- Email: [email protected]
polyisoprene elastomer blends
REACTIVE & FUNCTIONAL POLYMERS
Reactive & Functional Polymers 66 (2006) 1575–1588
www.elsevier.com/locate/react
Electric field induced stress moduli in polythiophene/polyisoprene elastomer blends Toemphong Puvanatvattana a, Datchanee Chotpattananont a, Piyanoot Hiamtup a, Sumonman Niamlang a, Anuvat Sirivat a,*, Alexander M. Jamieson b a
Conductive and Electroactive Polymer Research Unit, The Petroleum and Petrochemical College, Chulalongkorn University, Bangkok 10330, Thailand b Department of Macromolecular Science and Engineering, Case Western Reserve University, Cleveland, OH 44106, USA Received 18 November 2005; received in revised form 2 May 2006; accepted 7 June 2006 Available online 26 July 2006
Abstract Electrorheological properties of polyisoprene and polythiophene/polyisoprene blends were investigated for electroactive actuator applications. Experiments were carried out under the oscillatory shear mode and with an applied electric field strength varying from 0 to 2 kV/mm. The dynamic moduli, G 0 and G00 , of the pure polyisoprene were measured in terms of the crosslinking ratio and electric field strength. In the absence of an electric field, the storage modulus (G00 ) increased but the loss loss modulus (G00 ) decreased with increasing crosslinking ratio. In the uncrosslinked polyisoprene fluid, the storage modulus (G 0 ) exhibited no change in value with increasing electric field strength. For polyisoprene with crosslinking ratios of 2, 3, 5, and 7, the storage modulus sensitivity increased with electric field strength and attained maximum values of 10%, 60%, 25%, and 30%, respectively, at an electric field strength of 2 kV/mm. The dynamic moduli, G 0 and G00 , of blends of polythiophene with undoped particle concentrations of 5%, 10%, 20%, and 30% vol% were generally higher than those of crosslinked polyisoprene. The storage modulus sensitivity of these blends increased with electric field strength and attained maximum values of 50%, 35%, 110% ,and 45%, respectively, at an electric field strength of 2 kV/min. 2006 Elsevier B.V. All rights reserved. Keywords: Electrorheological properties; Polyisoprene; Polythiophene; Dynamic moduli
1. Introduction The exchange of electrical energy to mechanical energy has been of scientific and technological interests for many decades. Electromechanical energy * Corresponding author. Tel.: +66 2 218 4131; fax: +66 2 611 7221. E-mail address: [email protected] (A. Sirivat).
conversion is required in many applications such as muscle/insect-like actuators and robotic characteristics [1]. Electroactive polymers (EAPs) offer promising and novel characters such as lightweight, high energy density and high flexibility, and they are material candidates for muscle-like actuators. Dielectric elastomers belong to a type of electricfield-activated electroactive polymers that are capable of producing large strains, fast response, and
1381-5148/$ - see front matter 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.reactfunctpolym.2006.06.001
1576
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
high efficiency [2]. Polyisoprene or natural rubber is one type of dielectric material which has many useful characteristics: inexpensive due to its natural source, flexible polymer, low swelling in water, high tensile strength, good resilience, high hot tensile, and well behaved hysteresis. These characteristics are desirable properties required to induce large actuation strain when subjecting to an electric field. Recently, the insertion of a conductive polymer into a dielectric elastomer forming a blend has been of keen interest. Conductive polymers can offer a variety of benefits to the host elastomer: variable conductivity, better thermal stability, and mechanical properties [3]. Examples are a polyanilene-polyisoprene blend for a biosensor application [4], a polyanilene-EPDM blend [5], and TiO2 embedded in PDMS gels for actuator applications [6]. In our work, we are interested in developing and testing polythiophene/polyisoprene elastomer blends as a substitute for artificial muscles. The mechanical properties, viscoelastic properties and electrical properties will be investigated in terms of degree of crosslinking ratio, polythiophene particle concentration, and electric field strength. 2. The electrorheological effect For a dielectric elastomer, a simplified description of the ER effect in solid-like matrices is introduced and summarized here. The nature of the ER effect in polymer gels or elastomers can be explained using the one point dipole model. Polarization in the point dipole model occurs not at the surface of the particle but within it. If dipoles form within particles, an interaction between dipoles occurs and becomes stronger as the dipoles come close to each other. When the particles make contact with each other along the applied electric field, the interaction reaches a maximum. A balance between the electrostatic interaction and the elastic modulus of the solid matrix is important for the ER effect to transpire. If the elastic modulus of the solid matrix is larger than the sum of the interaction between particles, the ER effect may not be observed macroscopically. Therefore, the matrix should be a soft material such as gels or elastomers to produce a significant ER response [7]. We may discuss theoretically the influence of microscopic interaction between polarized particles on the macroscopic mechanical properties of the polymer blends, in particular, the elastic storage modulus. Let us consider a hard sphere in a contin-
uous matrix under an electric field. When the relative dielectric constant of the particle (ep) is larger than that of the matrix (em), a point dipole in the particle can be generated by application of an electric field. According to the classical theory [7], the point dipole moment is given by: l ¼ 4pr3 e0 em jE where j ¼ ðep em Þ=ðep þ 2em Þ
ð1Þ ð2Þ
Eq. (1) is an expression obtained formally for the volume polarizability of a particle, where r is the radius of the particle, e0 is the permittivity in vacuum (=8.854 · 1012 F/m), and E is the applied electric field strength. When two particles are aligned along the applied electric field and are in contact with each other, the dipole–dipole interaction (F) between two dipole particles is given by the following equation [7]: F ¼ ð3=2Þpr2 e0 em j2 E2
ð3Þ
In a cubical gel, there are many straight migration paths of dispersed particles, which remain parallel to the direction of the applied field. The cube gel is deformed by a small shear strain in a direction perpendicular to the applied field. We may assume that electrostatic interactions between the dispersed particles in a path are based only upon the interaction between two adjacent particles and the interaction between paths of particles is negligible. Macroscopic mechanical properties such as storage and loss moduli can be estimated by multiplying the electrostatic force between adjacent particles at a short range in the path, therefore, an increase in elastic modulus due to an applied electric field DG can be calculated as follows [7]: DG ¼ ð9=4ÞCem j2 E2
ð4Þ
where DG is the change in the storage modulus, C is the volume fraction of particles, em is the relative dielectric constant of matrix, and E is the intensity of the applied field. This states that DG is proportional to C, em, and E2. When these factors reach maximum values, DG is expected to become saturated. 3. Experimental 3.1. Materials and synthesis of poly(3thiopheneacetic acid) (Pth) 3-Thiopheneacetic acid, 3TAA (AR grade, Fluka), was used as the monomer. Anhydrous ferric chloride, FeCl3 (AR grade, Riedel-delHean), was
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
used as the oxidant. Chloroform, CHCl3 (AR grade, Lab-Scan), and methanol, CH3OH (AR grade, LabScan), dried over CaH2 for 24 h under nitrogen atmosphere and then distilled, were used as solvents. Diethyl ether was used as a solvent. Polyisoprene, PI (Mw = 40,000, viscosity of 400 Poise, Aldrich), was used as the polymer matrix. Dicumyl peroxide (AR grade, Fluka) was used as the crosslinking agent. The reaction was by the oxidative-coupling polymerization according to the method of Kim et al. [8]. 3-Thiopheneacetic acid (10.0 g) was refluxed for 24 h in 50 ml of dry methanol with one drop of concentrated H2SO4 in order to protect the oxidative decomposition of the carboxylic acid group of monomer during the oxidative-coupling polymerization. Methanol was evaporated, and the residue was extracted with fresh diethyl ether. The extract was washed with deionized water, dried with anhydrous MgSO4 and then filtered. Diethyl ether was evaporated from the filtrate by a rotating evaporator. Thiophene methylacetate (TMA) product was obtained after the diethyl ether was evaporated from the filtrate by a rotating evaporator. In a 100 ml three-necked flask, a solution of 10 mmol of protected monomer in 20 ml of chloroform was added dropwise to a solution of 40 mmol of ferric chloride, which had been dissolved in 30 ml of dry chloroform under nitrogen atmosphere. The molar ratio of the oxidant to monomer was 4:1 in all cases. The reaction was carefully maintained at 0 C (±0.5 C) for 24 h. The reaction mixture was precipitated, by pouring into a large excess amount of methanol (1 L) to obtain polythiophene methylacetate (PTMA). The product was repeatedly washed with methanol and deionized water. PTMA was hydrolyzed as follows: 0.5 g of PTMA was dissolved in 50 ml of 2.0 M NaOH solution and heated for 24 h at 100 C. The mixture was filtered, neutralized, and precipitated with a dilute HCl solution (0.5 M) to yield the polymer. The poly(3-thiopheneacetic acid) (Pth) was washed several times with deionized water before vacuum drying at room temperature for 2 days. 3.2. Characterization of poly(3-thiopheneacetic acid) (Pth) Each poly(3-thiopheneacetic acid) sample was identified for its functional groups by a FT-IR spectrometer (Thermo Nicolet, Nexus 670) operated in
1577
the absorption mode with 32 scans and a resolution of ±4 cm1, covering a wavenumber range of 4000– 400 cm1, using a deuterated triglycine sulfate detector. Optical grade KBr (Carlo Erba Reagent) was used as the background material. The synthesized Pth particles were ground with a mortar, mixed with KBr at a ratio of Pth:KBr = 1:20, and compression molded into pellets under the pressure of 8000 kg. UV–Vis spectra were recorded with an UV–Vis absorption spectrometer (Perkin–Elmer, Lambda 10). Measurements were taken in the absorbance mode in the wavelength range of 200– 800 nm. Synthesized Pth was ground into a fine powder, dissolved in DMSO at the concentration of 6.0 · 105 M, and pipetted into the sample holder. Scan speed was 240 mm/min, and a slit width of 2.0 nm was used with a deuterium lamp as the light source. The thermal stability of Pth was investigated using a thermogravimetric analyzer (DuPont, model TGA 2950) at a temperature range between 30 and 800 C, with a heating rate of 10 C/ min, and under O2 atmosphere [9]. Scanning electron micrographs were taken with a scanning electron microscope (JEOL, model JSM-5200) to determine the morphology of poly(3-thiopheneacetic acid) in powder forms and polythiophene/polyisoprene blends of various particle concentrations. The scanning electron micrographs of polythiophene and polymer blends were obtained by using an acceleration voltage of 20 kV with magnifications of 350 and 1500 times. The particle sizes of Pth were measured by a particle size analyzer (Malvern Instruments Ltd., Masterizer X). The electrical conductivity of undoped Pth was measured by a custom-built two-point probe coupled with an electrometer (Keithley, model 6517A). The specific conductivity r (S/cm) values of the pellets were obtained by measuring the bulk pellet resistance R (X). The relation r = (1/Rt)(1/K) = (I/Vt)(1/K) was used to calculate specific conductivity, where t is the pellet thickness (cm), I is the current change (A), V is the applied voltage (voltage drop) (V) and K is the geometric correction factor which is equal to the ratio w/l, where w and l are the probe width and the probe length, respectively. The geometrical correction factor (K) was determined by calibrating the two-point probe with semi-conducting silicon sheets of known resistivity values. These two probes were connected to a voltmeter (Keithley, model 6517A), which supplied a constant voltage source, and a resultant current was measured. Electrical conductivity values of several samples were
1578
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
first measured at various applied voltages to identify their linear Ohmic regimes. The crosslink density of PI was determined by measuring Mc at the so-called equilibrium swelling. Mc was determined from the swelling ability of certain solvents in a crosslinked network. The procedure of the crosslink density measurement was as follows. Initially the samples were cut to 10 · 10 mm, and then the weight of each specimen was measured. We put each specimen in a vial containing toluene for 3 days and the weight of the swollen sample was remeasured. For each crosslinking ratio, three samples were used and three experiments were carried out, and the data were averaged. The crosslink density was calculated from the following equations [10–13]:
Mc ¼
1=3 v1 q2 v2m v2m 2
½lnð1 v2m Þ þ v2m þ v1 v22m w0 v2m ¼ vequil q2 w0 ws w0 þ vequil ¼ q2 q1 v1 v ¼ 0:34 þ ðd1 d2 Þ2 RT 1 where crosslink density ¼ ½2M c ;
ð5Þ ð6Þ ð7Þ ð8Þ ð9Þ
v1 is the molar volume of solvent (Mw/density), q2 is the polyisoprene density equal to 0.92 g/cm3, q1 is the toluene density equal to 0.867 g/cm3, w0 is the original polymer weight, ws is the swollen polymer weight, v is the polymer-solvent interaction parameter, R is the universal gas constant, 8.29 Nm/mol K, T is the absolute temperature, 298 K, d1 is the solubility parameter of polyisoprene equal to 17.02 (MPa)1/ 2 , and d2 is the solubility parameter of toluene equal to 18.20 (MPa)1/2. 3.3. Electrorheological properties measurements and sample preparations The electrorheological properties of pure polyisoprene were investigated in terms of electric field strength and crosslinking ratio. A specific amount of dicumyl peroxide was mixed with polyisoprene and cast onto a mold (diameter 25 mm). Various amounts of dicumyl peroxide were used in order to vary the degree of crosslinking (crosslink ratios NDCP/NPI of 2, 3, 5, and 7). The existing bubbles were removed in a vacuum atmosphere at 25 C for 2 h. The fabrication of the crosslinked polyiso-
prenes was by compression molding at 175 C and 5 MPa for 7 min [14]. The polythiophene (Pth) powder was sieved with a mesh particle size of 38 lm and dried at room temperature for 24 h prior to their uses. The Pth/PI blends were prepared by mechanical blending of undoped synthesized polythiophene particles at various particle concentrations (5, 10, 20, and 30 vol%). Polyisoprene fluid with a specific amount of crosslinking agent was mixed together, and the mixture was mechanically stirred for about 5 min at room temperature. The optimum degree of crosslink was chosen (NDCP/NPI = 3) since this material had the highest G 0 sensitivity. A specific amount of Pth particles was then added and the mixture was mechanically stirred for 5 min at room temperature. The mixture was cast on the mold (diameter 25 mm) and bubbles were removed under a vacuum atmosphere at 25 C for 2 h. The fabrication of the Pth/crosslinked polyisoprenes blends (Pth_U/ PI_03) was carried out by compression molding at 175 C and 5 MPa for 7 min [14]. A melt rheometer (Rheometric Scientific, model ARES) was used to measure rheological properties. It was fitted with a custom-built copper parallel plates fixture (diameter of 25 mm). DC voltage was applied by a DC power supply (Instek, model GFG 8216A), which can deliver electric field strength up to 2 kV/mm. A digital multimeter was used to monitor voltage input. In these experiments, the oscillatory shear strain was applied and the dynamic moduli (G 0 and G00 ) were measured as functions of frequency and electric field strength. Strain sweep tests were first carried out to determine the suitable strains to measure G 0 and G00 in the linear viscoelastic regime. The appropriate strain was determined to be 220% for pure polyisoprene fluid (PI_00). For the crosslinked polyisoprenes (PI_02, PI_03, PI_05 and PI_07) and polythiophene/crosslinked polyisoprene blends (at blend compositions of 5, 10, 20, and 30 vol% corresponding to Pth_U5/PI_03, Pth_U10/PI_03, Pth_U20/PI_03, and Pth_U30/PI_03), the strain of 15% was used. The frequency sweep tests were carried out to measure G 0 and G00 of each sample as functions of frequency. The deformation frequency was varied from 0.1 to 100 rad/s. Prior to each measurement, each polyisoprene or polythiophene/polyisoprene blends sample was presheared at a low frequency (0.04 rad/s) with the electric field on for 10 min in order to attain the equilibrium polarization. Each measurement was carried
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
out at temperature of 27 C and was repeated at least two or three times. 4. Results and discussion 4.1. Characterization of poly(3-thiopheneacetic acid) An FT-IR spectrum of the synthesized Pth was taken to identify characteristic absorption peaks [8]. The characteristic peaks of Pth were at 3200– 3000 cm1, 3000–2800 cm1, 1700 cm1, 1300– 1200 cm1, and 830 cm1. These peaks can be assigned to the C–H bond stretching on the thiophene ring; the aliphatic C–H bond stretching; the carboxylic acid C@O stretching; the thiophene ring stretching; the carboxylic acid C–O stretching, and the out-of-plane thiophene C–H stretching, respectively [8]. The most distinct feature of this spectrum is the extremely broad O–H absorption occurring in the region from 3400 cm1 to 2400 cm1; this peak can be attributed to the strong hydrogen bonding of the dimmers. This absorption band often obscures the C–H stretching vibration peak occurring in the same region. It is obvious from the absorption peak at around 1700 cm1 that the ester groups were not deteriorated during the oxidative polymerization. A UV–Vis absorption spectrum of synthesized Pth/DMSO solution shows absorption peaks at 275 nm and 415 nm, corresponding to the p–p* transition of the bithiophene unit and the polymer backbone, respectively [8,15,16]. A TGA thermogram of synthesized Pth shows two degradation steps at 200 C and 440 C, corresponding to the side chain degradation and the backbone degradation, respectively [9]. The polythiophene/polyisoprene blends (Pth_U/PI_03) have a better thermal stability with increasing polythiophene particle concentration. The mean particle diameter of Pth was determined to be approximately 20 lm with standard deviation of 3 lm. The particle microstructure was observed using scanning electron microscopy (SEM). Fig. 1 shows the shapes of the undoped Pth and undoped Pth particles in polymer blends; they are quite irregular in shape. However, these irregular shaped particles appear to be well dispersed in the polyisoprene matrix [9]. The specific conductivity of undoped Pth (Pth_U) was measured by the custom-built two point probe (Keithley, model 6517A). The specific conductivity of Pth_U was 3.01 · 106 S/cm with
1579
a standard deviation of 1.01 · 107 S/cm. The specific conductivity of polyisoprene was 8.18 · 107 S/cm with a standard deviation of 1.13 · 107 S/cm. 4.2. Electrorheological properties of pure polyisoprene 4.2.1. Effect of crosslink in the absence of electric field The effect of crosslink on the rheological properties of pure polyisoprene (PI) was first investigated. The crosslink ratios, NDCP/NPI, were 2, 3, 5, and 7 (PI_02, PI_03, PI_05, and PI_07). For the PI_02 system, NDCP/NPI and the crosslink density were 2.00 and 4.058 · 105 mol/g, respectively. For the highly crosslinked system of PI_07, NDCP/NPI and the crosslink density were 7.02 and 16.495 · 105 mol/g, respectively. The crosslink density increases with increasing crosslink mole ratio; data are tabulated in Table 1. Fig. 2(a) and (b) shows the storage modulus (G 0 ) and the loss modulus vs. frequency at electric field strengths of 0 and 2 kV/mm, and at strain of 1%. G 0 increases monotonically with increasing crosslink ratio [1]. On the other hand, the loss modulus (G00 ) decreases with increasing crosslink ratio, as shown in Fig. 2(b). Increasing the crosslink ratio or the crosslink density prohibits free movements of polymer chains and consequently less molecular friction. The polyisoprene rheological characteristic changes from a fluid-like behavior (PI_03) to a solid-like behavior (PI_07) along with the corresponding increase in the storage modulus [7]. 4.2.2. Effect of crosslink under electric field The effect of electric field strength on the rheological properties of polyisoprene (PI) at various crosslink ratios was investigated in the electric field range between 0 and 2 kV/mm. Fig. 3 shows the storage modulus responses (DG 0 ) of various crosslinked polyisoprene systems as functions of electric field strength at frequency of 1 rad/s, and strain of 1%. Above a critical electric field of 500 V/mm, the storage modulus responses appear to increase monotonically with electric field strength. The storage modulus responses (DG 0 ) at electric field strength of 2 kV/mm are 500, 4097, 5002, and 8894 Pa for PI_02, PI_03, PI_05, and PI_07, respectively. The corresponding storage modulus sensitivity values, 0 ðxÞ defined as DG , of the crosslinked polyisoprene sysG0 ðxÞ 0
tems (PI_02, PI_03, PI_05, and PI_07) at electric
1580
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
Fig. 1. The morphology of polythiophene particles and polythiophene/ polyisoprene blends: (a) synthesized P3TAA at a magnification of 1500; (b) Pth_U5/PI_03; (c) Pth_U10/PI_03; (d) Pth_U20/PI_03; and (e) Pth_U30/PI_03 at a magnification of 350.
Table 1 The crosslink density of pure polyisoprene Systems
NDCP/NPI
Crosslink density (105 mol/g)
PI_02 PI_03 PI_05 PI_07
2.00 3.02 5.01 7.02
4.058 ± 0.312 6.638 ± 1.000 11.797 ± 0.625 16.495 ± 1.750
forces that act to pull the electrodes together [2,17]. This attraction causes a compressive force on the dielectric elastomeric in the direction of the electric field throughout the volume between the electrodes. The resulting effective (squeeze) pressure can be written as [2]: p ¼ e0 er E2
field strength of 2 kV/mm are 10%, 60%, 25%, and 30%, respectively. As an electric field is applied, electrical dipole moments are generated and the electrostatic interaction between the polymer chains is induced leading to an intermolecular interaction acting like an electrical network. In addition, a voltage differential between the electrodes can generate electromagnetic
ð10Þ
where e0 and er are the permittivity of free space and the relative permittivity of polymer, respectively. E is the applied electric field. The response of our polymer systems is functionally similar to those of electrostrictive polymers, in which the response is directly proportional to the square of the applied electric field [1,17,18]. The intermolecular interaction and the electrostatic interaction result in the loss of free chain movements, the higher chain rigid-
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
1581
G' (Pa)
1e+5
1e+4 PI_03, E = 0 PI_03, E = 2 PI_05, E = 0 PI_05, E = 2 PI_07, E = 0 PI_07, E = 2
kV/mm kV/mm kV/mm kV/mm kV/mm kV/mm
1e+3 .01
.1
1
10
100
1000
10
100
1000
ω (rad/s)
(a) 1e+5 PI_03, E = 0 PI_03, E = 2 PI_05, E = 0 PI_05, E = 2 PI_07, E = 0 PI_07, E = 2
G" (Pa)
1e+4
1e+3
kV/mm kV/mm kV/mm kV/mm kV/mm kV/mm
1e+2
1e+1
1e+0 .01
.1
(b)
1
ω (rad/s)
Fig. 2. Storage and loss moduli of polyisoprene systems at various crosslink ratios vs. frequency, strain 1%, 27 C, and at electric field strengths of 0 and 2 kV/mm: (a) storage modulus, G 0 (x) and (b) loss modulus, G00 (x).
ity, and correspondingly higher G 0 (x). Yang et al. [18] reported a similar finding on the effect of increasing effective (squeeze) pressure; the radial and circumferential stresses changed from a tensile state to a compressive state at a critical effective pressure. Fig. 4 shows the storage modulus responses (DG 0 ) as functions of crosslink density at electric field strengths of 1 kV/mm and 2 kV/mm. The storage modulus responses increase almost linearly with increasing crosslink density at both electric field strengths. The highest crosslink density system (PI_07, crosslink density = 16.495 · 105 mol/g) possesses the highest storage modulus response of
4000 and 8894 Pa at electric field strengths of 1 and 2 kV/mm, respectively. We note that the 0 ðxÞ storage modulus sensitivity value defined as DG G00 ðxÞ of the crosslinked polyisoprene system PI_03 (crosslink density = 6.638 · 105 gmol/ml), has the highest G 0 sensitivity value: 60% at electric field strength of 2 kV/mm. This result suggests that the PI_03 system contains a balanced rheological characteristic between the solid-like and the fluid-like behaviors; it has enough flexibility within the matrix in the absence of electric field in such a way that its structure rigidity can be further induced with electric field. On the other hand, the lower crosslinked system (PI_02)
1582
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
ΔG' (Pa)
10000
8000
PI_00, G'0 = 1.50 Pa PI_02, G'0 = 4812 Pa
6000
PI_03, G'0 = 7401 Pa PI_05, G'0 = 19155 Pa PI_07, G'0 = 27201 Pa
4000
2000
0
-2000 100
1000
E (V/mm) Fig. 3. Storage modulus responses of polyisoprene at various crosslink ratios vs. electric field strength, at frequency 1 rad/s, strain 1%, (PI_00 system, strain 220%), and at 27 C.
12000
E=1 E=1 E=2 E=2
10000
kV/mm; freq = 1.0 kV/mm; freq = 10 kV/mm; freq = 1.0 kV/mm; freq = 10
rad/s rad/s rad/s rad/s
ΔG'
8000
6000
4000
2000
0 0
5
10
15
20
Crosslink density (10 -5 mol/g) Fig. 4. Storage modulus responses vs. crosslink density at 27 C, and at electric field strengths of 1 and 2 kV/mm.
has a more fluid-like structure than other systems; free movement and relaxation of polymer chains are still allowed in the presence of electric field and consequently they are least affected by electric field. The higher crosslinked systems (PI_05 and PI_07) contain relatively more solid-like structures and thus they possess relatively higher storage modulus values. The polymer segments and chains are quite rigid and fixed in the absence of electric field; they simply cannot respond any further to intermolecular interaction under the applied electric field.
4.2.3. Effect of polythiophene composition in the absence of electric field The effect of polythiophene particle concentration on the rheological properties of polymer blends (Pth_U/PI) was investigated next. The PI_03 system, with crosslink density of 6.638 · 105 mol/g, was chosen and blended with poly(3-thiopheneacetic acid) particles. This PI_03 system exhibits the maximum G 0 sensitivity amongst the various systems studied. Figs. 5(a) and (b) show comparisons of the storage modulus (G 0 ) and the loss modulus
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
1583
1e+6
G' (Pa)
Pth_U5/PI_03, E = 0 Pth_U5/PI_03, E = 2 Pth_U30/PI_03, E = 0 Pth_U30/PI_03, E = 2
kV/ mm kV/ mm kV/ mm kV/ mm
1e+5
1e+4
.01
.1
1
1e+6
Pth_U5/PI_03, E = 0 Pth_U5/PI_03, E = 2 Pth_U30/PI_03, E = 0 Pth_U30/PI_03, E = 2
1e+5
G" (Pa)
10
100
1000
10
100
1000
ω (rad/s)
(a)
kV/ mm kV/ mm kV/mm kV/mm
1e+4
1e+3
1e+2
.01
.1
1
(b)
ω (rad/s)
Fig. 5. Comparison of the storage and the loss moduli of polythiophene/polyisoprene blends (Pth_U/PI_03) at various particle concentrations (5 and 30 vol%), strain 1%, 27 C, and at electric field strengths of 0 and 2 kV/mm: (a) the storage modulus, G 0 and (b) the loss modulus, G00 .
(G00 ) vs. frequency between polymer blends of various concentrations, at electric field strengths of 0 and 2 kV/mm. Polythiophene particle concentrations studied were 5, 10, 20, and 30 vol.% (Pth_U5/PI_03, Pth_U10/PI_03, Pth_U20/PI_03, and Pth_U30/PI_03). Both G 0 and G00 increase with polythiophene particle concentrations at all frequencies. The storage modulus G 0 (x) and the loss modulus G00 (x) of each polymer blends system are higher than those of the PI_03 system at any electric field strength. Close inspection of our data suggests that polythiophene particles in the polymer blends behave as fillers in the matrix; they can store or
absorb the forces/stresses within the matrix [1]. A polymer blend system with a higher particle concentration is expected to exhibit a higher internal stress response, a higher storage modulus G 0 (x), and a higher loss modulus G00 (x) response than those of pure PI_03 system. Concentration effect has been reported by Shiga [7] who found that the shear storage modulus of silicone gel containing PMACo particles increased with increasing volume fraction of dispersed particles embedded in the gel. Krause et al. [1] reported that the compression modulus, in the absence of an electric field, increased by a factor of 2 when
1584
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
the PANI particle concentration was increased from 0 to 20 wt%. These results were expected because the particles acted as filler particles capable of supporting internal stresses within the matrices. 4.2.4. Effect of polythiophene composition under electric field The effect of electric field strength on the rheological properties of polythiophene/polyisoprene blends (Pth_U/PI) at various polythiophene particle concentrations was investigated in the range between 0 and 2 kV/mm. Figs. 5(a) and (b) show the comparison of the storage modulus (G 0 ) and the loss modulus (G00 ) between the blends of various concentrations at electric field strengths of 0 and 2 kV/mm. The storage modulus (G 0 ) and the loss modulus (G00 ) of each polymer blend system generally increase with increasing electric field strength. Fig. 6 shows the storage modulus responses (DG 0 ) vs. electric field of various polymer blend systems at frequency of 1 rad/s, and strain equal to 1%. The increase in DG 0 with electric field is nonlinear within the range of 0.1–2 kV/mm. The storage modulus response values, DG 0 (x), of these systems at electric field strength of 2 kV/mm are 4097, 5446, 4026, 19161, and 33857 Pa for PI_03, Pth_U5/ PI_03, Pth_U10/PI_03, Pth_U20/PI_03, and Pth_U30/PI_03, respectively. The corresponding storage modulus sensitivity values, defined as DG0 ðxÞ , of these polymer blends systems (Pth_U5/ 0 G0 ðxÞ PI_03, Pth_U10/PI_03, Pth_U20/PI_03, and
Pth_U30/PI_03) at electric field strength of 2 kV/ mm are 50%, 35%, 110%, and 45%, respectively. In the absence of the electric field, the polythiophene particles are randomly dispersed within the polyisoprene matrix and there is no particle–particle interaction. When the electrical field is applied, both polyisoprene and polythiophene particles become polarized and induced dipole moments are generated, leading to intermolecular interactions. These intermolecular interactions induce the loss of chain free movements and the higher chain rigidity, as indicated by higher G 0 (x) values. The electric field evidently enhances the elastic modulus of our conductive polymer-polyisoprene blends, consistent with the results of other polymer blends [17–19]. Lui and Shaw [19] reported that for the silica/silicone elastomers, their moduli exhibited a nearly quadratic dependence on electric field when the particles were randomly dispersed. For the aligned structure, the shear modulus increased quadratically with low electric field values but became saturated at high values. Shiga [7] reported that for the PMACo particles in silicone gel, the increase in the elastic modulus induced by an electric field was 4 kPa with the particle volume fraction of 0.3, and the applied electric field of 2 kV/mm. Fig. 7 shows the storage modulus responses (DG 0 ) of polythiophene/polyisoprene blends vs. particle volume fraction at electric field strengths of 1 and 2 kV/mm. For both electric field strengths of 1 and 2 kV/mm, the storage modulus increases line-
40000 PI_03,
G'0 = 7401 Pa
Pth_U5/PI_03, G'0 = 11085 Pa Pth_U10/PI_03, G'0 = 12416 Pa
30000
ΔG' (Pa)
Pth_U20/PI_03, G'0 = 19428 Pa Pth_U30/PI_03, G'0 = 78098 Pa
20000
10000
0 100
1000
E (V/mm) 0
Fig. 6. Storage modulus responses (DG ) of the polythiophene/polyisoprene blends (Pth_U/PI_03) at various particle concentrations vs. electric field strength, at frequency 1 rad/s, strain 1%, and at 27 C.
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
1585
50000
Δ G' (Pa)
40000
E=1 E=1 E=2 E=2
kV/ mm; freq = 1.0 kV/ mm; freq = 10 kV/ mm; freq = 1.0 kV/ mm; freq = 10
rad/s r ad/s rad/s r ad/s
30000
20000
10000
0 0.00
.05
.10
.15
.20
.25
.30
φPth Fig. 7. Storage modulus responses (DG 0 ) of the polythiophene/polyisoprene blends as functions of particle volume fraction, 27 C, of various frequencies, and at electric field strengths of 1 and 2 kV/mm.
arly with particle volume fraction below 0.10; and it increases nonlinearly when particle volume fraction was above 0.10. For electric field strength of 1 kV/ mm, the storage modulus response values are 1728, 2592, 1996, 14053, and 8322 Pa for PI_03, Pth_U5/PI_03, Pth_U10/PI_03, Pth_U20/PI_03, and Pth_U30/PI_03, respectively. We also note that the storage modulus sensitivity value, defined 0 ðxÞ as DG , of polythiophene/polyisoprene blends G0 ðxÞ 0
Pth_U20/PI_03 system attains a maximum G 0 sensitivity value: 110% at electric field strength of 2 kV/ mm. At very low concentrations (Pth_U5/PI_03 and Pth_U10/PI_03), the number of particles is too small and the distances between particles are too large to allow a significant particle interaction [19]. The storage modulus response values of these systems are comparable to those of the crosslinked and the pure polyisoprene system (PI_03). At higher concentrations (Pth_U20/PI_03 and Pth_U30/PI_03), the distances between particles are smaller and hence there is a stronger interparticle interaction [7,19]. Evidently, the electric field strength of 1 kV/mm is presumably not strong enough to induce polarization interactions amongst all particles at high particle concentrations due to the steric hindrance effect [19]. The storage modulus response of Pth_U30/PI_03 system becomes saturated at approximately the same value as that of Pth_U20/PI_03 system. The storage modulus response shows a nonlinear
dependence on the particle volume fraction and it attains a saturation state at volume fractions above 0.2. On the other hand, at the higher electric field strength of 2 kV/mm, induced polarization interactions are more effective amongst particles at high particle concentrations. No saturation state was observed at high particle concentrations. Liu and Shaw [19] reported a similar effect for silica/silicone elastomers. They found that the enhancement of shear modulus was negligible below 8.0 vol%, but increased dramatically above this threshold concentration. At volume fraction above 55 vol%, the shear modulus decreased because the interparticle force decreased due to the steric hindrance effect. Fig. 8 shows electrical current vs. electric field of various blend systems studied. It can be seen that the leakage current remains below 2 lA at the electric field strength of 2 kV/mm. The case of the highest particle concentration of 30 vol%, 30Pth_U/ PI_03, in which the leakage current attains a value of about 7 lA at electric field strength of 2 kV/ mm, is an exception. 4.2.5. Electrorheological temporal response Finally, we investigated the temporal characteristics of the crosslinked polyisoprene (PI_03) and the polythiophene/polyisoprene blend at particle concentration of 20 vol% (Pth_U20/PI_03) at various electric field strengths. The temporal characteristic of each sample was recorded in the
1586
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588 8
Ι (μΑ)
6
PI_03 5Pth_U/PI_03 10Pt h_U/PI_03 20Pt h_U/PI_03 30Pt h_U/PI_03
4
2
0 0
500
1000
1500
2000
2500
E (V/mm) Fig. 8. Current (I) vs. applied electric field strength (E) of the polyisoprene system and polythiophene/polyisoprene blend systems at various polythiophene particle concentrations, and at 27 C.
linear viscoelastic regime at a strain of 1% and at frequency of 1 rad/s. Fig. 9(a) shows the change in G 0 of PI_03 system at electric field strengths of 1 and 2 kV/mm during a time sweep test, in which an electric field was turned on and off alternately. At the electric field of 1 kV/mm, G 0 immediately increased and rapidly reached a steady-state value. With the electric field turned off, G 0 decreased and nearly recovered its original value. Thus PI_03 appears to be a reversible system at 1 kV/mm. At the electric field of 2 kV/mm, the response of G 0 can be divided into two regimes: the initial regime in which G 0 rapidly overshooted to a large value on the first cycle followed by an irreversible decay with electric field off, and the steady state regime in which G 0 subsequently exhibited a reversible cyclic response. The time required for G 0 to reach the steady-state value on applying the field is called the induction time, sind. As shown in Table 2, sind decreases with increasing electric field strength; they are 107 and 75 s at electric field strengths equal to 1 and 2 kV/mm, respectively. The time required for G 0 to decay towards its steady-state value when the electric field is turned off is called the recovery time, srec. It is essentially independent of the electric field strength, as shown in Table 2. The recovery times are 76 and 83 s at electric field strengths equal to 1 and 2 kV/mm, respectively. The independence of srec on the field strength suggests that the strains induced were small such that the relaxations were nearly the same. We also found that tind is independent of crosslink density
or the elastic modulus, whereas trec decreases with increasing crosslink density, as shown in Table 2. The independence of sind with the crosslink density suggests that sind depends critically more on the electric field strength. The lower srec value required to relax with a higher crosslink density is because a higher elastic modulus system generally has a shorter relaxation time scale. Fig. 9(b) shows the temporal response of 20 vol% polythiophene/polyisoprene blend (Pth_U20/ PI_03) at electric field strengths of 1 and 2 kV/mm, respectively. The Pth_U20/PI_03 is clearly an irreversible system at both electric field strengths. Our result here suggests that there are some irreversible interactions between polythiophene particles, perhaps due to hydrogen bonding between adjacent polythiophene particles, and residual dipole moments inducing permanent interparticle interactions [9]. sind decreases with increasing field strength, as shown in Table 2. sind are 157 and 121 s at electric field strengths equal to 1 and 2 kV/mm, respectively. srec is nearly independent of electric field strength. As shown in Table 2, srec are 59 and 66 s at electric field strength equal to 1 and 2 kV/mm, respectively. Finally, we may note that the equilibrium G 0 values of PI_03 and Pth_U20/PI_03 systems in Figs. 9(a) and (b) are not the same as those shown in Figs. 2(a) and 6, respectively. This arises because the samples of Figs. 2(a) and 6 were subjected to preshearing under a steady electric field for a duration of 10 min. On the other hand,
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
1587
G' (Pa)
8000
on
off
7000
E = 1 kV/mm E = 2 kV/mm
on off
6000 0
2000
4000
(a)
6000
8000
10000
12000
time (s) 70000
G' (Pa)
65000
60000 on off on off
55000
E = 1 kV/mm E = 2 kV/mm
on off on off
50000 0
2000
4000
6000
8000
10000
12000
14000
time (s)
(b)
Fig. 9. Temporal response of storage modulus (G 0 ) of: (a) PI_03 system at various electric field strengths (1 and 2 kV/mm) and (b) Pth_U20/PI_03 system at various electric field strengths (1 and 2 kV/mm), and at temperature of 27 C.
the fresh samples of Fig. 9(a) and (b) were not presheared but the electric field was periodically turned on and off. Table 2 Induction time and recovery time of pure polyisoprene systems and polythiophene/polyisoprene blends Samples
Electric field (kV/mm)
Induction time (sind) (s)
Recovery time (srec) (s)
DG0ind (Pa)
DG0re (Pa)
PI_03
1 2 2 2 1 2
107 75 73 78 157 121
76 83 74 64 59 66
195 110 202 851 2257 4794
188 106 175 821 35 80
PI_05 PI_07 Pth_U20/ PI_03
5. Conclusion In our work, electrorheological properties of polyisoprene and polythiophene/polyisoprene blends were investigated by examining the effects of crosslink ratio and polythiophene particle concentration on the dynamic moduli, G 0 and G00 , under the oscillatory shear mode at electric field strength varying from 0 to 2 kV/mm. In pure PI systems, the storage modulus (G 0 ) increased but the loss modulus (G00 ) decreased with increasing crosslinking ratio. It exhibited a transition from a fluid-like to a solid-like behavior as crosslink density was varied. The storage modulus (G 0 ) and the loss modulus
1588
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
(G00 ) of the uncrosslinked polyisoprene fluid exhibited no change with increasing electric field strength. For polyisoprenes with the crosslink ratios of 2, 3, 5, and 7, the storage modulus increased with electric field strength. The maximum G 0 sensitivity was found to be about 0.6 for the PI_03 system at electric field strength of 2 kV/mm; it is frequency independent. Poly(3-thiopheneacetic acid) particles were synthesized via an oxidative polymerization and blended with the PI_03 system. For the electrorheological properties of Pth_U/PI_03 blends with the polythiophene particle concentrations of 5, 10, 20, and 30 vol%, the dynamic moduli G 0 and G00 of each polymer blend, were generally higher than those of pure polyisoprene due to polythiophene particles within the matrix acting as fillers; they can store or absorb the forces/stresses within the matrix. The storage modulus responses increased with electric field nonlinearly within the range of 0.1–2 kV/ mm. This can be attributed to the fact that polyisoprene and polythiophene particles became polarized and induced dipole moments were generated, leading to intermolecular interactions along the direction of electric field. The storage modulus 0 sensitivity, DG , attained a maximum G 0 sensitivity G00 value of 0.5, 0.35, 1.10, and 0.45 at particle concentrations of 5, 10, 20, and 30 vol% at the electric field strength of 2 kV/mm, respectively. Acknowledgements The authors acknowledge the financial supports to A.S. from Chulalongkorn University (through a grant from the Ratchadapesak Somphot Endowment Fund for the foundation of the
Conductive and Electroactive Polymers Research Unit), and the Petroleum and Petrochemical College Consortium. References [1] S. Krause, K. Bohon, Macromolecules 34 (2001) 7179–7189. [2] R. Kornbluh, R. Perlrine, Q. Pei, S. Oh, J. Joseph, in: Proceedings of the First World Congress on Biomimetics, 2002. [3] S. Ku¨c¸u¨kyavuz, M. Sankir, Z. Ku¨c¸u¨kyavuz, Synthetic Met. 128 (2002) 247–251. [4] Z. Shen, H. Xue, Y. Li, Synthetic Met. 124 (2001) 345–349. [5] R. Faez, R.-H. Schuster, M.-A. De Paoli, Eur. Polym. J. 38 (2002) 2459–2463. [6] M. Zrı´nyi, J. Fehe´r, G. Filipcsei, Macromolecules 33 (2000) 5751–5753. [7] T. ShigaAdvances in Polymer Science, vol. 134, Springer, Verlag, Berlin Heidelberg, 1997, pp. 133–163. [8] B. Kim, L. Chen, J.P. Gong, Y. Osada, Macromolecules 32 (1999) 3964–3969. [9] D. Chotpattananont, A. Sirivat, A.M. Jamieson, Colloid Polym. Sci. 282 (2004) 357–365. [10] F. Horkay, G.B. McKenna, P. Deschamps, E. Geissler, Macromolecules 33 (2000) 5215–5220. [11] S.-S. Choi, J. Anal. Appl. Pyrol. 52 (1999) 105–112. [12] P.C. Painter, M.M. Coleman, Fundamentals of Polymer Science: An Introductory Text, Technomic Publishing Company, USA, 1997. [13] D.W. Wang, 2nd ed., Encyclopedia of Polymer Science and Engineering, vol. 1, A Wiley-Interscience Publication, New York, 1992. [14] R. Faez, M.A. De Paoli, Eur. Polym. J. 37 (2001) 1139–1143. [15] F. Wang, Y.-H. Lai, M.-Y. Han, Macromolecules 37 (2004) 3222–3230. [16] F. Demanze, A. Yassar, F. Garnier, Macromolecules 29 (1996) 4267–4273. [17] R.E. Perline, R.D. Kornbluh, J.P. Joseph, Sensor Actuat. Phys. A64 (1998) 77–85. [18] E.E. Yang, M. Frecker, E. Mockensturm, in: Proceedings of ASME International Mechanical Engineering, 2003. [19] B. Liu, T.M. Shaw, J. Rheol. 45 (3) (2001) 641–657.
Reactive & Functional Polymers 66 (2006) 1575–1588
www.elsevier.com/locate/react
Electric field induced stress moduli in polythiophene/polyisoprene elastomer blends Toemphong Puvanatvattana a, Datchanee Chotpattananont a, Piyanoot Hiamtup a, Sumonman Niamlang a, Anuvat Sirivat a,*, Alexander M. Jamieson b a
Conductive and Electroactive Polymer Research Unit, The Petroleum and Petrochemical College, Chulalongkorn University, Bangkok 10330, Thailand b Department of Macromolecular Science and Engineering, Case Western Reserve University, Cleveland, OH 44106, USA Received 18 November 2005; received in revised form 2 May 2006; accepted 7 June 2006 Available online 26 July 2006
Abstract Electrorheological properties of polyisoprene and polythiophene/polyisoprene blends were investigated for electroactive actuator applications. Experiments were carried out under the oscillatory shear mode and with an applied electric field strength varying from 0 to 2 kV/mm. The dynamic moduli, G 0 and G00 , of the pure polyisoprene were measured in terms of the crosslinking ratio and electric field strength. In the absence of an electric field, the storage modulus (G00 ) increased but the loss loss modulus (G00 ) decreased with increasing crosslinking ratio. In the uncrosslinked polyisoprene fluid, the storage modulus (G 0 ) exhibited no change in value with increasing electric field strength. For polyisoprene with crosslinking ratios of 2, 3, 5, and 7, the storage modulus sensitivity increased with electric field strength and attained maximum values of 10%, 60%, 25%, and 30%, respectively, at an electric field strength of 2 kV/mm. The dynamic moduli, G 0 and G00 , of blends of polythiophene with undoped particle concentrations of 5%, 10%, 20%, and 30% vol% were generally higher than those of crosslinked polyisoprene. The storage modulus sensitivity of these blends increased with electric field strength and attained maximum values of 50%, 35%, 110% ,and 45%, respectively, at an electric field strength of 2 kV/min. 2006 Elsevier B.V. All rights reserved. Keywords: Electrorheological properties; Polyisoprene; Polythiophene; Dynamic moduli
1. Introduction The exchange of electrical energy to mechanical energy has been of scientific and technological interests for many decades. Electromechanical energy * Corresponding author. Tel.: +66 2 218 4131; fax: +66 2 611 7221. E-mail address: [email protected] (A. Sirivat).
conversion is required in many applications such as muscle/insect-like actuators and robotic characteristics [1]. Electroactive polymers (EAPs) offer promising and novel characters such as lightweight, high energy density and high flexibility, and they are material candidates for muscle-like actuators. Dielectric elastomers belong to a type of electricfield-activated electroactive polymers that are capable of producing large strains, fast response, and
1381-5148/$ - see front matter 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.reactfunctpolym.2006.06.001
1576
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
high efficiency [2]. Polyisoprene or natural rubber is one type of dielectric material which has many useful characteristics: inexpensive due to its natural source, flexible polymer, low swelling in water, high tensile strength, good resilience, high hot tensile, and well behaved hysteresis. These characteristics are desirable properties required to induce large actuation strain when subjecting to an electric field. Recently, the insertion of a conductive polymer into a dielectric elastomer forming a blend has been of keen interest. Conductive polymers can offer a variety of benefits to the host elastomer: variable conductivity, better thermal stability, and mechanical properties [3]. Examples are a polyanilene-polyisoprene blend for a biosensor application [4], a polyanilene-EPDM blend [5], and TiO2 embedded in PDMS gels for actuator applications [6]. In our work, we are interested in developing and testing polythiophene/polyisoprene elastomer blends as a substitute for artificial muscles. The mechanical properties, viscoelastic properties and electrical properties will be investigated in terms of degree of crosslinking ratio, polythiophene particle concentration, and electric field strength. 2. The electrorheological effect For a dielectric elastomer, a simplified description of the ER effect in solid-like matrices is introduced and summarized here. The nature of the ER effect in polymer gels or elastomers can be explained using the one point dipole model. Polarization in the point dipole model occurs not at the surface of the particle but within it. If dipoles form within particles, an interaction between dipoles occurs and becomes stronger as the dipoles come close to each other. When the particles make contact with each other along the applied electric field, the interaction reaches a maximum. A balance between the electrostatic interaction and the elastic modulus of the solid matrix is important for the ER effect to transpire. If the elastic modulus of the solid matrix is larger than the sum of the interaction between particles, the ER effect may not be observed macroscopically. Therefore, the matrix should be a soft material such as gels or elastomers to produce a significant ER response [7]. We may discuss theoretically the influence of microscopic interaction between polarized particles on the macroscopic mechanical properties of the polymer blends, in particular, the elastic storage modulus. Let us consider a hard sphere in a contin-
uous matrix under an electric field. When the relative dielectric constant of the particle (ep) is larger than that of the matrix (em), a point dipole in the particle can be generated by application of an electric field. According to the classical theory [7], the point dipole moment is given by: l ¼ 4pr3 e0 em jE where j ¼ ðep em Þ=ðep þ 2em Þ
ð1Þ ð2Þ
Eq. (1) is an expression obtained formally for the volume polarizability of a particle, where r is the radius of the particle, e0 is the permittivity in vacuum (=8.854 · 1012 F/m), and E is the applied electric field strength. When two particles are aligned along the applied electric field and are in contact with each other, the dipole–dipole interaction (F) between two dipole particles is given by the following equation [7]: F ¼ ð3=2Þpr2 e0 em j2 E2
ð3Þ
In a cubical gel, there are many straight migration paths of dispersed particles, which remain parallel to the direction of the applied field. The cube gel is deformed by a small shear strain in a direction perpendicular to the applied field. We may assume that electrostatic interactions between the dispersed particles in a path are based only upon the interaction between two adjacent particles and the interaction between paths of particles is negligible. Macroscopic mechanical properties such as storage and loss moduli can be estimated by multiplying the electrostatic force between adjacent particles at a short range in the path, therefore, an increase in elastic modulus due to an applied electric field DG can be calculated as follows [7]: DG ¼ ð9=4ÞCem j2 E2
ð4Þ
where DG is the change in the storage modulus, C is the volume fraction of particles, em is the relative dielectric constant of matrix, and E is the intensity of the applied field. This states that DG is proportional to C, em, and E2. When these factors reach maximum values, DG is expected to become saturated. 3. Experimental 3.1. Materials and synthesis of poly(3thiopheneacetic acid) (Pth) 3-Thiopheneacetic acid, 3TAA (AR grade, Fluka), was used as the monomer. Anhydrous ferric chloride, FeCl3 (AR grade, Riedel-delHean), was
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
used as the oxidant. Chloroform, CHCl3 (AR grade, Lab-Scan), and methanol, CH3OH (AR grade, LabScan), dried over CaH2 for 24 h under nitrogen atmosphere and then distilled, were used as solvents. Diethyl ether was used as a solvent. Polyisoprene, PI (Mw = 40,000, viscosity of 400 Poise, Aldrich), was used as the polymer matrix. Dicumyl peroxide (AR grade, Fluka) was used as the crosslinking agent. The reaction was by the oxidative-coupling polymerization according to the method of Kim et al. [8]. 3-Thiopheneacetic acid (10.0 g) was refluxed for 24 h in 50 ml of dry methanol with one drop of concentrated H2SO4 in order to protect the oxidative decomposition of the carboxylic acid group of monomer during the oxidative-coupling polymerization. Methanol was evaporated, and the residue was extracted with fresh diethyl ether. The extract was washed with deionized water, dried with anhydrous MgSO4 and then filtered. Diethyl ether was evaporated from the filtrate by a rotating evaporator. Thiophene methylacetate (TMA) product was obtained after the diethyl ether was evaporated from the filtrate by a rotating evaporator. In a 100 ml three-necked flask, a solution of 10 mmol of protected monomer in 20 ml of chloroform was added dropwise to a solution of 40 mmol of ferric chloride, which had been dissolved in 30 ml of dry chloroform under nitrogen atmosphere. The molar ratio of the oxidant to monomer was 4:1 in all cases. The reaction was carefully maintained at 0 C (±0.5 C) for 24 h. The reaction mixture was precipitated, by pouring into a large excess amount of methanol (1 L) to obtain polythiophene methylacetate (PTMA). The product was repeatedly washed with methanol and deionized water. PTMA was hydrolyzed as follows: 0.5 g of PTMA was dissolved in 50 ml of 2.0 M NaOH solution and heated for 24 h at 100 C. The mixture was filtered, neutralized, and precipitated with a dilute HCl solution (0.5 M) to yield the polymer. The poly(3-thiopheneacetic acid) (Pth) was washed several times with deionized water before vacuum drying at room temperature for 2 days. 3.2. Characterization of poly(3-thiopheneacetic acid) (Pth) Each poly(3-thiopheneacetic acid) sample was identified for its functional groups by a FT-IR spectrometer (Thermo Nicolet, Nexus 670) operated in
1577
the absorption mode with 32 scans and a resolution of ±4 cm1, covering a wavenumber range of 4000– 400 cm1, using a deuterated triglycine sulfate detector. Optical grade KBr (Carlo Erba Reagent) was used as the background material. The synthesized Pth particles were ground with a mortar, mixed with KBr at a ratio of Pth:KBr = 1:20, and compression molded into pellets under the pressure of 8000 kg. UV–Vis spectra were recorded with an UV–Vis absorption spectrometer (Perkin–Elmer, Lambda 10). Measurements were taken in the absorbance mode in the wavelength range of 200– 800 nm. Synthesized Pth was ground into a fine powder, dissolved in DMSO at the concentration of 6.0 · 105 M, and pipetted into the sample holder. Scan speed was 240 mm/min, and a slit width of 2.0 nm was used with a deuterium lamp as the light source. The thermal stability of Pth was investigated using a thermogravimetric analyzer (DuPont, model TGA 2950) at a temperature range between 30 and 800 C, with a heating rate of 10 C/ min, and under O2 atmosphere [9]. Scanning electron micrographs were taken with a scanning electron microscope (JEOL, model JSM-5200) to determine the morphology of poly(3-thiopheneacetic acid) in powder forms and polythiophene/polyisoprene blends of various particle concentrations. The scanning electron micrographs of polythiophene and polymer blends were obtained by using an acceleration voltage of 20 kV with magnifications of 350 and 1500 times. The particle sizes of Pth were measured by a particle size analyzer (Malvern Instruments Ltd., Masterizer X). The electrical conductivity of undoped Pth was measured by a custom-built two-point probe coupled with an electrometer (Keithley, model 6517A). The specific conductivity r (S/cm) values of the pellets were obtained by measuring the bulk pellet resistance R (X). The relation r = (1/Rt)(1/K) = (I/Vt)(1/K) was used to calculate specific conductivity, where t is the pellet thickness (cm), I is the current change (A), V is the applied voltage (voltage drop) (V) and K is the geometric correction factor which is equal to the ratio w/l, where w and l are the probe width and the probe length, respectively. The geometrical correction factor (K) was determined by calibrating the two-point probe with semi-conducting silicon sheets of known resistivity values. These two probes were connected to a voltmeter (Keithley, model 6517A), which supplied a constant voltage source, and a resultant current was measured. Electrical conductivity values of several samples were
1578
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
first measured at various applied voltages to identify their linear Ohmic regimes. The crosslink density of PI was determined by measuring Mc at the so-called equilibrium swelling. Mc was determined from the swelling ability of certain solvents in a crosslinked network. The procedure of the crosslink density measurement was as follows. Initially the samples were cut to 10 · 10 mm, and then the weight of each specimen was measured. We put each specimen in a vial containing toluene for 3 days and the weight of the swollen sample was remeasured. For each crosslinking ratio, three samples were used and three experiments were carried out, and the data were averaged. The crosslink density was calculated from the following equations [10–13]:
Mc ¼
1=3 v1 q2 v2m v2m 2
½lnð1 v2m Þ þ v2m þ v1 v22m w0 v2m ¼ vequil q2 w0 ws w0 þ vequil ¼ q2 q1 v1 v ¼ 0:34 þ ðd1 d2 Þ2 RT 1 where crosslink density ¼ ½2M c ;
ð5Þ ð6Þ ð7Þ ð8Þ ð9Þ
v1 is the molar volume of solvent (Mw/density), q2 is the polyisoprene density equal to 0.92 g/cm3, q1 is the toluene density equal to 0.867 g/cm3, w0 is the original polymer weight, ws is the swollen polymer weight, v is the polymer-solvent interaction parameter, R is the universal gas constant, 8.29 Nm/mol K, T is the absolute temperature, 298 K, d1 is the solubility parameter of polyisoprene equal to 17.02 (MPa)1/ 2 , and d2 is the solubility parameter of toluene equal to 18.20 (MPa)1/2. 3.3. Electrorheological properties measurements and sample preparations The electrorheological properties of pure polyisoprene were investigated in terms of electric field strength and crosslinking ratio. A specific amount of dicumyl peroxide was mixed with polyisoprene and cast onto a mold (diameter 25 mm). Various amounts of dicumyl peroxide were used in order to vary the degree of crosslinking (crosslink ratios NDCP/NPI of 2, 3, 5, and 7). The existing bubbles were removed in a vacuum atmosphere at 25 C for 2 h. The fabrication of the crosslinked polyiso-
prenes was by compression molding at 175 C and 5 MPa for 7 min [14]. The polythiophene (Pth) powder was sieved with a mesh particle size of 38 lm and dried at room temperature for 24 h prior to their uses. The Pth/PI blends were prepared by mechanical blending of undoped synthesized polythiophene particles at various particle concentrations (5, 10, 20, and 30 vol%). Polyisoprene fluid with a specific amount of crosslinking agent was mixed together, and the mixture was mechanically stirred for about 5 min at room temperature. The optimum degree of crosslink was chosen (NDCP/NPI = 3) since this material had the highest G 0 sensitivity. A specific amount of Pth particles was then added and the mixture was mechanically stirred for 5 min at room temperature. The mixture was cast on the mold (diameter 25 mm) and bubbles were removed under a vacuum atmosphere at 25 C for 2 h. The fabrication of the Pth/crosslinked polyisoprenes blends (Pth_U/ PI_03) was carried out by compression molding at 175 C and 5 MPa for 7 min [14]. A melt rheometer (Rheometric Scientific, model ARES) was used to measure rheological properties. It was fitted with a custom-built copper parallel plates fixture (diameter of 25 mm). DC voltage was applied by a DC power supply (Instek, model GFG 8216A), which can deliver electric field strength up to 2 kV/mm. A digital multimeter was used to monitor voltage input. In these experiments, the oscillatory shear strain was applied and the dynamic moduli (G 0 and G00 ) were measured as functions of frequency and electric field strength. Strain sweep tests were first carried out to determine the suitable strains to measure G 0 and G00 in the linear viscoelastic regime. The appropriate strain was determined to be 220% for pure polyisoprene fluid (PI_00). For the crosslinked polyisoprenes (PI_02, PI_03, PI_05 and PI_07) and polythiophene/crosslinked polyisoprene blends (at blend compositions of 5, 10, 20, and 30 vol% corresponding to Pth_U5/PI_03, Pth_U10/PI_03, Pth_U20/PI_03, and Pth_U30/PI_03), the strain of 15% was used. The frequency sweep tests were carried out to measure G 0 and G00 of each sample as functions of frequency. The deformation frequency was varied from 0.1 to 100 rad/s. Prior to each measurement, each polyisoprene or polythiophene/polyisoprene blends sample was presheared at a low frequency (0.04 rad/s) with the electric field on for 10 min in order to attain the equilibrium polarization. Each measurement was carried
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
out at temperature of 27 C and was repeated at least two or three times. 4. Results and discussion 4.1. Characterization of poly(3-thiopheneacetic acid) An FT-IR spectrum of the synthesized Pth was taken to identify characteristic absorption peaks [8]. The characteristic peaks of Pth were at 3200– 3000 cm1, 3000–2800 cm1, 1700 cm1, 1300– 1200 cm1, and 830 cm1. These peaks can be assigned to the C–H bond stretching on the thiophene ring; the aliphatic C–H bond stretching; the carboxylic acid C@O stretching; the thiophene ring stretching; the carboxylic acid C–O stretching, and the out-of-plane thiophene C–H stretching, respectively [8]. The most distinct feature of this spectrum is the extremely broad O–H absorption occurring in the region from 3400 cm1 to 2400 cm1; this peak can be attributed to the strong hydrogen bonding of the dimmers. This absorption band often obscures the C–H stretching vibration peak occurring in the same region. It is obvious from the absorption peak at around 1700 cm1 that the ester groups were not deteriorated during the oxidative polymerization. A UV–Vis absorption spectrum of synthesized Pth/DMSO solution shows absorption peaks at 275 nm and 415 nm, corresponding to the p–p* transition of the bithiophene unit and the polymer backbone, respectively [8,15,16]. A TGA thermogram of synthesized Pth shows two degradation steps at 200 C and 440 C, corresponding to the side chain degradation and the backbone degradation, respectively [9]. The polythiophene/polyisoprene blends (Pth_U/PI_03) have a better thermal stability with increasing polythiophene particle concentration. The mean particle diameter of Pth was determined to be approximately 20 lm with standard deviation of 3 lm. The particle microstructure was observed using scanning electron microscopy (SEM). Fig. 1 shows the shapes of the undoped Pth and undoped Pth particles in polymer blends; they are quite irregular in shape. However, these irregular shaped particles appear to be well dispersed in the polyisoprene matrix [9]. The specific conductivity of undoped Pth (Pth_U) was measured by the custom-built two point probe (Keithley, model 6517A). The specific conductivity of Pth_U was 3.01 · 106 S/cm with
1579
a standard deviation of 1.01 · 107 S/cm. The specific conductivity of polyisoprene was 8.18 · 107 S/cm with a standard deviation of 1.13 · 107 S/cm. 4.2. Electrorheological properties of pure polyisoprene 4.2.1. Effect of crosslink in the absence of electric field The effect of crosslink on the rheological properties of pure polyisoprene (PI) was first investigated. The crosslink ratios, NDCP/NPI, were 2, 3, 5, and 7 (PI_02, PI_03, PI_05, and PI_07). For the PI_02 system, NDCP/NPI and the crosslink density were 2.00 and 4.058 · 105 mol/g, respectively. For the highly crosslinked system of PI_07, NDCP/NPI and the crosslink density were 7.02 and 16.495 · 105 mol/g, respectively. The crosslink density increases with increasing crosslink mole ratio; data are tabulated in Table 1. Fig. 2(a) and (b) shows the storage modulus (G 0 ) and the loss modulus vs. frequency at electric field strengths of 0 and 2 kV/mm, and at strain of 1%. G 0 increases monotonically with increasing crosslink ratio [1]. On the other hand, the loss modulus (G00 ) decreases with increasing crosslink ratio, as shown in Fig. 2(b). Increasing the crosslink ratio or the crosslink density prohibits free movements of polymer chains and consequently less molecular friction. The polyisoprene rheological characteristic changes from a fluid-like behavior (PI_03) to a solid-like behavior (PI_07) along with the corresponding increase in the storage modulus [7]. 4.2.2. Effect of crosslink under electric field The effect of electric field strength on the rheological properties of polyisoprene (PI) at various crosslink ratios was investigated in the electric field range between 0 and 2 kV/mm. Fig. 3 shows the storage modulus responses (DG 0 ) of various crosslinked polyisoprene systems as functions of electric field strength at frequency of 1 rad/s, and strain of 1%. Above a critical electric field of 500 V/mm, the storage modulus responses appear to increase monotonically with electric field strength. The storage modulus responses (DG 0 ) at electric field strength of 2 kV/mm are 500, 4097, 5002, and 8894 Pa for PI_02, PI_03, PI_05, and PI_07, respectively. The corresponding storage modulus sensitivity values, 0 ðxÞ defined as DG , of the crosslinked polyisoprene sysG0 ðxÞ 0
tems (PI_02, PI_03, PI_05, and PI_07) at electric
1580
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
Fig. 1. The morphology of polythiophene particles and polythiophene/ polyisoprene blends: (a) synthesized P3TAA at a magnification of 1500; (b) Pth_U5/PI_03; (c) Pth_U10/PI_03; (d) Pth_U20/PI_03; and (e) Pth_U30/PI_03 at a magnification of 350.
Table 1 The crosslink density of pure polyisoprene Systems
NDCP/NPI
Crosslink density (105 mol/g)
PI_02 PI_03 PI_05 PI_07
2.00 3.02 5.01 7.02
4.058 ± 0.312 6.638 ± 1.000 11.797 ± 0.625 16.495 ± 1.750
forces that act to pull the electrodes together [2,17]. This attraction causes a compressive force on the dielectric elastomeric in the direction of the electric field throughout the volume between the electrodes. The resulting effective (squeeze) pressure can be written as [2]: p ¼ e0 er E2
field strength of 2 kV/mm are 10%, 60%, 25%, and 30%, respectively. As an electric field is applied, electrical dipole moments are generated and the electrostatic interaction between the polymer chains is induced leading to an intermolecular interaction acting like an electrical network. In addition, a voltage differential between the electrodes can generate electromagnetic
ð10Þ
where e0 and er are the permittivity of free space and the relative permittivity of polymer, respectively. E is the applied electric field. The response of our polymer systems is functionally similar to those of electrostrictive polymers, in which the response is directly proportional to the square of the applied electric field [1,17,18]. The intermolecular interaction and the electrostatic interaction result in the loss of free chain movements, the higher chain rigid-
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
1581
G' (Pa)
1e+5
1e+4 PI_03, E = 0 PI_03, E = 2 PI_05, E = 0 PI_05, E = 2 PI_07, E = 0 PI_07, E = 2
kV/mm kV/mm kV/mm kV/mm kV/mm kV/mm
1e+3 .01
.1
1
10
100
1000
10
100
1000
ω (rad/s)
(a) 1e+5 PI_03, E = 0 PI_03, E = 2 PI_05, E = 0 PI_05, E = 2 PI_07, E = 0 PI_07, E = 2
G" (Pa)
1e+4
1e+3
kV/mm kV/mm kV/mm kV/mm kV/mm kV/mm
1e+2
1e+1
1e+0 .01
.1
(b)
1
ω (rad/s)
Fig. 2. Storage and loss moduli of polyisoprene systems at various crosslink ratios vs. frequency, strain 1%, 27 C, and at electric field strengths of 0 and 2 kV/mm: (a) storage modulus, G 0 (x) and (b) loss modulus, G00 (x).
ity, and correspondingly higher G 0 (x). Yang et al. [18] reported a similar finding on the effect of increasing effective (squeeze) pressure; the radial and circumferential stresses changed from a tensile state to a compressive state at a critical effective pressure. Fig. 4 shows the storage modulus responses (DG 0 ) as functions of crosslink density at electric field strengths of 1 kV/mm and 2 kV/mm. The storage modulus responses increase almost linearly with increasing crosslink density at both electric field strengths. The highest crosslink density system (PI_07, crosslink density = 16.495 · 105 mol/g) possesses the highest storage modulus response of
4000 and 8894 Pa at electric field strengths of 1 and 2 kV/mm, respectively. We note that the 0 ðxÞ storage modulus sensitivity value defined as DG G00 ðxÞ of the crosslinked polyisoprene system PI_03 (crosslink density = 6.638 · 105 gmol/ml), has the highest G 0 sensitivity value: 60% at electric field strength of 2 kV/mm. This result suggests that the PI_03 system contains a balanced rheological characteristic between the solid-like and the fluid-like behaviors; it has enough flexibility within the matrix in the absence of electric field in such a way that its structure rigidity can be further induced with electric field. On the other hand, the lower crosslinked system (PI_02)
1582
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
ΔG' (Pa)
10000
8000
PI_00, G'0 = 1.50 Pa PI_02, G'0 = 4812 Pa
6000
PI_03, G'0 = 7401 Pa PI_05, G'0 = 19155 Pa PI_07, G'0 = 27201 Pa
4000
2000
0
-2000 100
1000
E (V/mm) Fig. 3. Storage modulus responses of polyisoprene at various crosslink ratios vs. electric field strength, at frequency 1 rad/s, strain 1%, (PI_00 system, strain 220%), and at 27 C.
12000
E=1 E=1 E=2 E=2
10000
kV/mm; freq = 1.0 kV/mm; freq = 10 kV/mm; freq = 1.0 kV/mm; freq = 10
rad/s rad/s rad/s rad/s
ΔG'
8000
6000
4000
2000
0 0
5
10
15
20
Crosslink density (10 -5 mol/g) Fig. 4. Storage modulus responses vs. crosslink density at 27 C, and at electric field strengths of 1 and 2 kV/mm.
has a more fluid-like structure than other systems; free movement and relaxation of polymer chains are still allowed in the presence of electric field and consequently they are least affected by electric field. The higher crosslinked systems (PI_05 and PI_07) contain relatively more solid-like structures and thus they possess relatively higher storage modulus values. The polymer segments and chains are quite rigid and fixed in the absence of electric field; they simply cannot respond any further to intermolecular interaction under the applied electric field.
4.2.3. Effect of polythiophene composition in the absence of electric field The effect of polythiophene particle concentration on the rheological properties of polymer blends (Pth_U/PI) was investigated next. The PI_03 system, with crosslink density of 6.638 · 105 mol/g, was chosen and blended with poly(3-thiopheneacetic acid) particles. This PI_03 system exhibits the maximum G 0 sensitivity amongst the various systems studied. Figs. 5(a) and (b) show comparisons of the storage modulus (G 0 ) and the loss modulus
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
1583
1e+6
G' (Pa)
Pth_U5/PI_03, E = 0 Pth_U5/PI_03, E = 2 Pth_U30/PI_03, E = 0 Pth_U30/PI_03, E = 2
kV/ mm kV/ mm kV/ mm kV/ mm
1e+5
1e+4
.01
.1
1
1e+6
Pth_U5/PI_03, E = 0 Pth_U5/PI_03, E = 2 Pth_U30/PI_03, E = 0 Pth_U30/PI_03, E = 2
1e+5
G" (Pa)
10
100
1000
10
100
1000
ω (rad/s)
(a)
kV/ mm kV/ mm kV/mm kV/mm
1e+4
1e+3
1e+2
.01
.1
1
(b)
ω (rad/s)
Fig. 5. Comparison of the storage and the loss moduli of polythiophene/polyisoprene blends (Pth_U/PI_03) at various particle concentrations (5 and 30 vol%), strain 1%, 27 C, and at electric field strengths of 0 and 2 kV/mm: (a) the storage modulus, G 0 and (b) the loss modulus, G00 .
(G00 ) vs. frequency between polymer blends of various concentrations, at electric field strengths of 0 and 2 kV/mm. Polythiophene particle concentrations studied were 5, 10, 20, and 30 vol.% (Pth_U5/PI_03, Pth_U10/PI_03, Pth_U20/PI_03, and Pth_U30/PI_03). Both G 0 and G00 increase with polythiophene particle concentrations at all frequencies. The storage modulus G 0 (x) and the loss modulus G00 (x) of each polymer blends system are higher than those of the PI_03 system at any electric field strength. Close inspection of our data suggests that polythiophene particles in the polymer blends behave as fillers in the matrix; they can store or
absorb the forces/stresses within the matrix [1]. A polymer blend system with a higher particle concentration is expected to exhibit a higher internal stress response, a higher storage modulus G 0 (x), and a higher loss modulus G00 (x) response than those of pure PI_03 system. Concentration effect has been reported by Shiga [7] who found that the shear storage modulus of silicone gel containing PMACo particles increased with increasing volume fraction of dispersed particles embedded in the gel. Krause et al. [1] reported that the compression modulus, in the absence of an electric field, increased by a factor of 2 when
1584
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
the PANI particle concentration was increased from 0 to 20 wt%. These results were expected because the particles acted as filler particles capable of supporting internal stresses within the matrices. 4.2.4. Effect of polythiophene composition under electric field The effect of electric field strength on the rheological properties of polythiophene/polyisoprene blends (Pth_U/PI) at various polythiophene particle concentrations was investigated in the range between 0 and 2 kV/mm. Figs. 5(a) and (b) show the comparison of the storage modulus (G 0 ) and the loss modulus (G00 ) between the blends of various concentrations at electric field strengths of 0 and 2 kV/mm. The storage modulus (G 0 ) and the loss modulus (G00 ) of each polymer blend system generally increase with increasing electric field strength. Fig. 6 shows the storage modulus responses (DG 0 ) vs. electric field of various polymer blend systems at frequency of 1 rad/s, and strain equal to 1%. The increase in DG 0 with electric field is nonlinear within the range of 0.1–2 kV/mm. The storage modulus response values, DG 0 (x), of these systems at electric field strength of 2 kV/mm are 4097, 5446, 4026, 19161, and 33857 Pa for PI_03, Pth_U5/ PI_03, Pth_U10/PI_03, Pth_U20/PI_03, and Pth_U30/PI_03, respectively. The corresponding storage modulus sensitivity values, defined as DG0 ðxÞ , of these polymer blends systems (Pth_U5/ 0 G0 ðxÞ PI_03, Pth_U10/PI_03, Pth_U20/PI_03, and
Pth_U30/PI_03) at electric field strength of 2 kV/ mm are 50%, 35%, 110%, and 45%, respectively. In the absence of the electric field, the polythiophene particles are randomly dispersed within the polyisoprene matrix and there is no particle–particle interaction. When the electrical field is applied, both polyisoprene and polythiophene particles become polarized and induced dipole moments are generated, leading to intermolecular interactions. These intermolecular interactions induce the loss of chain free movements and the higher chain rigidity, as indicated by higher G 0 (x) values. The electric field evidently enhances the elastic modulus of our conductive polymer-polyisoprene blends, consistent with the results of other polymer blends [17–19]. Lui and Shaw [19] reported that for the silica/silicone elastomers, their moduli exhibited a nearly quadratic dependence on electric field when the particles were randomly dispersed. For the aligned structure, the shear modulus increased quadratically with low electric field values but became saturated at high values. Shiga [7] reported that for the PMACo particles in silicone gel, the increase in the elastic modulus induced by an electric field was 4 kPa with the particle volume fraction of 0.3, and the applied electric field of 2 kV/mm. Fig. 7 shows the storage modulus responses (DG 0 ) of polythiophene/polyisoprene blends vs. particle volume fraction at electric field strengths of 1 and 2 kV/mm. For both electric field strengths of 1 and 2 kV/mm, the storage modulus increases line-
40000 PI_03,
G'0 = 7401 Pa
Pth_U5/PI_03, G'0 = 11085 Pa Pth_U10/PI_03, G'0 = 12416 Pa
30000
ΔG' (Pa)
Pth_U20/PI_03, G'0 = 19428 Pa Pth_U30/PI_03, G'0 = 78098 Pa
20000
10000
0 100
1000
E (V/mm) 0
Fig. 6. Storage modulus responses (DG ) of the polythiophene/polyisoprene blends (Pth_U/PI_03) at various particle concentrations vs. electric field strength, at frequency 1 rad/s, strain 1%, and at 27 C.
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
1585
50000
Δ G' (Pa)
40000
E=1 E=1 E=2 E=2
kV/ mm; freq = 1.0 kV/ mm; freq = 10 kV/ mm; freq = 1.0 kV/ mm; freq = 10
rad/s r ad/s rad/s r ad/s
30000
20000
10000
0 0.00
.05
.10
.15
.20
.25
.30
φPth Fig. 7. Storage modulus responses (DG 0 ) of the polythiophene/polyisoprene blends as functions of particle volume fraction, 27 C, of various frequencies, and at electric field strengths of 1 and 2 kV/mm.
arly with particle volume fraction below 0.10; and it increases nonlinearly when particle volume fraction was above 0.10. For electric field strength of 1 kV/ mm, the storage modulus response values are 1728, 2592, 1996, 14053, and 8322 Pa for PI_03, Pth_U5/PI_03, Pth_U10/PI_03, Pth_U20/PI_03, and Pth_U30/PI_03, respectively. We also note that the storage modulus sensitivity value, defined 0 ðxÞ as DG , of polythiophene/polyisoprene blends G0 ðxÞ 0
Pth_U20/PI_03 system attains a maximum G 0 sensitivity value: 110% at electric field strength of 2 kV/ mm. At very low concentrations (Pth_U5/PI_03 and Pth_U10/PI_03), the number of particles is too small and the distances between particles are too large to allow a significant particle interaction [19]. The storage modulus response values of these systems are comparable to those of the crosslinked and the pure polyisoprene system (PI_03). At higher concentrations (Pth_U20/PI_03 and Pth_U30/PI_03), the distances between particles are smaller and hence there is a stronger interparticle interaction [7,19]. Evidently, the electric field strength of 1 kV/mm is presumably not strong enough to induce polarization interactions amongst all particles at high particle concentrations due to the steric hindrance effect [19]. The storage modulus response of Pth_U30/PI_03 system becomes saturated at approximately the same value as that of Pth_U20/PI_03 system. The storage modulus response shows a nonlinear
dependence on the particle volume fraction and it attains a saturation state at volume fractions above 0.2. On the other hand, at the higher electric field strength of 2 kV/mm, induced polarization interactions are more effective amongst particles at high particle concentrations. No saturation state was observed at high particle concentrations. Liu and Shaw [19] reported a similar effect for silica/silicone elastomers. They found that the enhancement of shear modulus was negligible below 8.0 vol%, but increased dramatically above this threshold concentration. At volume fraction above 55 vol%, the shear modulus decreased because the interparticle force decreased due to the steric hindrance effect. Fig. 8 shows electrical current vs. electric field of various blend systems studied. It can be seen that the leakage current remains below 2 lA at the electric field strength of 2 kV/mm. The case of the highest particle concentration of 30 vol%, 30Pth_U/ PI_03, in which the leakage current attains a value of about 7 lA at electric field strength of 2 kV/ mm, is an exception. 4.2.5. Electrorheological temporal response Finally, we investigated the temporal characteristics of the crosslinked polyisoprene (PI_03) and the polythiophene/polyisoprene blend at particle concentration of 20 vol% (Pth_U20/PI_03) at various electric field strengths. The temporal characteristic of each sample was recorded in the
1586
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588 8
Ι (μΑ)
6
PI_03 5Pth_U/PI_03 10Pt h_U/PI_03 20Pt h_U/PI_03 30Pt h_U/PI_03
4
2
0 0
500
1000
1500
2000
2500
E (V/mm) Fig. 8. Current (I) vs. applied electric field strength (E) of the polyisoprene system and polythiophene/polyisoprene blend systems at various polythiophene particle concentrations, and at 27 C.
linear viscoelastic regime at a strain of 1% and at frequency of 1 rad/s. Fig. 9(a) shows the change in G 0 of PI_03 system at electric field strengths of 1 and 2 kV/mm during a time sweep test, in which an electric field was turned on and off alternately. At the electric field of 1 kV/mm, G 0 immediately increased and rapidly reached a steady-state value. With the electric field turned off, G 0 decreased and nearly recovered its original value. Thus PI_03 appears to be a reversible system at 1 kV/mm. At the electric field of 2 kV/mm, the response of G 0 can be divided into two regimes: the initial regime in which G 0 rapidly overshooted to a large value on the first cycle followed by an irreversible decay with electric field off, and the steady state regime in which G 0 subsequently exhibited a reversible cyclic response. The time required for G 0 to reach the steady-state value on applying the field is called the induction time, sind. As shown in Table 2, sind decreases with increasing electric field strength; they are 107 and 75 s at electric field strengths equal to 1 and 2 kV/mm, respectively. The time required for G 0 to decay towards its steady-state value when the electric field is turned off is called the recovery time, srec. It is essentially independent of the electric field strength, as shown in Table 2. The recovery times are 76 and 83 s at electric field strengths equal to 1 and 2 kV/mm, respectively. The independence of srec on the field strength suggests that the strains induced were small such that the relaxations were nearly the same. We also found that tind is independent of crosslink density
or the elastic modulus, whereas trec decreases with increasing crosslink density, as shown in Table 2. The independence of sind with the crosslink density suggests that sind depends critically more on the electric field strength. The lower srec value required to relax with a higher crosslink density is because a higher elastic modulus system generally has a shorter relaxation time scale. Fig. 9(b) shows the temporal response of 20 vol% polythiophene/polyisoprene blend (Pth_U20/ PI_03) at electric field strengths of 1 and 2 kV/mm, respectively. The Pth_U20/PI_03 is clearly an irreversible system at both electric field strengths. Our result here suggests that there are some irreversible interactions between polythiophene particles, perhaps due to hydrogen bonding between adjacent polythiophene particles, and residual dipole moments inducing permanent interparticle interactions [9]. sind decreases with increasing field strength, as shown in Table 2. sind are 157 and 121 s at electric field strengths equal to 1 and 2 kV/mm, respectively. srec is nearly independent of electric field strength. As shown in Table 2, srec are 59 and 66 s at electric field strength equal to 1 and 2 kV/mm, respectively. Finally, we may note that the equilibrium G 0 values of PI_03 and Pth_U20/PI_03 systems in Figs. 9(a) and (b) are not the same as those shown in Figs. 2(a) and 6, respectively. This arises because the samples of Figs. 2(a) and 6 were subjected to preshearing under a steady electric field for a duration of 10 min. On the other hand,
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
1587
G' (Pa)
8000
on
off
7000
E = 1 kV/mm E = 2 kV/mm
on off
6000 0
2000
4000
(a)
6000
8000
10000
12000
time (s) 70000
G' (Pa)
65000
60000 on off on off
55000
E = 1 kV/mm E = 2 kV/mm
on off on off
50000 0
2000
4000
6000
8000
10000
12000
14000
time (s)
(b)
Fig. 9. Temporal response of storage modulus (G 0 ) of: (a) PI_03 system at various electric field strengths (1 and 2 kV/mm) and (b) Pth_U20/PI_03 system at various electric field strengths (1 and 2 kV/mm), and at temperature of 27 C.
the fresh samples of Fig. 9(a) and (b) were not presheared but the electric field was periodically turned on and off. Table 2 Induction time and recovery time of pure polyisoprene systems and polythiophene/polyisoprene blends Samples
Electric field (kV/mm)
Induction time (sind) (s)
Recovery time (srec) (s)
DG0ind (Pa)
DG0re (Pa)
PI_03
1 2 2 2 1 2
107 75 73 78 157 121
76 83 74 64 59 66
195 110 202 851 2257 4794
188 106 175 821 35 80
PI_05 PI_07 Pth_U20/ PI_03
5. Conclusion In our work, electrorheological properties of polyisoprene and polythiophene/polyisoprene blends were investigated by examining the effects of crosslink ratio and polythiophene particle concentration on the dynamic moduli, G 0 and G00 , under the oscillatory shear mode at electric field strength varying from 0 to 2 kV/mm. In pure PI systems, the storage modulus (G 0 ) increased but the loss modulus (G00 ) decreased with increasing crosslinking ratio. It exhibited a transition from a fluid-like to a solid-like behavior as crosslink density was varied. The storage modulus (G 0 ) and the loss modulus
1588
T. Puvanatvattana et al. / Reactive & Functional Polymers 66 (2006) 1575–1588
(G00 ) of the uncrosslinked polyisoprene fluid exhibited no change with increasing electric field strength. For polyisoprenes with the crosslink ratios of 2, 3, 5, and 7, the storage modulus increased with electric field strength. The maximum G 0 sensitivity was found to be about 0.6 for the PI_03 system at electric field strength of 2 kV/mm; it is frequency independent. Poly(3-thiopheneacetic acid) particles were synthesized via an oxidative polymerization and blended with the PI_03 system. For the electrorheological properties of Pth_U/PI_03 blends with the polythiophene particle concentrations of 5, 10, 20, and 30 vol%, the dynamic moduli G 0 and G00 of each polymer blend, were generally higher than those of pure polyisoprene due to polythiophene particles within the matrix acting as fillers; they can store or absorb the forces/stresses within the matrix. The storage modulus responses increased with electric field nonlinearly within the range of 0.1–2 kV/ mm. This can be attributed to the fact that polyisoprene and polythiophene particles became polarized and induced dipole moments were generated, leading to intermolecular interactions along the direction of electric field. The storage modulus 0 sensitivity, DG , attained a maximum G 0 sensitivity G00 value of 0.5, 0.35, 1.10, and 0.45 at particle concentrations of 5, 10, 20, and 30 vol% at the electric field strength of 2 kV/mm, respectively. Acknowledgements The authors acknowledge the financial supports to A.S. from Chulalongkorn University (through a grant from the Ratchadapesak Somphot Endowment Fund for the foundation of the
Conductive and Electroactive Polymers Research Unit), and the Petroleum and Petrochemical College Consortium. References [1] S. Krause, K. Bohon, Macromolecules 34 (2001) 7179–7189. [2] R. Kornbluh, R. Perlrine, Q. Pei, S. Oh, J. Joseph, in: Proceedings of the First World Congress on Biomimetics, 2002. [3] S. Ku¨c¸u¨kyavuz, M. Sankir, Z. Ku¨c¸u¨kyavuz, Synthetic Met. 128 (2002) 247–251. [4] Z. Shen, H. Xue, Y. Li, Synthetic Met. 124 (2001) 345–349. [5] R. Faez, R.-H. Schuster, M.-A. De Paoli, Eur. Polym. J. 38 (2002) 2459–2463. [6] M. Zrı´nyi, J. Fehe´r, G. Filipcsei, Macromolecules 33 (2000) 5751–5753. [7] T. ShigaAdvances in Polymer Science, vol. 134, Springer, Verlag, Berlin Heidelberg, 1997, pp. 133–163. [8] B. Kim, L. Chen, J.P. Gong, Y. Osada, Macromolecules 32 (1999) 3964–3969. [9] D. Chotpattananont, A. Sirivat, A.M. Jamieson, Colloid Polym. Sci. 282 (2004) 357–365. [10] F. Horkay, G.B. McKenna, P. Deschamps, E. Geissler, Macromolecules 33 (2000) 5215–5220. [11] S.-S. Choi, J. Anal. Appl. Pyrol. 52 (1999) 105–112. [12] P.C. Painter, M.M. Coleman, Fundamentals of Polymer Science: An Introductory Text, Technomic Publishing Company, USA, 1997. [13] D.W. Wang, 2nd ed., Encyclopedia of Polymer Science and Engineering, vol. 1, A Wiley-Interscience Publication, New York, 1992. [14] R. Faez, M.A. De Paoli, Eur. Polym. J. 37 (2001) 1139–1143. [15] F. Wang, Y.-H. Lai, M.-Y. Han, Macromolecules 37 (2004) 3222–3230. [16] F. Demanze, A. Yassar, F. Garnier, Macromolecules 29 (1996) 4267–4273. [17] R.E. Perline, R.D. Kornbluh, J.P. Joseph, Sensor Actuat. Phys. A64 (1998) 77–85. [18] E.E. Yang, M. Frecker, E. Mockensturm, in: Proceedings of ASME International Mechanical Engineering, 2003. [19] B. Liu, T.M. Shaw, J. Rheol. 45 (3) (2001) 641–657.