High-Speed Electrically Actuated Elastomers with Strain Greater Than 100%

See allHide authors and affiliations

Science  04 Feb 2000:
Vol. 287, Issue 5454, pp. 836-839
DOI: 10.1126/science.287.5454.836


Electrical actuators were made from films of dielectric elastomers (such as silicones) coated on both sides with compliant electrode material. When voltage was applied, the resulting electrostatic forces compressed the film in thickness and expanded it in area, producing strains up to 30 to 40%. It is now shown that prestraining the film further improves the performance of these devices. Actuated strains up to 117% were demonstrated with silicone elastomers, and up to 215% with acrylic elastomers using biaxially and uniaxially prestrained films. The strain, pressure, and response time of silicone exceeded those of natural muscle; specific energy densities greatly exceeded those of other field-actuated materials. Because the actuation mechanism is faster than in other high-strain electroactive polymers, this technology may be suitable for diverse applications.

New high-performance actuator materials capable of converting electrical energy to mechanical energy are needed for a wide range of demanding applications, such as mini- and microrobots, micro air vehicles, disk drives, flat-panel loudspeakers, and prosthetic devices. Many types of candidate materials are under investigation, including single-crystal piezoelectric ceramics (1) and carbon nanotubes (2). Electroactive polymers are of particular interest because of the low cost of materials and the ability of polymers to be tailored to particular applications. Within the general category of electroactive polymers, many different types are under investigation, including electrostrictive polymers (3, 4), piezoelectric polymers (5), and electrochemically actuated conducting polymers and gels (6–11). Most electroactive polymers excel in some measures of performance (such as energy density or strain) but are unsatisfactory in others (such as efficiency and speed of response).

It has been well known for many years that the electric field pressure from free charges on the surface of all insulating materials induces stresses (Maxwell's stress) that strain the material. Zhenyiet al. (4) showed that a largely noncrystalline polymer (polyurethane) could produce actuated strains of 3 to 4% using metal electrodes such as 20-nm-thick gold; they estimated that 10% of their observed strain response was due to Maxwell stress. More recently, it has been suggested that Maxwell stress by itself can produce powerful electroactive responses in certain elastomers (12, 13). This mechanism of actuation (Fig. 1) distinguishes dielectric elastomers from most electrostrictive polymers previously reported. A dielectric elastomer film, typically 10 to 200 μm thick, is coated on each side with a compliant electrode material (e.g., carbon-impregnated grease). When a voltage is applied across the two electrodes, the electrostatic forces compress and stretch the film. Compression of the film thickness brings opposite charges closer together, whereas planar stretching of the film spreads out or separates similar charges. Both changes convert electrical energy to mechanical energy and provide the actuation mechanism.

Figure 1

The dielectric elastomers actuate by means of electrostatic forces applied via compliant electrodes on the elastomer film.

The actuation mechanism illustrated in Fig. 1 was previously shown to have high actuation pressures (0.1 to 2 MPa), fast response times (<1 ms), and potentially high efficiencies (>80 to 90%) (12). A variety of elastomer materials have been investigated, including silicones such as NuSil Technology's CF19-2186 and Dow Corning's HS3. Peak strains of 32% (CF19-2186) and 41% (HS3) were demonstrated, with specific elastic energy densities up to 0.15 J/g for CF19-2186 (12).

We have now conducted experiments that demonstrate extraordinarily high strains, five to six times those previously reported, with higher pressures (up to 7 MPa) and energy densities about 23 times those described earlier. The improvement is due to the identification of a new dielectric actuator material (3M's VHB 4910 acrylic) as well as the application of high prestrain in one planar direction, which enhances electrical breakdown strength and causes the material to actuate primarily in the low-prestrain planar direction. We also present data on applying higher prestrains to improve the performance of previously described silicones. Higher strains and actuation pressures can potentially be exploited to improve a wide range of existing actuator devices (e.g., pumps, motors, robot actuators, generators, and flat-panel loudspeakers) as well as enable new applications (e.g., small flapping-wing vehicles, lifelike prosthetics, noise suppression devices, and biologically inspired robots).

The compression and stretching modes of actuation are mechanically coupled for most elastomers because, at the stresses of interest, the elastomer volume is essentially fixed (the bulk modulus is much higher than the modulus of elasticity Y). We can use our electrostatic model to show that the effective compressive stress,p, compressing the film in thickness (13) isEmbedded Image(1)where ɛ is the relative dielectric constant of the material, ɛo is the permittivity of free space (8.85 × 10−12 F/m), and E is the electric field (volts per meter). The effective compressive stress in Eq. 1 is twice the stress normally calculated for two rigid, charged capacitor plates, because in an elastomer the planar stretching is coupled to the thickness compression. We refer to the stress in Eq. 1 as an effective stress because, strictly speaking, it is the result of both compressive stress acting in the thickness direction and tensile stresses acting in the planar directions. The compressive and tensile stresses are mechanically equivalent in a thin film to a single compressive stress acting in the thickness direction according to Eq. 1.

For low strains (e.g., <20%), the thickness strainsz can be approximated byEmbedded Image(2)For strains greater than about 20%, Eq. 2 is unsatisfactory because Y generally depends on the strain itself. The high actuated strains we observed require the modification of other conventional actuator material constitutive relations as well, even with an assumption of constant modulus. For example, the elastic strain energy density in an actuator material,u e (a common parameter for comparing the output capabilities of actuator materials), is typically expressed asu e = ½psz = ½Ysz 2, but this formula assumes low strains. For high strains, the planar area over which the compression acts increases substantially as the material is compressed (12). For high-strain, nonlinear materials, where the compressive stress is known, a more useful measure of performance might be the electromechanical energy density e, which we define as the amount of electrical energy converted to mechanical energy per unit volume of material for one cycle. The electromechanical energy density can be written asEmbedded Image(3)(14), where p is the constant compressive stress. If we substitute p =Ysz on the right side of Eq. 3 and expand the logarithm for small sz , it follows that ½e = u e, thus providing a valid comparison between the high-strain, nonlinear materials discussed here and conventional low-strain energy density formulas for materials such as piezoelectrics.

We tested many types of polymeric elastomer films. Here we focus on three promising types: Dow Corning HS3 silicone, NuSil CF19-2186 silicone, and the 3M VHB 4910 acrylic adhesive system (15). As noted above, results for the silicone films were reported in earlier publications, but new high-prestrain results using these polymers are reported here.

Strain measurements were made with elastomer films stretched on a rigid frame. Compliant electrodes were stenciled with conductive carbon grease (Chemtronics Circuit Works CW7200) on the top and bottom of the films. The active, electroded portion of the stretched film was small relative to the film's total area. Thus, the inactive portions of the film acted as a spring force on the boundaries of the active regions. When a voltage difference was applied between the top and bottom electrodes, the active region expanded while the inactive region contracted. Removing the applied voltages caused the reverse change. A digital video optical system was used to measure the actuated strain. Measurements were taken about 1 s after application of the voltage. The stretched film technique for measuring strains introduces some boundary constraints from the inactive portion of the film, but it circumvents the difficulty of trying to achieve free-boundary conditions with a soft flexible material (12).

For an elastomer, the absolute strain under actuation depends on the prestrain. A more useful quantity is the relative strain under actuation:Embedded ImageThe relative strain equals the absolute strain if there is zero prestrain in the film. The relative area strain is defined similarly, with the active planar area replacing length in the above expression.

Two types of strain tests were performed, circular (biaxial) and linear (uniaxial). In the circular tests, a small circular active region (5 mm in diameter) was used to decrease the likelihood of a fabrication defect causing an abnormally low breakdown voltage. The film was stretched uniformly on the frame, and the circle expanded in area when a voltage was applied (Fig. 2). The expansion of the circle is equal in both x and y planar directions because there is no preferred planar direction for the film. By contrast, the linear strain tests used a high prestrain in one planar direction and little or no prestrain in the other planar direction. High prestrain effectively stiffens the film in the high-prestrain planar direction, which causes the film to actuate primarily in the softer, low-prestrain planar direction and in thickness. Figure 3 shows a linear strain test. The relative strain was measured in the central region of the elongated (black) active area, away from the edge constraints.

Figure 2

The circular strain test measures the expansion of an actuated circle on a larger stretched film. The photo shows 68% area expansion during actuation of a silicone film.

Figure 3

(A and B) Linear strain test of HS3 silicone film with a high horizontal prestrain for the field off (A) and on (B) with a field of 128 V/μm; 117% relative strain was observed in the central region of (B). (C andD) Activation of acrylic elastomers, producing about 160% relative strain, for the field off (C) and on (D); the dark area in (C) indicates the active region.

The circular test results for three elastomers under different conditions of prestrain are given in Table 1. The peak relative area strain was measured directly, and the relative thickness strain was calculated from the constant volume constraint. The breakdown field was calculated from the known voltage and the measured film thickness (corrected for the given relative thickness strain). No attempt was made to minimize voltage with these relatively thick films, and voltages were typically 4 to 6 kV. Thinner films generally yield lower but comparable performance at lower voltage. For example, preliminary measurements showed 104% relative area strain at 980 V using a thinner acrylic film. The electromechanical energy density e was estimated from the peak field strength (Eq. 1) and the relative thickness strain. The value ½e is listed in Table 1 for convenient comparison to conventional elastic energy densities available for other actuator materials.

Table 1

Circular and linear strain test results.

View this table:

As indicated by the values, the VHB 4910 acrylic elastomer gave the highest performance in terms of strain and actuation pressure. Extensive lifetime tests have not been made, but acrylic films have been operated continuously for several hours at the 100% relative area strain level with no apparent degradation in relative strain performance. However, the acrylic elastomer has relatively high viscoelastic losses that limit its half-strain bandwidth (the frequency at which the strain is one-half of the 1-Hz response) to about 30 to 40 Hz in the circular strain test. By comparison, HS3 silicone has been used for prototype loudspeakers at frequencies as high as 2 to 20 kHz (16, 17). The actuation of CF19-2186 silicone, albeit at lower strains and fields than reported here, has been measured directly via laser reflections with full strain response up to 170 Hz (resonance effects prevented measurement at higher speeds) (12). The only apparent fundamental limits on actuation speed are the viscoelastic losses, the speed of sound in the material, and the time to charge the capacitance of the film (electrical response time).

The strains in the linear strain test can be quite large, up to 215% for the VHB 4910 acrylic adhesive (Table 1). The VHB 4910 acrylic elastomer, when undergoing ∼160% strain in a linear strain test, exhibited buckling (the vertical wrinkles in Fig. 3D) that was not seen in properly stretched silicone films. Buckling indicates that the film is no longer in tension in the horizontal direction during actuation, and that the overall relative thickness strain is greater than indicated by measurements of the electrode boundaries. That is, the relative strain numbers for VHB 4910 in Table 1 may be undervalued.

The dielectric elastomer films presented here appear promising as actuator materials because their overall performance can be good. The available literature indicates that the actuated strains of silicone are greater than for any known high-speed electrically actuated material (that is, a bandwidth above 100 Hz). Silicone elastomers also have other desirable material properties such as good actuation pressures and high theoretical efficiencies (80 to 90%) because of the elastomers' low viscoelastic losses and low electrical leakage (12).

The VHB 4910 acrylic adhesive appears to be a highly energetic material. The energy density of the acrylic adhesive is three times that reported for single-crystal lead–zinc niobate/lead titanate (PZN-PT) piezoelectric (about 1 MJ/m3) (1), itself an energetic new material with performance much greater than that of conventional piezoelectrics. The density of both the silicones and the acrylic adhesive is approximately that of water and about one-seventh that of ceramic piezoelectric materials. Hence, the energy density of the acrylic adhesive on a per-weight basis (the specific energy density) is about 21 times that of single-crystal piezoelectrics and more than two orders of magnitude greater than that of most commercial actuator materials.

Potential applications for dielectric elastomer actuators include robotics, artificial muscle, loudspeakers, solid-state linear actuators, and any application for which high-performance actuation is needed. A variety of actuator devices have been made with the silicone elastomers, including rolled actuators, tube actuators, unimorphs, bimorphs, and diaphragm actuators (12, 18, 19). Their performance is promising, but most of this work did not exploit the benefits of high prestrain or the new acrylic material. We have built an actuator using 2.6 g of stretched acrylic film that demonstrated a force of 29 N and displacement of 0.035 m, a high mechanical output for such a small film mass. The very high strains recently achieved suggest novel applications for shape-changing devices, and the specific energy density of the acrylic adhesive is so high that, if it could be realized in a practical device, it could replace hydraulic systems at a fraction of their weight and complexity. However, practical applications require that a number of other issues be addressed, such as high-voltage, high-efficiency driver circuits, fault-tolerant electrodes, long-term reliability, environmental tolerances, and optimal actuator designs.

  • * To whom correspondence should be addressed. E-mail: pelrine{at}


View Abstract

Navigate This Article