Dielectric elastomer transducers with enhanced force output and work density

Hristiyan Stoyanov, Paul Brochu, Xiaofan Niu, Enrico Della Gaspera, and Qibing Pei Citation: Appl. Phys. Lett. 100, 262902 (2012); doi: 10.1063/1.4730953 View online: http://dx.doi.org/10.1063/1.4730953 View Table of Contents: http://apl.aip.org/resource/1/APPLAB/v100/i26 Published by the American Institute of Physics.

Related Articles
Development of a surface plasmon resonance and nanomechanical biosensing hybrid platform for multiparametric reading Rev. Sci. Instrum. 84, 015008 (2013) Tunnel-field-effect-transistor based gas-sensor: Introducing gas detection with a quantum-mechanical transducer Appl. Phys. Lett. 102, 023110 (2013) Harnessing electromechanical membrane wrinkling for actuation Appl. Phys. Lett. 101, 171906 (2012) Advances for dielectric elastomer generators: Replacement of high voltage supply by electret Appl. Phys. Lett. 101, 162901 (2012) Capacitive micromachined ultrasonic transducer based tilt sensing Appl. Phys. Lett. 101, 153502 (2012)

Additional information on Appl. Phys. Lett.

Journal Homepage: http://apl.aip.org/ Journal Information: http://apl.aip.org/about/about_the_journal Top downloads: http://apl.aip.org/features/most_downloaded Information for Authors: http://apl.aip.org/authors

Downloaded 05 Mar 2013 to 148.206.47.72. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

APPLIED PHYSICS LETTERS 100, 262902 (2012)

Dielectric elastomer transducers with enhanced force output and work density
Hristiyan Stoyanov,a) Paul Brochu,a) Xiaofan Niu, Enrico Della Gaspera, and Qibing Peib)
Soft Materials Research Laboratory, Department of Materials Science and Engineering, University of California Los Angeles, 420 Westwood Plaza, Los Angeles, California 90095-1595, USA

(Received 25 April 2012; accepted 4 June 2012; published online 25 June 2012) We demonstrate that the force output and work density of polydimethylsiloxane (PDMS) based dielectric elastomer transducers can be signicantly enhanced by the addition of high permittivity titanium dioxide nanoparticles. The nanocomposites are capable of maintaining the actuation strain performance of the pure PDMS at relatively low electric elds while increasing the force output and work density due to mechanical reinforcement. A model relating the Maxwell stress to the measured force from the actuator was used to determine the dielectric permittivity at high electric elds thus providing results that can be directly correlated to device performance. This approach toward higher work density materials should enable smaller, lighter, and less intrusive actuator C systems ideal for biomedical and robotic devices in particular. V 2012 American Institute of Physics. [http://dx.doi.org/10.1063/1.4730953] Dielectric elastomer transducers are of great interest, particularly for elds where soft actuators with muscle-like characteristics are required.13 Promising applications such as self-commutating articial muscle motors,4,5 soft generators,6,7 haptic displays,2 as well as robotic and biomedical devices8,9 have emerged; however the broad utilization is still limited by the requirement of high voltages for actuation and low work density. Soft elastic materials able to generate high work density at relatively low voltages are strongly desirable and can accelerate progress. The general approach toward tackling these issues has been to increase the permittivity of the elastomer material, thereby increasing the electrostatic pressure produced for a certain voltage and thus increasing the strain performance at low electric elds. Several methods have been explored with promising results, including the addition of high dielectric constant nanoparticles10,11 or encapsulated conductive nanollers1215 and chemical modication of the elastomer via grafting/blending of polarizable side chains1618 or high dipole moment moieties.19 Among these approaches, permittivity increase due to hyperelectronic polarization of grafted side chains or molecular dipoles has proven to be an effective way to enhance the energy density and the actuation sensitivity while preserving the elastic properties of the host elastomer. High dielectric constant nanoparticles and encapsulated conductive nanollers have shown the tendency to reinforce the polymer host offsetting the permittivity increase and thus leading to moderate improvement of actuation sensitivity.10,13 Proper tuning of the mechanical properties via crosslinking density optimization has to be performed to obtain signicant improvement in actuation sensitivity.15 While reducing the required electric elds is indeed a noble and perhaps necessary task, dielectric elastomers also require high strain, high force, and high work density to compete with current state of the art technologies. A higher
a)

b)

H. Stoyanov and P. Brochu contributed equally to this work. Electronic address: qpei@seas.ucla.edu.

work density allows for a reduction in the size, mass, and intrusiveness of actuator devices. We approach the issue from a different perspective: rather than focusing solely on attempting to reduce the driving voltage by increasing the permittivity, we choose a soft elastomer host that can produce relatively high strains at low to moderate electric elds and use the nanoparticle composite approach to increase the force output and work density. The mechanical reinforcement and increase in permittivity, caused by the addition of the nanoparticles to the elastomer host, work in tandem to increase the overall performance of our composite material. We show that indeed, this nanocomposite approach leads to signicant enhancements of the actuator blocking force and work density while maintaining relatively low electric eld operation, high actuation strains, and high breakdown elds. For the host elastomer, we selected a high temperature vulcanizing (HTV) two part platinum catalyzed PDMS with a low elastic modulus (Wacker Silpuran 6000/10). In order to prevent aggregation of the nanoparticles, we have selected 20 nm rutile TiO2 nanoparticles (er 128) that have been functionalized with short PDMS chains. The tendency of the nanoparticle ller to reinforce the PDMS was offset by decreasing the crosslink density of the PDMS. This was achieved by mixing unequal parts of the two part silicone; the optimal ratio to produce the softest host was a 1:9 ratio of part A to part B. Composites were prepared by dissolving the two silicone parts in isopar G (ExxonMobil), adding the TiO2 nanoparticles to part B and sonicating in a 400 W bath sonicator for 1 h. The solutions were cooled and the part A solution was added and briey sonicated to homogenize the mixture. The solutions were drop cast onto glass and cured at 160  C for 30 min. Linear actuators were fabricated by prestraining the resulting lms by stretches ax ay 1.75 to prevent any issues related to electromechanical instability20,21 and attaching them to rigid frames as seen in Fig. 3 (inset). Electrodes were patterened by spraying a dispersion of single walled carbon nanotubes (P3-SWCNTs) in a combination of 80% isopropyl alcohol and 20% deionized water through a contact mask. The resulting freestanding linear
C V 2012 American Institute of Physics

0003-6951/2012/100(26)/262902/3/$30.00

100, 262902-1

Downloaded 05 Mar 2013 to 148.206.47.72. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

262902-2

Stoyanov et al.

Appl. Phys. Lett. 100, 262902 (2012)

FIG. 1. Blocking force as a function of electric eld measured at room temperature for varying TiO2 loadings.

FIG. 2. Elastic modulus and mechanical loss tangent tan d in the frequency range of 0.1 to 20 Hz.

actuators were used for electromechanical characterization including blocking force measurements, dynamic mechanical analysis, and constant load linear actuation tests. Blocking force measurements were performed by loading the linear actuators in a TA Instruments RSAIII dynamic mechanical analysis machine (DMA). The samples were preloaded to maintain the initial prestrain and allowed to stress relax for several hours prior to testing. A high voltage power supply built in-house was used to apply high voltage across the lm from 0 V to failure using a 200 V step. The device was maintained under isostrain conditions and the blocking force was monitored using the DMAs load cell. Fig. 1 presents the blocking force measured at different electric elds for systematic variation of the nanoparticle content. The blocking force is observed to progressively increase with TiO2 loading, reaching a value of 705 mN for 20%wt which corresponds to 2.2 times increase compared to the pure PDMS (320 mN). Importantly, no decrease in the breakdown eld strength was observed up to a loading of 20%wt. This is important as both electrostatic pressure and strain vary with the square of the electric eld and the energy density varies with the fourth power of electric eld.3 The blocking force measured in the direction perpendicular to the applied electric eld (Fy) is directly related to the Maxwell pressure generated in the thickness direction and for prestrained actuators can be derived for physically measurable quantities as shown and experimentally validated by Kofod et al.22 Fy x 0 z0 x0 eo er E2 a2 ay eo er V 2 ; ay z0 x (1)

where x0 is the actuator width, z0 is the lm thickness before pre-strain, ax and ay are the stretch ratios in width and length direction, respectively, eo is the vacuum permittivity, er is the relative permittivity of the dielectric lm, E is the electric eld, and V is the applied voltage. The blocking force is a function of the applied voltage, permittivity of the active material, and geometry of the actuator only. It provides an efcient way to determine the permittivity of the composites at high electric eld, since all other parameters are known. In order to determine the change in permittivity of the composite, the blocking force data were t with the force equation

(Fig. 1 (black lines)). The results of the t are plotted in the inset of Fig. 1 as a relative increase over the pure PDMS. A substantial increase of 2.7 times is observed for relatively low weight fractions. The advantage of using this method to determine the permittivity of the composite over other standard methods, such as dielectric spectroscopy and LCR meters, is that the measurements are made at high electric elds (0-300 V/lm compared to 0.001 to 0.1 V/lm for standard methods) under normal operating conditions for dielectric elastomers and thus provide results that can be directly correlated to device performance. As reported previously, dielectric permittivity is a complex parameter and measurements made at low electric elds for composite materials can give misleading results due to increases in the dielectric loss (tan d) related to Maxwell-Wagner polarization or increased conductivity which leads to substantial decreases in the electric eld that the material can support and thus greatly reducing overall performance.23 The mechanical properties of the material were characterized using the same setup and actuator conguration under a dynamic load in the frequency range of 0.1 to 20 Hz at a strain of 1% (Fig. 2). As expected, the stiffness increases with TiO2 nanoparticle loading as the nanoparticles act to restrict chain motion by effectively acting as physical crosslinking sites. The elastic modulus is seen to increase from 1.3 MPa at 1 Hz for the pure material to 3.9 MPa for the composite with 30%wt loading. Importantly, the loss tangent is observed to remain essentially unchanged from the neat PDMS up to loadings of 30%wt (and perhaps higher), indicating that the TiO2 had no effect on the viscoelastic behavior of the material. The actuation strain was measured by suspending a load from the bottom of the devices and applying a voltage ramp at a rate of 20 V/s and monitoring the elongation of the active region using digital video and LABVIEW software. The results of the actuation tests (Figure 3 top) show an increase in the actuation strain sensitivity for loadings of 5%wt and 10%wt over the pure PDMS, indicating that at low loadings the inuence of the nanoparticles on the permittivity of the composite outweighs their inuence on the mechanical properties. Above 10%wt the added mechanical reinforcement begins to dominate and the strain response is slightly

Downloaded 05 Mar 2013 to 148.206.47.72. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

262902-3

Stoyanov et al.

Appl. Phys. Lett. 100, 262902 (2012)

ment of work density. The increase in work density is very valuable for most actuator applications as it decreases the required size and mass of the actuator device to perform a task, allowing for smaller, lighter weight, and less intrusive devices. In conclusion, we have demonstrated that the addition of high permittivity nanoparticle llers to a soft high breakdown strength silicone host can yield high work density and high strain actuation at relatively low electric elds. This approach seems well suited for the development of dielectric elastomer articial muscle actuators for biomedical and mobile robotic applications where low mass, small size, and low intrusiveness are of the utmost importance. The work was supported by General Motor Corporation and the University of California Discovery Program.
R. Pelrine, R. Kornbluh, Q. Pei, and J. Joseph, Science 287, 836 (2000). F. Carpi, S. Bauer, and D. De Rossi, Science 330, 1759 (2010). 3 P. Brochu and Q. Pei, Macromol. Rapid Commun. 31, 10 (2010). 4 I. A. Anderson, T. C. H. Tse, T. Inamura, B. M. OBrien, T. McKay, and T. Gisby, Appl. Phys. Lett. 98, 123704 (2011). 5 B. M. OBrien, T. G. McKay, T. A. Gisby, and I. A. Anderson, Appl. Phys. Lett. 100, 074108 (2012). 6 T. G. McKay, B. M. OBrien, E. P. Calius, and I. A. Anderson, Appl. Phys. Lett. 98, 142903 (2011). 7 R. Kaltseis, C. Keplinger, R. Baumgartner, M. Kaltenbrunner, T. Li, P. Maechler, R. Schwoediauer, Z. Suo, and S. Bauer, Appl. Phys. Lett. 99, 162904 (2011). 8 F. Carpi, D. DeRossi, R. Kornbluh, R. Perline, and P. Sommer-Larsen, eds., Dielectric Elastomers as Electromechanical Transducers (Elsevier, Amsterdam, 2008). 9 Q. Pei, R. Pelrine, M. Rosenthal, S. Stanford, H. Prahlad, and R. Kornbluh, Proc. SPIE 5385, 4150 (2004). 10 H. Stoyanov, M. Kollosche, S. Risse, D. N. McCarthy, and G. Kofod, Soft Matter 7, 194 (2011). 11 H. C. Nguyen, V. T. Doan, J. Park, J. C. Koo, Y. Lee, J. Nam, and H. R. Choi, Smart Mater. Struct. 18, 015006 (2009). 12 C. Huang, Q. Zhang, G. deBotton, and K. Bhattacharya, Appl. Phys. Lett. 84, 4391 (2004). 13 M. Molberg, D. Crespy, P. Rupper, F. Nueesch, J.-A. E. Manson, C. Loewe, and D. M. Opris, Adv. Funct. Mater. 20, 3280 (2010). 14 G. Kofod, S. Risse, H. Stoyanov, D. N. McCarthy, S. Sokolov, and R. Kraehnert, ACS Nano 5, 1623 (2011). 15 D. M. Opris, M. Molberg, C. Walder, Y. S. Ko, B. Fischer, and F. A. Nueesch, Adv. Funct. Mater. 21, 3531 (2011). 16 C. Huang and Q. Zhang, Adv. Funct. Mater. 14, 501 (2004). 17 F. Carpi, G. Gallone, F. Galantini, and D. DeRossi, Adv. Funct. Mater. 18, 235 (2008). 18 H. Stoyanov, M. Kollosche, D. N. McCarthy, and G. Kofod, J. Mater. Chem. 20, 7558 (2010). 19 B. Kussmaul, S. Risse, G. Kofod, R. Wache, M. Wegener, D. N. McCarthy, H. Krueger, and R. Gerhard, Adv. Funct. Mater. 21, 4589 (2011). 20 X. Zhao and Z. Suo, Appl. Phys. Lett. 91, 061921 (2007). 21 B.-X. Xu, R. Mueller, M. Klassen, and D. Gross, Appl. Phys. Lett. 97, 162908 (2010). 22 G. Kofod, P. Sommer-Larsen, R. Kronbluh, and R. Pelrine, J. Intell. Mater. Syst. Struct. 14, 787 (2003). 23 H. Stoyanov, D. Mc Carthy, M. Kollosche, and G. Kofod, Appl. Phys. Lett. 94, 232905 (2009).
2 1

FIG. 3. Linear actuation strain (top) and calculated work density (bottom).

reduced. At higher loadings, the composites are still capable of respectably high linear actuation strains, reaching 19.5% at 80 V/lm for 15%wt and 16.8% at 81 V/lm for 20%wt. The mechanical reinforcement of the materials due to the presence of the nanoparticles leads to signicant increase in the load that the actuators are capable of driving (inset of Fig. 3) which, in combination with the enhanced permittivity, leads to signicant improvements in the work density. The work density of the material was calculated as the work carried out by the actuator divided by the volume and since the actuator drives a constant load the work carried out can be written as: (2) W Fdx Fy0 sz ; where F is the applied load, y0 is the initial actuator length, and sz is the strain at given electric eld. The data plotted in Figure 3 (bottom) show clearly the benecial effect of the nanoparticles on the actuators work density. Although the maximum actuation strain is slightly reduced for the 20%wt nanocomposite compare to the pure PDMS, the driving load is increased by 3 times thus leading to signicant improve-

Downloaded 05 Mar 2013 to 148.206.47.72. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions