James 2014
James 2014
James 2014
composites
This content has been downloaded from IOPscience. Please scroll down to see the full text.
(http://iopscience.iop.org/0964-1726/23/5/055001)
View the table of contents for this issue, or go to the journal homepage for more
Download details:
IP Address: 131.204.172.32
This content was downloaded on 31/08/2014 at 08:06
E-mail: n.kunnamkuzhakkaljames@tudelft.nl
Abstract
Piezoelectric ceramic–polymer composites with 0–3 connectivity were fabricated using lead
zirconium titanate (PZT) powder dispersed in an ionomer (Zn ionomer) and its reference
ethylene methacrylic acid copolymer (EMAA) polymer matrix. The PZT–Zn ionomer and
PZT–EMAA composites were prepared by melt extrusion followed by hot pressing. The
effects of poling conditions such as temperature, time and electric field on the piezoelectric
properties of the composites were investigated. The experimentally observed piezoelectric
charge coefficient and dielectric constant of the composites were compared with theoretical
models. The results show that PZT–Zn ionomer composites have better piezoelectric
properties compared to PZT–EMAA composites. The static and fatigue properties of the
composites were investigated. The PZT–Zn ionomer composites were found to have excellent
fatigue resistance even at strain levels of 4%. Due to the self-healing capabilities of the
ionomer matrix, the loss of piezoelectric properties after high strain tensile cyclic loading
could be partially recovered by thermal healing.
contact with one another while the polymer phase is connected piezoelectric ceramic particles with a low dielectric constant.
in all three dimensions. The 0–3 composites are considered to Different approaches have been taken to increase the poling
be the least complicated in terms of manufacturing and ease of efficiency, for example the use of piezoelectrically active
molding in complex, predesigned shapes. Although there have polymers, the use of various conductive additives such as
been many reports about piezoelectric–polymer composites carbon black or graphite and the use of functionalized PZT
with 0–3 connectivity, only a handful of investigations have ceramics by a silane coupling agent to enhance the binding
been carried out on the effect of poling conditions on the piezo- between PZT and the polymer matrix [13, 19–21].
electric properties of the composites for polymer matrices with Yamada et al proposed a model to predict the piezoelectric
different electrical conductivities [9–15]. charge constant and dielectric constant of a composite by
In the present study an ionomer has been selected as the assuming a perfect degree of poling of individual ellipsoidal
polymer matrix because of its high flexibility, good mechanical piezoelectric particles which are dispersed in a continuous low
properties, excellent adhesion to the ceramic phase and self- dielectric polymer matrix [18]. The final equations are given
healing potential [16, 17]. The effective poling conditions for below.
nφ(εc − εp )
PZT–Zn ionomer composites were investigated and the results ε = εp 1 + (3)
were compared with the reference PZT–EMAA (ethylene nεp + (εc − εp )(1 − φ)
methacrylic acid copolymer) composites and PZT ceramics. nε
The experimentally observed dielectric and piezoelectric coef- d33 = φ d33c (4)
(nε + (εc − ε))
ficient are compared with Yamada’s model [18]. Furthermore,
where n is a parameter reflecting the aspect ratio of the ferro-
to investigate the suitability of potential applications of such
electric particle and d33 and d33c are the piezoelectric charge
composites, the tensile properties and high cycle fatigue of the
constant of the composite and ceramic phase respectively [19].
composites for large strain levels has been studied. Finally, for
The piezoelectric voltage constant of the composite, g33 , can
the first time ever it is demonstrated that for a well chosen
be calculated according to the following equation:
self-healing polymer matrix the loss of sensorial functionality
of the composites after high cyclic tensile fatigue can be d33
partially recovered by thermal healing. g33 = (5)
εo ε
2
Smart Mater. Struct. 23 (2014) 055001 N K James et al
(a) (b)
self-healing phenomena was either initiated by high energy of both the as-produced composites and cyclically loaded
impact or a puncture test, such as bullet penetration or by composites were measured. The specimens were fatigue tested
more controlled quasi-static penetration experiments [23] or under different strain levels, by properly selecting the stress
by mild thermal treatments and looking at the healing of ratio defined as the ratio of the minimum stress to the maximum
surface scratches [24]. As shown elsewhere [17], the Zn-based stress. The shape of the loading waveform was sinusoidal and
ionomers show the highest degree of self-healing (i.e. the the employed frequency was restricted to 1 Hz to avoid sample
ability to close cracks or other local mechanical damage upon heating.
application of a modest temperature well below the melting Finally, to measure the functional healing, the degraded
point of the polymer). An additional desirable characteristic of tensile fatigue loaded composites were annealed at 70 ◦ C in a
the ionomeric polymer is the relaxation of the ionic species at conventional circulating hot air oven while the piezoelectric
elevated temperatures which increases the electrical conduc- properties were measured at intermediate times.
tivity significantly, which should be beneficial to the poling
efficiency.
4. Results and discussion
The composites were prepared by mixing a defined
ratio of polymeric matrix and PZT powder in a labscale 4.1. Microstructure
counter rotating twin screw extruder (DSM Xplore Research
Netherlands). The processing temperature was kept at 170 ◦ C The as-produced composites are in the form of opaque whitish
for both the PZT–Zn ionomer and the PZT–EMAA composites. flexible sheets of about 1 mm thickness. An SEM micrograph
The rotating force was set to 4.5 kN. In order to get a of the cross-sectional view of a 30 vol.% filled PZT–Zn
homogeneous distribution of PZT ceramic particles in the ionomer composite is shown in figure 2(a). The micrograph
polymer matrix, the residence time was kept for 5 min, which confirms (figure 2(b)) that fully dense PZT–Zn ionomer
ensured multiple passes of the material through the mixing composites with a spatially uniform distribution of the PZT
section of the extruder. The composites were finally extruded particles were created. The PZT particles are well distributed,
through a 2 mm diameter outlet. In order to get planar sheets show limited agglomeration (hence a 0–3 configuration) and
of composites, the extruded wire-shaped materials were hot adhere well to the polymer matrix which is a key requirement
pressed at temperature of 150 ◦ C with a pressure of 1 MPa for for optimal piezoelectric and mechanical properties. A higher
5 min, yielding flexible planar sheets with a thickness of about magnification SEM micrograph showing the primary particles
1 mm. Electrodes on the samples were made by sputtering inside the composite is given in figure 2(c).
of gold. The samples were poled in an organic oil (rape
seed oil) bath under different poling conditions (electric field 4.2. Poling behavior
strength, temperature and time). The dielectric constants of the
composites were measured using the parallel plate capacitor The effect of poling temperature on the d33 of 30 vol.% filled
method using an Agilent 4263B LCR meter at 1V AC signal PZT–Zn ionomer and PZT–EMAA composites is shown in
and 1 kHz. The d33 of the composites was measured using a figure 3 for a poling field of 15 kV mm−1 and a time of
Berlincourt type d33 meter (KCF technology PM3001). The 60 min. The results are also compared with those for bulk
static tensile properties of the composite were investigated PZT sintered ceramics. It is evident that for the PZT–polymer
using a Zwick Roell 20 kN tensile bench with a 1 kN load cell composites the poling temperature has a marked effect on
and a cross-head speed of 0.5 mm min−1 at room temperature. the d33 , especially around 50–80 ◦ C, whereas the d33 of
Rectangular specimens with dimensions 50 mm × 5 mm × the bulk PZT ceramics remains more or less constant. The
1 mm were tested. The average value over four samples are dielectric breakdown of the composites prevents higher poling
reported. Fatigue tests were conducted at room temperature temperatures being applied. It is clear that PZT–Zn ionomer
on an MTS 831 Elastomer test system. Rectangular specimens composites show significantly better piezoelectric properties
with dimensions 50 mm × 5 mm × 1 mm were prepared and than the corresponding PZT–EMAA composites. The increase
within these samples two circular regions were metalized and of d33 with poling temperature observed for the PZT–Zn
poled at 15 kV mm−1 prior to the fatigue test. The d33 values ionomer composite can be attributed to the relaxation of the
3
Smart Mater. Struct. 23 (2014) 055001 N K James et al
Figure 2. (a) Photograph of 30 vol.% filled ionomer composite, (b) SEM micrograph of cross-sectional view of 30 vol.% filled PZT–Zn
ionomer composite and (c) SEM micrograph showing the primary particles.
4
Smart Mater. Struct. 23 (2014) 055001 N K James et al
800
PZT ceramic
780 PZT- Zn ionomer composite
PZT- EMAA composite
760
740
d33 (pC.N )
-1
0
0 20 40 60 80 100 120 140 160 180 200
Poling time (min)
770
PZT ceramic
765 PZT - Zn ionomer composite
PZT - EMAA composite
760
755
d33 (pC.N )
-1
750
5
4
3
2
1
0 Figure 6. (a) The dielectric constant and (b) loss tangent of 30 vol.%
4 6 8 10 12 14 16 18 20 22 PZT–EMAA and PZT–ionomer composites as a function of
-1
Poling field (kV.mm ) frequency and temperature (the number inside the brackets in the
legend indicates the temperature in ◦ C).
Figure 5. Variation of the piezoelectric charge coefficient (d33 ) as a
function of poling field. 30 vol.% PZT–EMAA and PZT–ionomer composites. For all
conditions the dielectric constant shows an almost frequency
piezoelectric charge coefficient of the composites increases independent behavior with a small decrease in the low fre-
with the poling field for both PZT–Zn ionomer and PZT– quency region. The loss tangent spectra are characterized by a
EMAA composites and reaches a constant level at about peak appearing in the low frequency region of PZT–ionomer
17 kV mm−1 . In PZT ceramics, the maximum is already composites especially at a temperature of 60 ◦ C (i.e. at the
observed at 4 kV mm−1 . The poling field is the driving force selected poling temperature). The strength and frequency of
for the reversal of domains. The higher the field, the faster and relaxation depend on the characteristics of molecular and dipo-
more complete the reorientation of the domains in the direction lar relaxation [26]. In PZT–ionomer composites, the relaxation
of the poling field. In a 0–3 polymer–ceramic composite, and mobility of ionic clusters is evident by the peak shift
a large part of the applied electric field during poling is towards the higher frequency side. The relatively high mobility
obscured by the polymer phase. According to equation (1), at higher temperatures enhances the transport properties which
the effective electric field experienced by the ceramic phase in turn increase the loss tangent of PZT–ionomer composites.
is of the order of 1.7–28 V mm−1 for the range of applied In figure 7 the measured dielectric constant and piezoelec-
electric fields of 1–15 kV mm−1 . Hence, a polymer–ceramic tric charge coefficient of the composites are compared to the
composite requires a high electric field for effective poling of values predicted using Yamada’s model. This model predicts
the piezoelectric phase. Based on these results, the optimum that these properties mainly depend on the volume fraction
poling conditions for the described composites are an electric of PZT and the aspect ratio of the particles. The dielectric
field of 17 kV mm−1 to be applied at a temperature of 60 ◦ C constant and piezoelectric charge coefficient increases signifi-
for 60 min. cantly with increasing PZT volume fraction. The theoretically
Figures 6(a) and (b) show the temperature and frequency predicted values match well with the experimentally observed
dependence of the dielectric constant and loss tangent of data for the composites. The d33 and dielectric constant of the
5
Smart Mater. Struct. 23 (2014) 055001 N K James et al
8 40
g33 (mV.m.N )
-1
6 30
4 20
2 10
0 0
0 10 20 30 0 10 20 30
Volume fraction of PZT loading (%) volume fraction of PZT loading (%)
(a)
Figure 8. Variation of piezoelectric voltage coefficient (g33 ) of the
composites as a function of PZT volume percentage.
and dynamic mechanical properties of such composites were
studied in more detail.
6
Smart Mater. Struct. 23 (2014) 055001 N K James et al
b
Figure 9. Variation of (a) Young’s modulus and (b) elongation and
strength at break as a function of PZT volume percentage in a Figure 10. Piezoelectric charge coefficient of the 30 vol.% filled
Zn-ionomer-based composite. PZT–Zn ionomer composite (a) as a function of the applied strain
during single pass loading and (b) after cyclic loading to three
different maximal strain values.
that such a drop in d33 is caused by delamination between
PZT particles and the polymer matrix as there are no intrinsic
changes in the properties of the constituent phases are to be
expected at room temperature testing. The assumed debonding
is also evident from a SEM micrographs (figure 12(a)) of the
fracture surface of the PZT–Zn ionomer composite.
7
Smart Mater. Struct. 23 (2014) 055001 N K James et al
References
8
Smart Mater. Struct. 23 (2014) 055001 N K James et al
[20] Sakamoto W K, De Souza E and Das-Gupta D K 2001 [24] van der Zwaag S 2007 Self Healing Materials: An Alternative
Electroactive properties of flexible piezoelectric composites Approach to 20 Centuries of Materials Science (Dordrecht:
Mater. Res. 4 201–4 Springer) ISBN 978-1-4020-6250-6 pp 95–114
[21] Capsal J F, Dantras E, Laffont L, Dandurand J and [25] Chure M C, Chen P C, Wu L, Chen B H and Wu K K 2011
Lacabanne C 2010 Nanotexture influence of BaTiO3 Influence of poling conditions on the characteristics of PZT
particles on piezoelectric behavior of PA 11/BaTiO3 ceramics Adv. Mater. Res. 284 1375–80
[26] Kremer F and Schonhals A 2003 Broadband Dielectric
nanocomposites J. Non-Cryst. Solids 356 629–34
Spectroscopy (Berlin: Springer)
[22] Mendiola J, Alemany C, Jimenez B and Maurer E 1985 Poling [27] Carlsson L, Crane R L and Uchino K 2000 Comprehensive
of (Pb, La)(Zr, Ti)O3 J. Mater. Sci. Lett. 4 1383–6 Composite Materials vol 5 (New York: Elsevier) pp 523–32
[23] Varley R J and van der Zwaag S 2010 Development of a [28] Tavman I H 1996 Thermal and mechanical properties of
quasi-static test method to ionomer Polym. Int. aluminum powder filled high density polyethylene
59 1031–8 composites J. Appl. Polym. Sci. 62 2161–7