4 shows SEM and EDAX images of KNN nanopowder
obtained by calcinations of the gel precursor at 750 °C for 2 h.
The powder consisted of regular cubic nanoparticles, and had
low agglomeration. The average particle size of the KNN powder
was estimated to be 450 nm from Fig. 4(b). The distribution of O,
Na, Nb, and K elements in the powder is shown in Fig. 4(c)–(f),
respectively. The even color of the four images indicates the uni-
form distribution of the four elements in the sample. The quantita-
tive results are shown in Fig. 4(g). The table inset in Fig. 4(g)
indicated that the molar ratio of K:Na:Nb:O in the sample was
0.0439
lm/V and 0.02785 N/m, respectively. The impedance of the device under
different frequencies was characterized using
a
precision impedance analyzer
(4294A Agilent, USA). The d33 value was measured by using a quasi-static d33 tester
(ZJ-4A Institute of Acoustics, Chinese Academy of Sciences, China). The output cur-
rent and voltage signals of the energy harvesting device were collected by using an
oscilloscope (3014B Tektronix, USA) equipped with a low-noise current preampli-
fier (SR570 SRS, USA) and a low-noise voltage preamplifier (SR560 SRS, USA),
respectively.
3. Results and discussion
Fig. 2 shows the TG–DSC results obtained for the sol–gel derived
precursor powder up to 900 °C. Endothermic peaks appeared at
120 °C and 227 °C, and one exothermic peak was observed at
603 °C. Four regions of mass loss were observed in the TG curve.
The endothermic peak at 120 °C and the minor mass loss of about
6.8% between room temperature and 165 °C were attributed to
the removal of moisture and alcohol. The endothermic peak at
227 °C accompanied by a mass loss of about 56.2% from 165 °C to
260 °C was caused by the removal of excess citric acid. The third
region with a mass loss of about 15.7% between 260 °C and 550 °C
was attributed to the decomposition of organic groups from citric
acid. The mass loss of 14.5% in the temperature range of
550–690 °C was related to the formation of the KNN phase. The
exothermic peak at 603 °C is considered to be the crystallization
temperature of intermediate KNN phase [19]. A mass loss of about
Fig. 1. Schematic diagram of an energy harvesting device, where the three origin
color layers are copper electrodes, the dark red color layer is Kapton, the white
spots in the mediate layer are the KNN nanopowder, the gray layer is the PDMS. (For
interpretation of the references to color in this figure legend, the reader is referred
to the web version of this article.)
Fig. 3. XRD patterns of KNN powder obtained from (a) sol–gel method and (b) solid
state reaction method. The powder was annealed at 750 °C for 2 h.

Y. Zhuang et al. / Journal of Alloys and Compounds 629 (2015) 113–117
115
Fig. 4. SEM images and EDAX images of the KNN nanopowder: (a) low magnification image of KNN powder; (b) high magnification of (a); (c), (d), (e) and (f) are the
distribution maps of O, Na, Nb and K elements of the rectangular region from (a); (g) is the EDAX image of KNN powder, the insert table is the quantitative result of KNN
powder.
(011) crystal plane of orthorhombic K0.5Na0.5NbO3. This is coinci-
dent with the selected area diffraction pattern (SAED) shown in
Fig. 5(d). The view direction of the diffraction pattern is along
the ½111ꢂ crystal direction.
The mechanical properties of the flexible energy harvesting
device fabricated using the KNN powder is shown in Fig. 6. The
width, length, and thickness of the device were 2.5 cm, 4.5 cm,
and 80 lm, respectively. Fig. 6(a) shows a topographical image of
the PDMS/KNN mixture layer. The surface roughness of the whole
layer was on the order of 20 nm. An image of this layer under
external force is shown in Fig. 6(b). The distribution of lateral force
across the layer can be calculated using the following equation:
FL ¼ S ꢃ K ꢃ VL
ð5Þ
where FL is the lateral force, S is the sensitivity (0.0439
lm/V), K is
the spring constant of the probe (0.02785 N/m), and VL is the lateral
signal. The measured lateral force was on the order of 1 nN. The
force profiles are shown in Fig. 6(c). The AFM probe was controlled
to repeatedly approach (a–b), contact (b–c), and retract (d–e–f)
from the PDMS/KNN layer. Suitable mechanical properties are
essential for the fabrication of durable and reliable devices. The
Young’s modulus of the fabricated device was calculated from the
contact (b–c) part using the Hertz model [21]:
Fig. 5. TEM images and SAED pattern of KNN nanopowder: (a) low magnification;
(b) high magnification; (c) high resolution TEM image of (b); (d) SAED pattern from
the rectangle area in (b).
nearly 1:1:2:6. The K:Na:Nb ratio was coincident with the
designed molar ratio. The excess of O element was caused by the
specimen holder and conductive adhesive. Accordingly, the chem-
ical formula of the KNN powder obtained from the gel precursor
should be K0.5Na0.5NbO3. Fig. 5 shows TEM images of the KNN pow-
der and the selected area electron diffraction pattern (SAED) of a
KNN nanoparticle. The particle size of the KNN nanopowder was
found to be on the order of 200 nm. The high magnification image
of the red rectangular area of Fig. 5(b) was shown in Fig. 5(c). The
interplanar spacing of the direction marked with the blue arrow in
Fig. 5(c) was determined to be 2.842 Å, which corresponded to
F ¼ 4=3ER1=2d3=2
;
ð6Þ
ð7Þ
1
1
1
¼
þ
R1 R2
;
R
where F is the loading force, d is the indentation depth, E is Young’s
modulus, R is the radius of the effective radius, R1 is the radius of
the probe (2 nm), and R2 is the radius of the PDMS/KNN layer
(1). The obtained average Young’s modulus was 10.4 GPa. The

116
Y. Zhuang et al. / Journal of Alloys and Compounds 629 (2015) 113–117
Fig. 6. Atomic force microscopy (AFM) images of the energy harvesting device with (a) topography image; (b) lateral signal image; (c) force profiles on the PDMS/KNN layer.
harvesting device. The external force was applied to the device
using fingers. When an external force was placed on the device, a
voltage can be induced via the piezoelectric response (the d33 of
the device was ꢀ6 pC/N). As a result, a current through an outer
load was observed in the external circuit. This process generated
output current and voltage signals, with positive/negative signals,
as shown in Fig. 8. When the external force was removed, the
strain transferred to the device caused by the external force disap-
peared, and the voltage between the two electrodes also disap-
peared. Therefore, the accumulated electrons flowed back
through the outer load in the external circuit. This process gener-
ated reverse output current and voltage signals, as shown in
Fig. 8. The largest output current and voltage signals observed for
the KNN based device were 4.48 nA and 0.17 V, respectively. The
power delivered to the outer load under an external force was esti-
mated using the following equation [24]:
Z
2
1
T
V0ðtÞ
Fig. 7. Impedance versus frequency loop of the energy harvesting device.
PL ¼
dt;
ð8Þ
RL
where PL is the delivered power, V0(t) is the real-time voltage, RL is
the resistance of the outer load, and T is the time period of external
force application. A delivered power of up to 0.13 lW was obtained
impedance properties of a device affect its energy harvesting effi-
ciency. The impedance of the fabricated energy harvesting device
under different frequencies is shown in Fig. 7. The impedance of
the device was observed to decrease with increasing frequency.
with a load resistance of 100 MX.
The maximum and minimum impedance values were 800 K
2.76 at 10 kHz and 79 MHz, respectively.
X and
X
4. Conclusions
To investigate the performance of the flexible energy harvesting
device, the short-circuit current and open-circuit voltage were
measured under external force. Ni et al. [22] have verified that
the current and voltage signals obtained under such conditions
are related to the piezoelectricity of the piezoelectric materials.
The power generation mechanism previously proposed [18,22,23]
for similar energy generating devices based on piezoelectric mate-
rials can be applied to the present KNN nanopowder-based energy
Single phase and well crystallized K0.5Na0.5NbO3 (KNN) nano-
powder was prepared using a sol–gel method, with a calcination
temperature of 750 °C. The average grain size of the KNN powder
was determined to be about 450 nm. The Young’s modulus of an
energy harvesting device fabricated using the powder was about
10.4 GPa. The power delivered by the fabricated energy harvesting
device was 0.13 lW with a load resistance of 100 MX. This work is
Fig. 8. Short-circuit current (a) and open-circuit voltage (b) of the flexible energy harvesting device when subject to external force. The inset of (a) is a photograph of the
energy harvesting device, the size of the device is 2.5 cm ꢃ 4.5 cm ꢃ 80
lm. The inset of (b) is a SEM image of the PDMS/KNN layer, the distribution of the KNN powder in
PDMS is uniform.

Y. Zhuang et al. / Journal of Alloys and Compounds 629 (2015) 113–117
117
[7] C. Corgnale, T. Motyka, S. Greenway, J.M. Perez-Berrios, A. Nakano, H. Ito, T.
Maeda, J. Alloys Comp. 580 (2013) S406–409.
[8] L. Schlapbach, Nature 460 (2009) 809–811.
[9] H.G. Pan, R. Li, Y.F. Liu, M.X. Gao, H. Miao, Y.Q. Lei, Q.D. Wang, J. Alloys Comp.
463 (2008) 189–195.
expected to represent a general high-efficiency method of fabricat-
ing low-cost flexible, implantable, or wearable mechanical energy
harvesting devices.
[10] S. Xu, Y. Qin, C. Xu, Y. Wei, R. Yang, Z.L. Wang, Nat. Nanotechnol.
(2010) 366–373.
5
Acknowledgements
[11] Q.L. Liao, Z. Zhang, X.H. Zhang, M. Mohr, Y. Zhang, H.J. Fecht, Nano Res. 7
(2014) 917–928.
[12] S. Xu, B.J. Hansen, Z.L. Wang, Nat. Commun. 1 (2010) 93–97.
[13] Y. Qi, N.T. Jafferis, K.L. Jr, C.M. Lee, H. Ahmad, M.C. McAlpine, Nano Lett. 10
(2010) 524–528.
[14] S. H Shin, Y.H. Kim, M.H. Lee, J.Y. Jung, J. Nah, ACS Nano 8 (2014) 2766–2773.
[15] L.Q. Cheng, K. Wang, J.F. Li, Chem. Commun. 49 (2013) 4003–4005.
[16] X.P. Wang, J.G. Wu, D.Q. Xiao, J.G. Zhu, X.J. Cheng, T. Zheng, B.Y. Zhang, X.J. Lou,
X.J. Wang, J. Am. Chem. Soc. 136 (2014) 2905–2910.
The work was supported by the National Nature Science Foun-
dation of China (Grant Nos. 51102193 and 51372196), China Post-
doctoral Science Foundation (Grant Nos. 2012M521761 and
2014T70914) and 111 Project B14040.
References
[17] S. Xu, Y.W. Yeh, G. Poirier, M.C. McAlpine, R.A. Register, N. Yao, Nano Lett. 13
(2013) 2393–2398.
[18] W. Wu, S. Bai, M.M. Yuan, Y. Qin, Z.L. Wang, T. Jing, ACS Nano 6 (2012)
6231–6235.
[19] A. Chowdhury, J. Bould, J. Nanopart. Res. 12 (2010) 209–215.
[20] A. Chowdhury, J. Bould, M.G.S. Londesborough, S.J. Milne, Chem. Mater. 22
(2010) 3862–3874.
[21] M. Nikkhah, J.S. Strobl, E.M. Schmelz, J. Biomech. 44 (2001) 762–766.
[22] X. Ni, F. Wang, A. Lin, Q. Xu, Z. Yang, Y. Qin, Sci. Adv. Mater. 5 (2013)
1781–1787. 7.
[1] A.F. Y u, M. Song, Y. Zhang, J.Z. Kou, J.Y. Zhai, Z.L. Wang, Nanotechnology 25
(2014). 455503-1-6.
[2] L. Gu, N.Y. Cui, L. Cheng, Q. Xu, S. Bai, M.M. Yuan, W.W. Wu, J.M. Liu, Y. Zhao, F.
Ma, Y. Qin, Z.L. Wang, Nano Lett. 13 (2013) 91–94.
[3] D. Dhakars, Y. Grawli, S. Chhatre, P. Wadgaonkar, S. Ogale, Phys. Chem. Chem.
Phys. 16 (2014) 22874–22881.
[4] P.K. Mokhtar, S.N. Masoud, H.M.S. Mostafa, J. Alloys Comp. 617 (2014)
627–632.
[5] A. Lamberti, A. Sacco, M. Laurenti, M. Fontana, C.F. Pirri, S. Bianco, J. Alloys
Comp. 615 (2014) S487–490.
[23] S. Bai, Q. Xu, L. Gu, F. Ma, Y. Qin, Z.L. Wang, Nano Energy 1 (2012) 789–795.
[24] X. Chen, S.Y. Xu, N. Yao, Y. Shi, Nano Lett. 10 (2010) 2133–2137.
[6] A.I. Boukai, Y. Bunimovich, J.T. Kheli, J.K. Yu, W.A. Goddard, J.R. Heath, Nature
451 (2008) 168–171.