Effects of humidity on the performance of ionic polymer-metal composite actuators: Experimental study of the back-relaxation of actuators

Article abstract of DOI:10.1021/jp074611q

This article focuses on the dependence of water uptake on the displacement, velocity, mechanical force, and charging profiles of perfluorinated ionomer-platinum/Li⁺-based actuators. Both the displacement and force generation were found to be strongly dependent on the humidity. The primary reason for this effect is a decrease in the stiffness as a result of the humidity. The actuators demonstrated a dramatic reverse motion and a negative force, and this subsequent relaxation was dramatically decreased by decreasing humidity. This relaxation process can be explained by the slow diffusion of water into the elastically softened anode and out of the stiffened camode. There are no clear inflection points on the charging profile during me reverse relaxation, and this suggests that the relaxation process does not involve a major redistribution of counter cations. An increase in water uptake resulted in an enhancement of the velocity of the displacement. A continuous generation of force was also examined by scanning potential, and the force was proportional to the potential. Humidities near 50-60% (i.e., water uptakes of ca. 5 wt %) gave a better actuator bending performance.

Full text of DOI:10.1021/jp074611q

J. Phys. Chem. B 2007, 111, 11915-11920
11915
Effects of Humidity on the Performance of Ionic Polymer-Metal Composite Actuators:
Experimental Study of the Back-Relaxation of Actuators
Eiichi Shoji* and Daisuke Hirayama
Department of Human and AI Systems, Intelligent Materials Science and Technology Laboratory, UniVersity of
Fukui, 3-9-1 Bunkyo, Fukui 910-8507, Japan
ReceiVed: June 14, 2007; In Final Form: July 31, 2007
This article focuses on the dependence of water uptake on the displacement, velocity, mechanical force, and
charging profiles of perfluorinated ionomer-platinum/Li+-based actuators. Both the displacement and force
generation were found to be strongly dependent on the humidity. The primary reason for this effect is a
decrease in the stiffness as a result of the humidity. The actuators demonstrated a dramatic reverse motion
and a negative force, and this subsequent relaxation was dramatically decreased by decreasing humidity.
This relaxation process can be explained by the slow diffusion of water into the elastically softened anode
and out of the stiffened cathode. There are no clear inflection points on the charging profile during the reverse
relaxation, and this suggests that the relaxation process does not involve a major redistribution of counter
cations. An increase in water uptake resulted in an enhancement of the velocity of the displacement. A
continuous generation of force was also examined by scanning potential, and the force was proportional to
the potential. Humidities near 50-60% (i.e., water uptakes of ca. 5 wt %) gave a better actuator bending
performance.
Introduction
SCHEME 1
To date, various polymers, such as polymer gels,1-4 ion-
exchange polymers,5-13 and conducting polymers,14-17 among
others, have attracted attention as active sensors, flexible and
lightweight actuators, artificial muscles, and energy transducers.
For the past decade, ionic polymer-metal composites (IPMCs)
have been extensively studied as new materials. IPMC actuators
consist of a host polymer bound by thin metal surfaces, where
the two metallic surfaces are conductive. The host polymer
usually consists of perfluorinated ionomers such as Nafion (E.I.
DuPont de Nemours & Co. Inc.) and Flemion (Asahi Glass Co.
Ltd.). Scheme 1 shows the chemical structure of Nafion, where
x ) 6.5, y ) 1, z ) 1 corresponds to Nafion 117.18 The
polyelectrolyte matrix was neutralized by the presence of
counterions, balancing the charge of the anions covalently fixed
to the membrane. This polymer has a unique morphological
structure that has a two-phase system of an ion cluster network
surrounded by a hydrophobic poly(tetrafluoroethylene) (PTFE)
medium,19 The cluster network contributes a dynamic bending
performance to the actuators. In various modeling and experi-
mental works, the phase-separated domains have been revealed
to be spherical inverted micellar structures connected by short
narrow channels.20,21 The length of the side chains affects the
separation between the ionic and nonpolar domains, and the
counterions such as Li+ or Na+ are mobile and solvated by water
within nanoclusters with sizes of 3-5 nm. Nobel metals such
as platinum and gold are usually plated as IPMC electrodes on
both sides of the film by an electroless plating method.
and can be used under water and in the air. It is noted that their
mechanical characteristics are similar to those of biological
muscles in terms of flexibility, softness, and ability to undergo
large deformations.22 These unique features are attractive for
potential applications of cutting-edge actuators in artificial
muscles, robotic arms, micromanipulators, and micropumps.23
To understand the actuation mechanism of IPMCs, several
feasible numerical and experimental models based on hypotheses
and analytical chemical processes have been proposed.22 For
example, a model based on the spatial distribution of the ions
in an electric field and a pH gradient,24 an Euler-Bernoulli beam
model,25 a linear dynamic lumped-parameters model,26 a black-
box model based on a network of a capacitor and resistor,27
a
higher-order lumped-parameter model dealing with the relax-
ation phenomena,28 a model based on microstructure of Nafion
and a phase separation morphology,29 a model based on water
transport across the membrane dealing with deformation and
relaxation phenomena,30 and a model based on the concentration
of water molecules dealing with mass transport and chemical
reactions at boundaries have been reported to understand the
deformation of the actuators.22
Although these models are well proposed, a detailed mech-
anism for the generation of the large displacement and back-
relaxation phenomena of the actuator is not yet fully elucidated.
It seems that the mobility of ion-water clusters and water
redistribution affect the deformation and stress relaxation
phenomena. Here, understanding water migration under an
IPMC actuators give larger displacements than other chemical
actuators, and they are lightweight, noiseless, easy to produce,
and generate no EMI (electromagnetic interference). They can
be operated at relatively low voltages in the range of 1-3 V
* To whom correspondence should be addressed. E-mail: shoji@
chem.his.fukui-u.ac.jp.
10.1021/jp074611q CCC: $37.00 © 2007 American Chemical Society
Published on Web 09/21/2007

11916 J. Phys. Chem. B, Vol. 111, No. 41, 2007
Shoji and Hirayama
applied electric field is an important subject for IPMC-type
actuators. To develop an understanding of actuator performance,
more experimental findings regarding the effects of water uptake
on the performance of actuators are still needed.
The purpose of the present contribution is to report on such
experimental findings and the performance of IPMC-type
actuators. In this work, the effects of water migration on the
behavior of the displacement were examined by changing the
humidity to manipulate the water uptake in the film. Profiles
of the displacement, velocity, and force were carefully examined.
The stress relaxation process of an actuator film due to water
redistribution could be a key subject for understanding the
displacement and force processes.
SCHEME 2
Instruments Inc.). The programmable power supply was a
monopolar system that was converted to a bipolar system. The
delay of the bipolar switching was less than 5 ms, and this timing
was carefully adjusted by the software. Both the current and
the potential were monitored through the DAQ. The measure-
ment of the displacement was performed in a humidity-
controlled chamber. The length of the actuator was carefully
set to 16 mm on the stage by clamping 2 mm of the 18-mm
actuator. The displacement was evaluated horizontally at 12 mm
from the root. The mechanical force of the actuators was
monitored by a balance having a homemade attachment to the
actuators. All of the experiments were conducted at 25 °C.
Experimental Section
Materials. Nafion 117 (equiv wt 1100, thickness ca. 180 µm),
sodium borohydride (>98% grade), and [Pt(NH3)4]Cl2 [tet-
raamineplatinum(II) chloride, >98%] were purchased from
Aldrich and used as received. Nitric acid (concentrated, 70%)
and lithium hydroxide monohydrate (>98%) were purchased
from Wako Pure Chemical Industries, Ltd. Water (18 MΩ) from
a Millipore purification system was used for all experiments
unless otherwise noted.
Results and Discussion
Preparation of the Actuator. The fabrication of IPMC-type
actuators includes three basic stages: (a) ion exchange from
H+ to the metal complexes in the matrix of Nafion, (b) chemical
reduction using ionic reductants (i.e., surface metallization can
be done if the reducing agent is negatively charged for a static
repulsive inhibition of the reaction in the film), and (c) ion
exchange to the final form.5,32,33 In the first process, Nafion
film was soaked in [Pt(NH3)4]2+ solution to allow this platinum
salt diffused into the film. Then, the film was transferred to a
solution containing the anionic reduction agent BH4-. Scheme
2shows the primary reaction of the deposition of platinum. It is
known that the penetration of the conductive layers onto Nafion
results in dendritic branching and fractals.33 After the metal-
lization reaction, the film was finally soaked in lithium
hydroxide solution to allow for the cation exchange of lithium
and well washed and rinsed with water.
Actuator Performances. A square waveform of +2 and -2
V was applied to the actuator electrodes. The interval of each
step was 20 s. As reported in many journal articles,34 a bending
motion toward the anode occurs as a result of an electric field.
Here, it is known that the velocity of the actuation strongly
depends on the species of cations such as Li+, Na+, K+, Rb+,
Cs+, and tetraalkylammonium cations.34 The relaxation time in
this work means the period during which the bending motion
of the actuator is completely halted without manipulating the
electric field. In fact, an interval time of 20 s might not be
enough time to achieve full relaxation. In the strict sense of
relaxation, it will take several hundreds of seconds to achieve
full relaxation. To avoid unexpected deterioration of the
electrode upon application of a potential for a long period, we
limited this time to 20 s. During this time period, the relaxation
bending process was clearly recorded.
As shown in Figure 1, the water uptake of IPMC was
examined as a function of the relative humidity. Clearly, the
film absorbs water effectively as a function of the humidity,
and the maximum variation in the mass gain was ca. 22% at
98% RH. The tendency of the mass change of water uptake
agrees with similar experimental results for Nafion membranes.35
The average number of water molecules per sulfonic acid group
at 0% RH has been reported to be 1.55.35 In the humidity range
of 0-50%, an additional water molecule was added to each
sulfonic acid group. The ratio of H2O/SO3H was ca. 3-5. At
higher humidities, this process becomes much more rapid, and
water finally starts to condense on all regions of the membrane
until the resulting pressure within its clusters is balanced by
the elastic stresses that develop within the membrane. Nemat-
Nasser et al. previously reported the relationship between
Preparation of the Actuator. The preparation method used
herein is based on a reported procedure.31 The IPMC composite
film was carefully washed and soaked in pure water and used
as freshly fabricated. In this study, Li+ form was examined for
all experiments. Electroless plating was carried out to deposit
platinum on both surfaces of a pure Nafion film. To clean the
surfaces of the film, the film was immersed in a beaker of a
boiling water/nitric acid (1/2) mixture for 30 min. This process
also removed organic contaminants in the material. Then, the
film was washed with water and boiled in water for 30 min.
This process ensured complete hydration. Then, the film was
soaked in a 0.01 M tetraamineplatinum(II) chloride solution at
25 °C for 2 h with agitation. The film was then rinsed with
water and transferred to freshly prepared 0.01 M sodium
borohydride at 25 °C for 2 h. Then, the film was rinsed with
water, immersed in 1 M nitric acid solution for 30 min, and
boiled in water for 30 min. This plating process was repeated
three times. The film was finally soaked in 0.1 M lithium
hydroxide solution for 10 h and rinsed well with water so that
the counter cations were exchanged to Li+.
Water Uptake of IPMC with Humidity. To identify a
possible relationship between water uptake and relative humid-
ity, first, a piece of the IPMC actuator (3 × 20 mm) was dried
at 100 °C in vacuum for 10 h and weighed (21.1 mg). Then,
the film was suspended in an environment with 30% relative
humidity (RH). The humidity was increased using water vapor,
the film was allowed to equilibrate and then weighed, and the
process was then repeated. A graph of percentage mass gain
versus humidity was plotted from the data.
Evaluation of Dynamic Performance. We developed a
quantitative static and dynamic performance evaluation system
especially for chemical actuators such as IPMC actuators.
Because the actuator undergoes a large bending motion, a CCD
camera was employed to follow the displacement. The evalu-
ation system was composed of a programmable power supply
(HP model 6632B), data acquisition board (DAQ) (National
Instruments Inc.), CCD camera, image processing board, and
balance. The images from the camera were preprocessed through
the image processing board. The waveform of the potential
applied to the actuators was fully controlled by the software.
Here, homemade software was developed in LabView (National

Performance of Ionic Polymer-Metal Composite Actuators
J. Phys. Chem. B, Vol. 111, No. 41, 2007 11917
Figure 1. Percentage mass gain versus relative humidity for the
actuator film.
Figure 3. Maximum forward and backward displacements versus
humidity, showing (a) each gap of the forward displacement (open
symbols) and (b) each gap of the backward displacement (solid
symbols).
Figure 2a clearly shows that the actuator exhibits a dramatic
backward motion (i.e., a back-relaxation process34,36) just after
the maximum forward displacement. This subsequent reverse
motion is dramatically suppressed by decreasing the humidity.
The theoretical and experimental frameworks have been dis-
cussed for several models of the mechanism of IPMC actua-
tion.22,34,36,37 Hydraulic models of IPMC actuation are important
to understand this reverse mechanism. That is, when the potential
is applied, the voltages cause the migration of ions (i.e., cations
redistribute and migrate toward the cathode), and the hydraulic
effect is assigned to the motion of water molecules that are
contributing ions by solvation. As seen in Scheme 1, because
Nafion has a hydrophobic backbone of fluorocarbon polymer
that is separated from the hydrophilic clusters containing
unbound solvated cations and fixed sulfonate groups, water
redistribution affects the cluster volume and distribution of
cations. These changes also impact the stiffness of the film.
The stiffness is inversely proportional to the hydration volume,34
that is, an increase in water uptake leads to a decrease in the
stiffness. The secondary hydraulic effect involves electro-
osmotic effects that are derived from the initial migration of
ions.36 After the migration of both cations and water toward
the cathode, subsequent redistribution and reconfiguration of
those species can be promoted by a slow buildup of a voltage
and electro-osmotic effects.36,38 The back-relaxation comes from
the rearrangement of small mobile molecules.36 From these
findings, the slow diffusion of water into the elastically softened
anode and out of the stiffened cathode can be considered to
give the slow reverse relaxation.
The dependences of the maximum gaps of the forward
displacement (a, open circles) and the backward displacement
(b, solid symbols) on RH are shown in Figure 3. Both the
forward and reverse displacements are increased by RH,
primarily because of a decrease in the stiffness. Moreover, it is
noted that the forward motion is strongly influenced by RH.
Here, the two plots are very close when the RH is less than ca.
40%, and the reverse motion is eventually weakened.
Figure 4 shows the dependence of the charge on the humidity.
Basically, the charge during the actuation is based on a
compensation for migration of cations, faradic process for the
electrolysis of water, and/or unexpected redox reactions of the
electrode. A potential of 2 V is being applied to the electrodes
in the experiment, and this potential is more than the value of
ca. 1.23 V that is the theoretical potential for water electrolysis.
Thus, except for unexpected degradation, both compensation
Figure 2. Displacement versus time profiles under various humidity
conditions: (a) 90% (solid line), (b) 50% (dotted line), and (c) 30%
(solid line).
hydration volume and stiffness as “stiffness is in inverse
proportion to hydration volume”.36 That is, from the relation
between the cross-sectional area and the hydration volume, the
sample’s cross-sectional area can be estimated at various
hydration levels by simply measuring its weight. The mass of
the sample can be calculated by the corresponding hydration
volume and then the cross-sectional area. These results gave
the stiffness as a function of hydration volume. Because there
is no information on any relation between the hydration volume
and the hydration state, however, it is difficult to correlate
hydration state to stiffness quantitatively. Thus, our discussion
here was to point out that the stiffness can be decreased if the
hydration (state) is enhanced by humidity, because the increase
in hydration volume is directly based on the absorption of water.
Figure 2 shows the displacement under various RH conditions
of 90% (water uptake ca. 18%, solid line, curve a), 50% (ca.
3%, dotted line, curve b), and 30% (ca. 1%, solid line, curve c)
obtained by applying a potential. Clearly, the displacement
strongly depends on the humidity (water uptake), and curve a
shows the largest displacement of ca. (3.5 mm. This value is
nearly 10 times larger than the (0.3 mm value for curve c at
30% RH. The result agrees with the finding that the displace-
ment depends on the level of hydration.36 Here, the nature of
the transport of water and hydrated ions within the IPMC film
affects the moduli at different frequencies.33 Thus, the effect of
the humidity on the stiffness34 (modulus of elasticity) should
be an important factor in facilitating the bending motion and
back-relaxation. We will discuss this behavior further
below.

11918 J. Phys. Chem. B, Vol. 111, No. 41, 2007
Shoji and Hirayama
Figure 4. Dependence of charge on humidity.
of ions and electrolysis of water are expected to be a primary
source of the charge. The charge would also depend on the
mobility of the ions, the level of hydration, and the form of the
cations. Given that the charge of ca. 0.25C near 90% RH is
about 3 times higher than the charge of 0.09C near 20% RH,
the increase in the charge is mainly due to the higher mobility
of the hydrated cations and the electrolysis of water. Regarding
the electrolysis of water, even if all of the 0.25C were being
used for the electrolysis, the loss of water is estimated to be
only 0.05 mg. Thus, this amount of water can easily be supplied
under 90% RH conditions, so that the actuator can repeat the
bending motion continuously at 2 V.
The velocities of both the forward and backward displace-
ments are proportional to the humidity (Figure 5). Note that
the velocities in Figure 5A are based on average velocities
during forward and backward displacements, and the velocities
in Figure 5B are based on the maximum differential velocities
for forward and backward displacement. As can be seen in
Figure 5B, the maximum forward velocity is ca. 14 mm/s, and
this velocity is about 10 times faster than the backward
displacement of 1.3 mm/s. The stiffness of the film is decreased
by water uptake, but the diffusion of cations can be facilitated.34
Therefore, as shown in Figure 5, an increase in water uptake
results in an enhancement of the velocity of the displacement.
Figure 5. Velocity versus humidity for both forward and backward
motion during actuation. The velocity was evaluated as (a) the average
of the forward and backward displacements and (b) the maximum
velocity of the forward and backward displacements.
Figure 6 shows the time course of the displacement as the
RH is changed from ca. 30% to ca. 90%. The displacement at
90% RH is about 10 times larger than the displacement at 40%
RH. Note that a back-relaxation can be gradually observed with
increasing RH above 50%. Above ca. 60% RH, the IPMC
actuator shows a remarkable backward displacement that is
nearly one-half of the maximum displacement. Figure 7A and
7B shows the time course of the force as a function of humidity,
and a back-relaxation behavior can be recognized in the force
generation. The maximum peak forces are 1 and 3 mN at 50%
and 90% RH, respectively. This finding indicates that the
hydration impacts an internal stress (i.e., modulus) of the film
and this stress affects both the displacement (i.e., without
loading) and the force (i.e., with loading). In the time course of
the force generation, three patterns of profiles without relaxation
(a), in transient (b), and with relaxation (c) can be seen in Figure
7A. Note that each scale is not the same in the comparison.
The transient is located near ca. 50% RH. A similar transient
can also be recognized in the displacement in Figure 6 at ca.
50% RH. Both the displacement and the force are enhanced by
an increase in the humidity. Thus, the force can be increased if
the displacement is increased. A highly hydrated actuator gives
a larger bending displacement because of a decrease in the
Figure 6. Displacement versus time under a (2 V square waveform.
The interval of each step is 20 s. The humidity was controlled from
ca. 30% to ca. 90%.
stiffness and much more water transport. After the larger
bending motion, a dramatic backward motion is seen clearly.
The stiffness of the film, the hydration state of the film (for
both Nafion and mobile ions), and the hydration volume are
affected by humidity. Thus, the relaxation process should be
influenced by at least these parameters and should be intricate.
If the humidity is increased, however, the relaxation time would
be decreased as a result of a decrease in the stiffness, giving a
smooth bending motion of the film. By compiling the results
of the displacement and force profiles, the bending phenomena
are revealed to be a result of internal stress of the film (or vice

Performance of Ionic Polymer-Metal Composite Actuators
J. Phys. Chem. B, Vol. 111, No. 41, 2007 11919
Figure 7. Force versus time and humidity under the (2 V square
waveform shown in Figure 1. The humidity was controlled from ca.
30% to ca. 90%. Three types of profiles (a-c) are shown. The time
region of a from 1000 to 2500 s is expanded to b. In both plots a and
b, the force after 20 s at each cycle is evaluated as a net force, and the
net force is indicated by the supported dotted lines.
Figure 8. Force versus time and humidity. Data in Figure 8 are partially
enlarged and shown at humidity values of (a) 80%, (b) 50%, and (c)
30%.
versa). Here, because a reaction to the stress necessitates a
deformation of the film, both the displacement and the force
generation would lead to a back-relaxation. In terms of an
evaluation of the durability performance of the actuators, we
focused on a metastable force during the back-relaxation process.
In comparison to a quick forward displacement within seconds,
this relaxation approaching a metastable state required tens or
even hundreds of seconds.36 To elucidate the metastable state,
supported dotted lines were added to Figure 6A and 6B to
indicate the force after 20 s at each cycle. As can be seen, the
maximum plateau exists near 50% RH.
Force profiles a-c in Figure 7A are enlarged and presented
in Figure 8with charging profiles at each cycle. Figure 8A-C
corresponds to force profiles around 80%, 50%, and 30% RH,
respectively. The maximum force and back-relaxation processes
during force generation are very sensitive to the humidity. Under
lower-humidity conditions (for example, ca. <50%), the actuator
becomes “rigid”, and the total displacement (i.e., bending
performance) gets smaller. Under higher-humidity conditions
(ca. >50%), the actuator becomes “soft”, and the total displace-
ment increases. Thus, the production of the force naturally
relates to a balance of the rigidness and the bending performance
of the film. One question might arise as to whether a larger
force could be produced by an actuator that is rigid and has
smaller bending capability or by an actuator that is soft and
has a larger bending capability. In the comparison of experi-
mental data in Figure 8A and 8C, a capability giving larger
displacement is of primary importance in obtaining a larger
force. In the charging profiles in Figure 8, the charge is balanced
by the force generation and the humidity. During the force
production consuming 0.15 C near 90% RH in Figure 7A, the
charge is continuously consumed without any inflection points
during the relaxation process. Here, the charge consumption can
basically be understood as an accumulation of charge compen-
sation for counterions, electrolysis of water, unexpected stray
currents through the film, or platinum-related redox processes
in the film. As long as charge compensation is for the migration
of ions by the electric field conditions, having no clear inflection
point on the charging process during back-relaxation suggests
that the back-relaxation process might involve not the charge
redistribution of ions but, rather, water redistribution. The
experimental finding here would relate to water redistribution
based on osmotic pressure and stiffness/hydration effects.34,36
We have evaluated continuous force generation by scanning
the potential at 50 mV/s from 0 to 3.5 V. Before this experiment,
we confirmed that potentials above 4 V were too high and
caused serious damage to the actuator. Thus, the potential was
limited to 3.5 V. Figure 9A shows the profile of the force under
40% RH, for which the maximum force is 2.6 mN at 3.5 V.
Figure 9A also shows a long-period back-relaxation giving a
negative force after open circuit. This negative force would come
from an accumulated internal stress (as a modulus) during
continuous force production. We are currently carefully focusing
on this process, and the results will be reported elsewhere. The
same experiments as depicted in Figure 9A were repeated under
various RH values, giving the results shown in Figure 9B.
Interestingly, Figure 9B demonstrates that the actuator can
produce a maximum force of ca. 3.7 mN near 50% RH. Again,
this would be a result of a balance of the rigidity and bending
capability of the film. This finding clearly suggests that the

11920 J. Phys. Chem. B, Vol. 111, No. 41, 2007
Shoji and Hirayama
(2) Eichenbaum, G. M.; Kiser, P. F.; Simon, S. A.; Needham, D.
Macromolecules 1998, 31, 5084-5093.
(3) Tanaka, T.; Fillmore, D.; Sun, S. T.; Nishio, I.; Swislow, G.; Shah,
A. Phys. ReV. Lett. 1980, 45, 1636-1639.
(4) Harmona, M. E.; Tangb, M.; Frank, C. W. Polymer 2003, 44 (16),
4547-4556.
(5) Millet, P.; Pineri, M.; Durand, R. J. Appl. Electrochem. 1989, 19,
162-166.
(6) Sadeghipour, K.; Salomon, R.; Neogi, S. Smart Mater. Struct. 1992,
1, 172-179.
(7) Oguro, K.; Asaka, K.; Takenaka, H. In Proceedings of the 4th
International Symposium on Micro Machine and Human Science; IEEE
Press: Piscataway, NJ, 1993; p 39.
(8) Asaka, K.; Oguro, K.; Nishimura, Y.; Mizuhata, M.; Tanaka, H.
Polym. J. 1995, (27), 436-440.
(9) Shahinpoor, M.; Bar-Cohen, Y.; Simpson, J. O.; Smith, J. Smart
Mater. Struct. 1998, 7 (6), R15-R30.
(10) Onishi, K.; Sewa, S.; Asaka, K.; Fujiwara, N.; Oguro, K. Electro-
chim. Acta 2001, 46 (8), 1233-1241.
(11) Kim, K. J.; Shahinpoor, M. Polymer 2002, 43, 797-802.
(12) Noh, T.-G.; Tak, Y.; Nam, J.-D.; Choi, H. Electrochim. Acta 2002,
47, 2341-2346.
(13) Jung, K.; Nam, J.; Choi, H. Sens. Actuators A 2003, 107, 183-
192.
(14) Baughman, R. H. Synth. Met. 1996, 78 (3), 339-353.
(15) Mazzoldi, A.; Santa, A. D.; Rossi, D. D. In Polymer Sensors and
Actuators; Osada, Y., De Rossi, D. E., Eds.; Springer-Verlag: Heidelberg,
Germany, 1999.
(16) Kaneko, M.; Fukui, M.; Takashima, W.; Kaneto, K. Synth. Met.
1997, 84, 795-796.
(17) Madden, J. D.; Cush, R. A.; Kanigan, T. S.; Brenan, C. J.; Hunter,
I. W. Synth. Met. 1999, 105, 61-64.
(18) Kawano, Y.; Wang, Y.; Palmer, R. A.; Aubuchon, S. R. Polim.:
Cienc. Tecnol. 2002, 12 (2), 96-101.
(19) Shahinpoor, M.; Kim, K. J. Smart Mater. Struct. Int. J. 2005, 14
(1), 197-214.
(20) Heitner-Wirguin, C.; Bauminger, E. R.; A., L.; Kanter, F. B.; Ofer,
S. Polymer 1980, 21, 1327-1329.
Figure 9. Force measurement of a Nafion-Pt-based actuator at 40%
RH. The potential was scanned from 0 to 3.5 V at 50 mV/s. (A) The
maximum force was evaluated at 3.5 V, and then the potential was set
to 0 V to record a negative force generation. (B) Dependence of the
force at 3.5 V on the humidity. The experimental procedure was the
same as for A, and the force was evaluated at various humidities.
(21) Hsu, W. Y.; Timothy, T. D.; Gierke, D. Macromolecules 1982,
15, 101-105.
humidity (water uptake) has to be carefully considered to obtain
the maximum performance of IPMC-type actuators.
(22) Enikov, E. T.; Seo, G. S. Sens. Actuators A 2005, 122, 264-272.
(23) Lee, J. H.; Lee, J. H.; Nam, J.-D.; Choi, H.; Jung, K.; Jeon, J. W.;
Lee, Y. K.; Kim, K. J.; Tak, Y. Sens. Actuators A 2005, 118, 98-106.
Conclusion
(24) Shahinpoor, M. In Proceedings of the SPIE Conference on Smart
Structures and Materials; SPIE Press: Bellingham, WA, 1993; Vol. 1916,
pp 40-50.
(25) Shahinpoor, M. In Proceedings of SPIE Conferences on Electro-
actiVe Polymer Actuators and DeVices; SPIE Press: Bellingham, WA, 1999;
Vol. 3669, pp 109-121.
(26) Kanno, R.; Kurata, A.; Hattori, M.; Tadokoro, S.; Takamori, T. In
Proceedings of the Japan-U.S.A. Symposium on Flexible Automation;
ISCIE/ASME: New York, 1994; Vol. 2, pp 691-698.
(27) Kanno, R.; Tadokoro, S.; Takamori, T.; Hattori, M. In Proceedings
of the IEEE International Conference on Robotics and Automation; IEEE
Press: Piscataway, NJ, 1996; Vol. 1, pp 219-225.
(28) Bao, X.; Bar-Cohen, Y.; Lih, S. In Proceedings of the SPIE Smart
Structures and Materials Symposium; SPIE Press: Bellingham, WA, 2002;
pp 4695-4727.
(29) Nemat-Nasser, S.; Li, J. Y. J. Appl. Phys. 2000, 87 (7), 3321-
3331.
Both forward and reverse displacements were dramatically
enhanced by humidity. For instance, the forward displacement
at 90% RH was nearly 10 times larger than the displacement at
30% RH. The primary reason for this observation is a decrease
in stiffness resulting from water uptake. Here, water molecules
were added to each sulfonate group, and at higher humidities,
all regions of the membrane were condensed by water until the
resulting pressure within its clusters was balanced. The hydraulic
effect involves electro-osmotic effects that are derived from the
migration of ions. These events enhance the velocity of the
displacement and also promote a reverse relaxation process. The
reverse relaxation process does not involve a major redistribution
of counter cations. A reverse relaxation was also recognized
during force generation, and this relaxation was dramatically
sensitive to the humidity. During continuous force loading, the
accumulated internal stress in the film gives a negative force.
These results are consistent with the view that the bending
phenomena are a result of the balance of internal stress and
water transport in the film.
(30) Shahinpoor, M.; Kim, K. J. Smart Mater. Struct. 2004, 13 (6),
1362-1388.
(31) Kucernak, A. R.; Muir, B. Electrochim. Acta 2001, 46, 1313-1322.
(32) Takenaka, H.; Torikai, E.; Kawami, Y.; Wakabayashi, N. Int. J.
Hydrogen Energy 1982, 7, 397-403.
(33) Shahinpoor, M. Electrochim. Acta 2003, 48 (14-16), 2343-2353.
(34) Nemat-Nasser, S.; Wu, Y. J. Appl. Phys. 2003, 93 (9), 5255-5267.
(35) James, P. J.; Elliott, J. A.; McMaster, T. J.; Newton, J. M.; Elliott,
A. M. S.; Miles, M. J.; Hanna, S. J. Mater. Sci. 2000, 35, 5111-5119.
(36) Nemat-Nasser, S. J. Appl. Phys. 2002, 92 (5), 2899-2915.
(37) Shahinpoor, M. Electrochim. Acta 2003, 48 (14-16), 2343-2353.
(38) Bar-Cohen, Y. ElectroactiVe Polymer (EAP) Actuators as Artificial
Muscles; SPIE Press, 2001.
Acknowledgment. This work was partially supported by
research grants from the University of Fukui.
References and Notes
(1) Osada, Y.; Gong, J. Prog. Polym. Sci. 1993, (18), 187.