1. Introduction
Ionic electroactive polymers (EAP) are among the most promising smart materials with applicability as electromechanical transducers. Compared with the other types of EAP based actuators, ionic actuators exhibits large deformations under low applied voltages (1–5 V) [
1]. Historically, the most common polyelectrolyte for preparation of ionic actuators is sulfonated tetrafluoroethylene copolymer, since the first electromechanical transducers were made using the commercially available Nafion (DuPont) [
2,
3]. Even though ion-polymer metal composites (IPMC) based on Nafion demonstrate high performance, there are drawbacks which limit its further development and practical applications such as fixed ion exchange capacitance (IEC) and proton conductivity, reduction mechanical and electrochemical properties at elevated temperature, low generated force, back-relaxation under direct current voltage and high cost. Many studies have been focused on the search for new polymers that can replace Nafion membranes [
4,
5]. Sulfonated commercially available polymers such as sulfonate polystyrene [
6], sulfonated poly(ether ether ketone) [
7], sulfonated poly(styrene-ran-ethylene) [
8], sulfonated poly(styrene-b-ethylene-co-butylene-b-styrene) [
9,
10,
11], sulfonated polyetherimide [
12], sulfonated polyimide [
13] sulfonated styrenic pentablock copolymer [
14,
15,
16], sulfonated polyphenylsulfone [
17], sulfonated poly(1,4-phenylene ether-ether-sulfone) [
18] were employed to prepare ionic actuators. IPMC actuators based on these polymers could operate only in the swollen state. The electrolyte is usually water or an aqueous salt solution. The main disadvantage of actuators with aqueous electrolyte is their unsuitability for prolonged operation in the open air. Both evaporation and electrolysis due to a narrow electrochemical window of water quickly and significantly reduce the efficiency of electromechanical devices. Two approaches were proposed to solve this problem: the use of non-volatile, stable electrolytes and encapsulation of actuators. Although encapsulated actuators are able to work in air for a long time [
19,
20,
21], this approach does not solve the problem of the electrochemical stability of water. The capsule also increases the total stiffness of the actuator, which reduces their efficiency. Another approach is to replace water with room temperature ionic liquids (IL) as an electrolyte [
22,
23,
24,
25]. ILs have unique and advantageous properties such as high ionic conductivity, thermal stability, wide electrochemical window, and immeasurably low vapor pressure. Polymer gel electrolytes based on ILs demonstrate conductivity at a room temperature above 10
−3 S/cm, which is enough for practical use [
26]. High efficiency of carbon nanotubes-based electrodes (bucky gel electrodes) in actuators comprising ionic liquid was shown by Asaka [
25,
27,
28]. On the one hand, membranes based on ionomers demonstrate conductivity significantly lower than pure electrolytes, on the other hand, the appearance of additional mobile cations increases the deformation of actuators based on such membranes [
29]. Watanabe described high performance printable polymer actuators based on soluble sulfonated polyimide comprising ionic liquid with carbon electrodes [
30]. More recently, fast response bucky gel actuator based on specially synthesized sulfonated block copolymer containing IL was described [
31,
32]. The most common ionic liquid for actuators based on inomers is 1-ethyl-3-methylimidazolium tetrafluoroborate (EMImBF4), and the actuators typically show a deformation of 0.4–0.6% at a voltage of 1–3 V.
Highly conductive sulfonated polymer based polynorbornene prepared using ring-opening metathesis polymerization (ROMP) for proton exchange membrane was described by Ramani [
33]. This polymer was synthesized by co-polymerization of norbornene substituted benzenesulfonyl chloride, norborbene and dicyclopentadiene and subsequent hydrolysis to form an ionomer. In this study, we describe the preparation of poly(phenyl norbornenes) synthesized by two polymerization routes, namely, ROMP and addition polymerization [
34], their sulfonation under homogeneous conditions, and their use as electrolyte membranes of ionic actuators. Homogeneous sulfonation of the polymer in different chlorinated solvents under mild conditions by propionyl sulfate was studied. The polymers were characterized by
1H NMR, FTIR, GPC and DSC. For electromechanical applications cation of the sulfonated polymer was exchanged by imidazolium and 1-methylimidazolium. The resulting polymers were blended with polyvinylidene fluoride (PVDF) and ionic liquid to achieve polyelectrolyte membranes. Finally, bucky gel actuators were fabricated and tested in terms of deformation and blocking force under 2 V DC and compared with classical counterpart.
2. Materials and Methods
2.1. Materials
Unless otherwise stated, all manipulations were carried out using standard Schlenk techniques under an argon atmosphere. 5-phenyl-2-norbornene was synthesized by the Diels-Alder reaction from styrene and dicyclopentadiene using standard procedure [
35]. Toluene, o-xylene was refluxed over Na and distilled in an argon atmosphere, stored over sodium wire. Dichloromethane, 1,2-dichloroethane and chloroform was distilled over CaH
2 in an argon atmosphere prior to use. Methanol, 2,2′-methylenebis (6-tert-butyl-4-methylphenol), p-toluenesulfonyl hydrazide, the first-generation Grubbs catalyst, Pd(OAc)
2, tricyclohexylphosphine, propionic anhydride, imidazole, 1-methylimidazole, ethyl bromide, sodium tetrafluoroborate, sodium tetrakis [3,5-bis(trifluoromethyl)phenyl]borate (NaBAr
F) were purchased from commercial sources (Acros) and were used without prior purification. PVDF purchased from the Konstantinov Kirovo-Chepetsk Chemical Combine as Fluoroplast-2. Single wall carbon nanotubes (SWCNT) purchased from the OCSiAl. 1-Ethyl-3-methylimidazolium tetrafluoroborate (EMImBF
4) was synthesized by literature procedure [
36].
2.2. Polymers Characterization
All NMR spectra were acquired at 600 MHz (
1H) on a Bruker Avance III Ultrashield spectrometer. NMR spectra were recorded in deuterated chloroform for non-ionic compounds and in deuterated dimethyl sulfoxide for sulfonated polymers. Chemical shifts were referenced to residual solvent signals. Fourier transform infrared (FTIR) spectra were acquired in the range of 4000–400 cm
−1 on Bruker Tensor-27 spectrophotometer using KBr pellets. The differential scanning calorimetry (TA Instruments Q20 V24.11) data were obtained in a sealed aluminum pan with a heating rate of 10 °C/min under N
2 purge. The molecular weights of the polymers were evaluated by gel permeation chromatography (GPC) on a Waters high pressure chromatograph (Microgel mix column 1–5 μm 500 × 7.7 mm Chrompack, sample volume was 100 μL, sample concentration was 1–2 mg/mL in chloroform) equipped with a refractometric detector. Calibration was performed according to polystyrene standards with molecular weights of 1 × 10
3 ÷ 1 × 10
6. The calculation of molecular mass characteristics was performed according to the calibration dependence, which is linear in the range 10
3 ÷ 10
6. Some characterization of sulfonated polymer prepared via ROMP is published previously as raw data without discussion [
37].
2.3. Polymerization of 5-phenyl-2-norbornene (ROMP)
The procedure is described on the example of an experiment with a monomer/catalyst ratio of 3000/1 and an initial monomer concentration in the reaction mixture of 0.28 M. Other experiments were carried out similarly. A solution of the first-generation Grubbs catalyst (1.0 × 10−3 M) in dry toluene was prepared immediately before polymerization. 1.00 g (5.9·mmol) of 5-phenyl-2-norbornene and 18 mL of dry toluene were introduced into a vial under argon atmosphere. Polymerization was initiated by adding 2.00 mL (2.0 × 10−6 mol) of the catalyst solution. Stirring was continued for 2 h. Each time the reaction mixture became so viscous that stirring was difficult, it was diluted with 5.0 mL of dry toluene (total volume: 40.0 mL). Polymerization was terminated by the addition of vinyl ethyl ether. The polymer was precipitated into methanol containing an oxidation inhibitor (2,2′-methylene bis (6-tert-butyl-4-methylphenol)). Then the polymer was filtered, washed 3 times with methanol and dried. The polymer was re-precipitated twice from toluene to methanol and dried under vacuum at 40 °C to constant weight. Yield: 0.88 g (88%). Mw = 1.2 × 106, Mw/Mn = 3.5.
1H NMR (δ, ppm, CDCl3): 7.40–6.71 (m., 5H, Ar−H), 5.67–4.44 (m., 2H, −HC=CH−), 3.42–0.72 (m., 7H).
2.4. Synthesis of Hydrogenated Poly(5-phenyl-2-norbornene) HPPNB
To a two-neck flask (1 L) equipped with a reflux condenser and a magnetic stirrer 15 g (88 mmol) of 5-phenyl-2-norbornene and 300 mL of dry xylene were added. Polymerization was initiated by adding a solution of 24·mg (2.9 × 10−5 mol) of the first-generation Grubbs catalyst in 3 mL of xylene. Stirring was continued for 1 h. Each time the reaction mixture became so viscous that stirring was difficult, it was diluted with 50 mL of dry xylene (total volume: 200 mL). Vinyl ethyl ether was added to reaction mixture to terminate polymerization. Then 66 g (0.35 mol) of p-toluenesulfonyl hydrazide was added to the reaction mixture. The mixture was boiled under stirring for 4 h. The resulting viscous polymer solution with unreacted p-toluenesulfonyl hydrazide was decanted and the product was precipitated into methanol. The product was filtered, washed 3 times with methanol and dried under vacuum. The polymer was re-precipitated twice from toluene to methanol and dried under vacuum at 40 °C to constant weight. Yield: 13.2 g (87%). Mw = 8.2 × 105, Mw/Mn = 3.5.
1H NMR (δ, ppm, CDCl3): 7.44–6.74 (m., 5H, Ar−H), 3.38–0.27 (m., 11H).
2.5. Polymerization of 5-phenyl-2-norbornene (Addition Polymerization)
The polymerization was carried out at a 5-phenyl-2-norbornene/Pd(OAc)2/NaBArF/PCy3 ratio of 3000/1/5/2 and an initial monomer concentration in the reaction mixture of 2.4 M. In a glovebox under argon atmosphere 6.8 mL of Pd(OAc)2 (0.01 M, 0.068 mmol) in dry chloroform, 6.8 mL of NaBARF (0.05 M, 0.34 mmol) in dry chloroform were mixed in a vial and 1 drop absolute methanol was added then 6.8 mL of a solution of PCy3 (0.02 M, 0.136 mmol) in dry chloroform and the resulting solution was stirred for 5 min. In a separate 500 mL flask 10.2 g of 5-phenyl-2-norbornene (0.06 mol) was dissolved in 9 mL of dry chloroform. The polymerization was initiated by adding 6.0 mL of the catalytic mixture (0.02 mmol Pd) to the flask with the monomer under stirring. Stirring was continued for another 20 h. As the viscosity increased, the reaction mixture was diluted stepwise with dry chloroform (the total volume of the mixture was 160 mL). The reaction mixture was precipitated into methanol, the polymer was washed 3 times with methanol and dried in vacuo. Then the polymer was twice reprecipitated from toluene into methanol and dried in vacuum at 80 °C to constant weight. Yield: 6.1 g (60%). Mw = 3.1·105, Mw/Mn = 1.4.
1H NMR (δ, ppm, CDCl3): 8.37–6.07 (m., 5H, Ar–H), 3.38–1.27 (m., 11H).
2.6. Copolymerization of 5-phenyl-2-norbornene and 5-docecyl-2-norbornene (Addition Polymerization)
The copolymer was synthesized by a procedure similar to that described above with the same ratio of catalyst and monomers. The procedure was repeated twice with a molar ration of 5-phenyl-2-norbornene to 5-docecyl-2-norbornene 10/1.3 (total mass 2.05 g) and 10/2.6 (total mass 2.20 g). The reaction mixture was stirred for 5 h instead of 20 h and was diluted 4.5 times. Yield: 0.86 (42%), Mw = 1.17 × 106, Mw/Mn = 1.69, monomer ratio 27/73 (Dodecyl/Ph); 0.77 g (35%), Mw = 2.88 × 106, Mw/Mn = 2.51 monomer ratio 1/1 (Dodecyl/Ph).
1H NMR (δ, ppm, CDCl3): 8.07–6.20 (m., Ar−H), 3.34–0.57 (m., aliphatic).
2.7. Sulfonation Procedure
3.00 g of polymer was dissolved in 60 mL of chlorinated solvent (dichloromethane, chloroform or 1,2-dichloroethane) under argon in an oil bath heated at 40 °C. After the polymer was dissolved, in a separate flask propionic anhydride (9.52 g, 4.2 eq.) was added to 10 mL of solvent cooled to 0 °C. To the solution 6.82 g (4.0 eq.) of 98% sulfuric acid was added dropwise at such a rate that the temperature of the solution did not exceed 5 °C. The sulfonating mixture was added to the solution of the polymer and the mixture was stirred under argon at 40 °C for 3 h. Immediately after addition of propionyl sulfate solution, the reaction mixture changed color to dark brown. After 3 h, the gelatinous reaction mixture was poured into 50 mL of 2-propanol to terminate sulfonation. All volatiles were isolated on rotary evaporator under vacuum at 60 °C. The residue was transferred to a flask contains dimethyl sulfoxide (50 mL). The mixture was stirred at 150 °C under reduced pressure (400–500 mm Hg) to remove remaining solvents. After 4 h clear brown solution was obtained. The resulting solution was cooled to room temperature and poured into 300 mL of diethyl ether. The precipitated product was filtered and washed 3 times with diethyl ether and dried under vacuum at 60 °C for 12 h.
SHPPNB: 1H NMR (δ, ppm, DMSO-d6):7.75–7.38 (m., otho 2H of Ph-SO3H), 7.32–6.87 (m., 5H of Ph + meta 2H of Ph-SO3H), 3.71–3.11 (m., 1H), 2.21–0.26 (m., 10H).
SAPPNB: 1H NMR (δ, ppm, DMSO-d6):8.35–8.10 (m., otho 2H of Ph-SO3H), 7.76–7.38 (m., 2H of Ph-SO3H), 7.37–6.96 (m.,1.32H of Ph), 3.78–3.56 (m., 1H), 2.41–0.86 (m., 8H).
2.8. Synthesis of Polymer with Imidazolium and 1-methylimidazolium Cation
1.00 g of sulfonated hydrogenated poly(5-phenyl-2-norbornene) was added to 25 mL of dimethylformamide. The mixture was stirred at 120 °C 1 h. To the resulting gelatinous mass 270 mg of imidazole or 325 mg of 1-methylimidazole was added. The reaction mixture was stirred until it became homogeneous and then it was cooled to room temperature. The clear brown solution was poured into a 400 mL beaker containing 250 mL of diethyl ether. After filtration and washing with 2 × 50 mL of diethyl ether the product was dissolved in 30 mL of methanol. The solution was poured again into tetrahydrofuran. In addition, the procedure was repeated once more time. Polymer precipitate was dried under vacuum.
SPPhNB-MIm: 1H NMR (δ, ppm, DMSO-d6): 8.91 (br. s, 1 H, NCHN),7.66–7.42 (m., otho 2 H of Ph-SO3H + CH=CH of MIm), 7.28–6.89 (m., 5 H of Ph + meta 2 H of Ph-SO3MImH+), 3.82 (s, 3 H), 3.40–2.82 (m., 1 H), 2.24–0.29 (m., 10 H).
SPPhNB-Im: 1H NMR (δ, ppm, DMSO-d6): 8.58 (br. s, 1 H, NCHN),7.63–7.32 (m., otho 2 H of Ph-SO3H + CH=CH of Im), 7.30–6.78 (m., 5 H of Ph + meta 2 H of Ph-SO3MImH+), 3.38–2.73 (m., 1H), 2.21–0.27 (m., 10H).
SAPhNB-MIm: 1H NMR (δ, ppm, DMSO-d6): 8.86 (br. s, 1 H, NCHN), 8.53–8.06 (m., otho 2 H of Ph-SO3H), 7.76–6.75 (m., 5 H of Ph + meta 2 H of Ph-SO3MImH++ CH=CH of MIm as two narrow singlets at 7.57 and 7.51 ppm), 3.79 (s, 3 H, CH3 of MImH+), 3.61–3.28 (m., 1 H), 2.35–0.84 (m., 8 H).
SAPhNB-Im: 1H NMR (δ, ppm, DMSO-d6): 8.53 (br. s, 1 H, NCHN), 8.35–8.08 (m., ortho 2 H of Ph-SO3MImH+), 7.84–6.85 (m., 5 H of Ph + meta 2 H of Ph-SO3H + CH=CH of Im as a narrow singlet at 7.40 ppm), 3.72–3.55 (m., 1 H), 2.41–0.86 (m., 8 H).
2.9. Degree of Sulfonation Measurement
The sulfonation degree was determined by reverse titration method. To a 100 mL flask 100–200 mg of polymer was weighed and 10 mL of methanol, 10 mL of 0.1 M of NaOH and phenolphthalein were added. The mixture was stirred for 24 h and titrated with 0.1 M HCl. The procedure was repeated triple times and ion-exchange capacity (IEC) was calculated using average acid amount.
2.10. Membrane Preparation
The polymer membranes were prepared by casting method as follows. The casting solutions were prepared by dissolving 750 mg of ionomer and 750 mg of EMImBF4 or by 750 mg of PVDF, 375 mg of ionic polymer and 375 mg of EMImBF4 in DMF at 100 °C on a hot plate at 1400 rpm for 6 h. The mixtures were degassed under vacuum prior to casting. Resulting homogeneous solutions were cast on a Petri dishes (Ø 105 mm) and left to dry in oven at 80 °C for 12 h. Then the films were stripped from the dishes and weighted to control full removal of the solvent.
2.11. Membrane Characterization
The ionic conductivity of the polymer membranes was determined by the ac complex impedance technique over the frequency range from 0.1 Hz to 5 MHz using a P-45X potentiostat/galvanostat equipped with FRA-24M module (Electrochemical Instruments). The samples were sandwiched between symmetrical cells containing two coin-shaped steel electrodes with area of A (0.25 cm2) at the open circuit potential with a small amplitude ac voltage of 5 mV to measure membrane impedance, Z (Ω). The thickness of the sample L (cm) was measured with a micrometer. The conductivity (σ, S/cm) was then calculated from the equation: σ = L/(Z × A). Scanning electron microscopy (SEM) was performed on an TESCAN Vega 3 instrument with an accelerating voltage of 20 kV. The samples were sputter-coated with approximately 10 nm of gold before analysis. The membranes were fractured in liquid nitrogen to observe the cross-section morphology. Mechanical properties of the membranes were measured with the Universal Testing Machine (Instron 5985) at room temperature. The membranes were cut into small pieces (~50 mm × 9 mm) for testing. The exact width and thickness were measured with a caliper and micrometer, respectively, for each specimen and these values were used for strength calculations.
2.12. Preparation of Actuators
Bucky-gel actuators were prepared similar to the method previously reported [
27]. In an agate mortar, 100 mg of an ionic liquid (EMImBF4) was added to 40 mg of carbon nanotubes. The mixture was triturated and 0.5 mL of DMF was added and ground again. The operation was repeated three more times. The resulting gel was added to a solution of 60 mg of PVDF in 4 mL of DMF and the mixture was placed in an ultrasonic bath for 12 h at 50 °C. The resulting thick paste was transferred into a Petri dish (Ø 95 mm) and 12 mL of DMF was added. The mixture was evenly distributed in shape, either using an ultrasonic bath or using a spatula. The dish was placed in an oven for 3 h (100 °C). The thickness of the obtained electrode film was 15–20 μm. Samples of rectangular shape of membrane (~65 × 6 mm) and two electrodes (~60 × 5 mm) were cut. The polymer membranes were sandwiched between two bucky-gel electrodes via hot pressing at 120 °C. Then the actuators were cut to 60 × 5 mm for tip displacement and blocking force tests.
2.13. Characterization of Actuators
The actuator strip was connected to the glassy carbon electrodes over graph paper and the displacement was registered by digital camera. The constant and alternating square-wave voltages were applied to the actuator strip by a P-45X potentiostat/galvanostat equipped with FRA-24M module (Electrochemical Instruments). Constant voltage +3 V and alternating voltages ±2 V with 0.1, 0.05, 0.025 and 0.0125 Hz were used. The voltage and current were monitored simultaneously with a software ES8. The actuator strip showed a bending motion when the voltage was applied. The strain is calculated from the displacement [
38] by,
where D—displacement, L—free length, h—thickness.
The blocking force was measured using analytical balances similar to the method described elsewhere [
39]. The maximum generated stress (σ) during actuation motion was calculated by using two parameters, the maximum strain (ε
max) and the Young’s modulus (Y
el) of electrodes, according to Hooke’s law: σ = εY
el. The mechanical stress was also calculated according to the equation σ = 6Fl/bh
2. where F is the blocking force generated by the sample, l, b and h are its length, width and thickness, respectively.