Behaviors of Polypyrrole Soft Actuators in LiTFSI or NaCl Electrolyte Solutions Containing Methanol

Organic soft linear actuators were fabricated using galvanostatic electropolymerization of the polypyrrole (PPy) thin film using a methyl benzoate electrolyte solution of N,N-Diethyl-N-methyl-N-(2-methoxyethyl) ammonium bis (trifluoromethanesulfonyl) imide. The electrochemical deformation behaviors of the PPy actuators were investigated in aqueous solutions of an electrolyte, lithium bis (trifluoromethanesulphonyl) imide (LiTFSI) or sodium chloride (NaCl), containing different concentrations of methanol. The actuating strain of approximately 9% was achieved when the actuator was driven by a potential between –1 and 1 V with the potential sweep rate of 10 mV/s corresponding to 0.0025 Hz in the LiTFSI electrolyte containing 40% to 50% of methanol under a load stress of 0.3 MPa. However, the PPy actuator could not catch up with the higher frequency. On the other hand, the PPy actuator caught up with the potential sweep up to 0.1 Hz in the NaCl solutions with a methanol concentration between 40% and 60% with the expense of the actuating strain to approximately 1%.

In the research of PPy actuators, not only larger electrochemical strain and stress but their operation speed is another important issue, and a PPy bending actuator was reported to operate at the frequency up to 90 Hz [15].However, the force generated by the bending actuator was much smaller compared to the PPy linear actuators [16,17].Similar PPy linear actuators were also reported to operate with at the frequency of 30 Hz [18,19].On the other hand, the performance of PPy actuators were reported to strongly dependent on the different kind of cations in the electrolyte solutions during actuation [20,21].Kaneto et al. made systematic researches using different kinds of electrolytes such LiCl, NaCl, etc. for [22][23][24].Hara et al. also reported that their TFSI-doped porous PPy films exhibited increased actuating strains when their aqueous lithium bis (trifluoromethansulfonyl)imide (LiTFSI) electrolyte solutions contained propylene carbonate [25].They attributed those effects to the swelling of the PPy film caused by the penetration of propylene carbonate.The swelled PPy film could more easily pass TFSI anions.Hoshino et al. also found that the PPy films showed notable increase of actuating strains when they were functioned in LiTFSI solutions containing 2-propanol [26] or methanol [27].However, the PPy actuators in the electrolyte solutions showed notable electrochemical creep after repeated actuation processes.
In this paper, we report on increased actuating strain of PPy actuators but with minimal increase of electrochemical creep in a LiTFSI electrolyte solution contain-

Experiments
The polymerization of PPy films was carried out using a computer-controlled potentio-galvanostat.A counter electrode (Ti), a reference electrode (Ag/AgCl), and a working electrode (Ti) were immersed into methyl benzoate solutions of 0.25 M pyrrole and 0.2 M N,N-diethyl-Nmethyl-N-(2-methoxyethyl) ammonium bis (trifluoromethanesulfonyl) imide, and the potential voltage was controlled to keep a constant current of 0.2 mA/cm 2 for 4 h at 20˚C between the counter electrode and the working electrode.These chemicals were purchased from Sigma-Aldrich Inc.
The PPy actuator was used as the working electrode in the 1 M LiTFSI aqueous electrolyte solutions or in the 1 M NaCl aqueous solutions with different methanol concentrations of 0% -100%.The PPy actuator exhibited expansion and contraction motions under the alternating potential with the triangular wave shape applied between the PPy actuator and the counter electrode.The peak values of the potential voltage were −1 and +1 V, and the potential sweep rates were 10 -400 mV/s that correspond to the frequencies between 0.0025 and 1 Hz.The extension and contraction of the PPy actuator was measured by monitoring the displacement of the weight position using a laser displacement sensor as described in the previous publications [26][27][28][29].A load stress of 0.3 MPa was applied on the PPy actuator.

Results and Discussion
Figure 1 shows a typical measurement result for the relationship between the measured strain and time during repeated actuations at 0.0025 Hz.This measurement was performed in the LiTFSI electrolyte solution containing 40% of methanol.The averaged strain continuously shifts to the positive strain direction due to electrochemical creep.Here, the electrochemical strain is defined as the change of the averaged strain as shown the dotted line in Figure 1.The difference between the peak values and the bottom values of the strain is defined as the actuating strain as indicated by the arrow in Figure 1.
Figures 2(a) and (b) show the comparison of the strain as a function of time under the repeated potential voltage change for actuators that were functioned in the LiTFSI solutions containing various concentrations of methanol.The actuators in the electrolyte solution with 20% and 80% methanol exhibited increased actuating strain compared to that in the electrolyte solution without methanol, and the actuating strains stayed at the similar level after 10 cycles of actuations as shown in Figure 2(a).The electrochemical creep (continuous back ground change) gradually increased and approached to approximately 4%.On the other hand, when the methanol concentration was increased from 60% to 100%, the actuating strains of these actuators rapidly decreased after the repeated actuations as shown in Figure 2(b).On the other hand, the electrochemical creep seemed to be significantly smaller.
Figure 3 shows the relationships between the actuating strain and the methanol concentration after 10 cycles of actuations.The actuating strain showed the maximum value of 9% for the actuators functioned in the electrolyte solutions containing 40% or 50% of methanol.The actuating strain for the actuators functioned in the electrolyte solutions containing more than 60% of methanol exhibited rapid decreases.
Figure 4 shows the relationship between the electrochemical creep and the methanol concentration, and the electrochemical creep continuously decreased as a function of the methanol above 40%.The largest current with a large hysteresis occurred for the actuator functioned in the electrolyte solution containing 40% methanol, and the hysteresis curves continuously decreased as the methanol concentration increased above 60%.This may explain the decreased actuating strain above the methanol concentration of 60%.
However, consistent explanations for these behaviors of the PPy actuators functioned in the LiTFSI solutions containing various concentration of methanol have not been obtained.Table 1 compares the surface tension and viscosity of pure water and methanol at 20˚C.The data   were taken from the web page of the National Institute of Standards and Technology (NIST).The surface tensions of methanol are 22.6 dyn/cm, which are nearly 30% that of water.Therefore, when the PPy actuator is positively biased, TFSI anions along with methanol molecules might more easily penetrate into the porous structure of PPy.Thus, the increased expansion was observed in the LiTFSI electrolyte solutions with 20% -50% of methanol.The TFSI ions diffused into the PPy porous structure in the positive potential region could be disturbed to escape from the PPy structure due to viscosity of the electrolyte solution in the negative potential region.In contrast, the reduced disturbance for the out diffusion of the TFSI anions from the PPy film was expected because the viscosity of the methanol was smaller than that of water.However, the decreased actuating strains of the actuators in the electrolyte solutions containing more than 60% of methanol after the repeated actuations can not be fully explained by the discussions above.
Higashi et al. recently reported the increase of Young's modulus after repeated actuation in aqueous LiTFSI electrolyte solutions containing 0% (water), and 20% of methanol as summarized in Table 2 [29].Young's modulus increased from 0.13 to 0.45 GPa after 10 cycles of actuations in the electrolyte solution without methanol, and it increased to 4.15 GPa after 10 cycles of actuations in the electrolyte solution containing 20% of methanol.In addition, notable reduction in the tensile strength and the strains at break of the PPy films actuated in the electrolyte solutions containing methanol were also observed.Although the introduction of methanol in the LiTFSI electrolyte solutions improves the actuating strain of the PPy actuators, the hardening of the PPy actuators after the repeated actuations could be more significant when the concentration of methanol was larger than 60%, which may explain the decreased actuating strains after the repeated actuations in the electrolyte solutions containing methanol more than 60%.methanol did not function at 0.075 Hz.However, the actuator in the solutions containing 20%, or more methanol functioned at 0.075 Hz.In addition, the PPy actuator in the solutions containing 0% and 20% metha-nol did not catch up with the 0.1 Hz actuation.This phenomenon was attributed to the fact that NaCl was not sufficiently ionized.The actuator caught up with 0.1 Hz actuation in the solutions containing 40% and 60% methanol.
The electrochemical strains of the solutions containing 20% methanol became the largest.These results attribute this phenomenon to the low viscosity and low surface tension of methanol comparison with the water.Therefore dopant might diffuse into the PPy film smoothly.
Figure 7 shows 1) relationships between the strain and time of the PPy actuators in NaCl electrolyte solution at the potential sweep rate of 300 mV/s corresponding 0.075 Hz, and 2) relationships between the strain and time of PPy actuators in NaCl electrolyte solution at the potential sweep rate of 400 mV/s corresponding 0.1 Hz.Improvement of the strain was observed by mixing methanol in the NaCl solution.The actuating strain was increased by mixing 20% or more of methanol.The actuating strain was the largest by mixing 20% methanol concentration at 0.075 Hz.On the other hand, the actuating strain was larger for the NaCl solution containing 40% -60% at 0.1 Hz.Thus, the most suitable NaCl concentration depends on the actuation frequency.The actuating strain in each methanol concentration at each operating speed was summarized in Table 3.

Summary
Soft actuators were fabricated using galvanostatic electropolymerization of a PPy thin film using a methyl benzoate electrolyte solution of N,N-Diethyl-N-methyl-N-(2-methoxyethyl) ammonium bis (trifluorome-thanesulfonyl) imide.The electrochemical deformation behaviors of the PPy actuator were investigated in aqueous solutions of an electrolyte, LiTFSI NaCl, containing different concentrations of methanol.The actuating strain of approximately 9% was achieved when the actuator was driven by a potential between -1 and 1 V with the potential sweep rate of 10 mV/s corresponding to 0.0025 Hz in the LiTFSI electrolyte containing 40% to 50% of metha-  nol under a load stress of 0.3 MPa.However, the PPy actuator could not catch up with the higher frequency.On the other hand, the PPy actuator caught up with the potential sweep up to 0.1 Hz in the NaCl solutions with a methanol concentration between 40% and 60% with the expense of the actuating strain to approximately 1%.

Figure 1 .
Figure 1.Relationship between strain and time for PPy actuator in LiTFSI electrolyte solution containing 40% methanol during electrochemical actuations at 0.0025 Hz.

Figure 2 .
Figure 2. Relationship between strain and time during electrochemical actuations of PPy actuators in LiTFSI electrolyte solutions with different methanol concentrations at 0.0025 Hz.

Figures 5 (
a) and (b) compares the corresponding cyclic voltammograms of the PPy actuators functioned in the electrolyte solutions containing various concentrations of methanol.The current in the positive potential voltage range corresponds to the motion of large TFSI − anions, and the current in the negative potential voltage range corresponds to the motion of small sized Li + cations.Thus, a large volume change occurs in the positive potential voltage range.The largest hysteresis of the PPy actuator driven in the electrolyte solution with methanol implicates the enhancement of the TFSI ionic motions into or outwards the PPy actuator.

Figure 3 .
Figure 3. Change of actuating strain of PPy actuators in LiTFSI electrolyte solutions with different methanol concentrations as measured at 0.0025 Hz.

Figure 4 .Figure 5 .
Figure 4. Change of electrochemical creep of PPy actuators in LiTFSI electrolyte solutions with different methanol concentrations as measured at 0.0025 Hz.

Figure 6 (Figure 6 .
Figure 6.(a) Comparison of relationships between the strain and time of PPy actuators in LiTFSI electrolyte solution and in NaCl electrolyte solution at the potential sweep rate of 200 mV/s corresponding 0.05 Hz, and (b) Relationship between the strain and time during electrochemical actuations of PPy actuators in NaCl electrolyte solutions with different methanol concentrations at 0.05 Hz.The initial strain was adjusted for better view of the plots.

7 *Figure 7 .
Figure 7. (a) Relationships between the strain and time of PPy actuators in NaCl electrolyte solutions at the potential sweep rate of 300 mV/s corresponding 0.075 Hz, and (b) relationships between the strain and time of PPy actuators in NaCl electrolyte solution at the potential sweep rate of 400 mV/s corresponding 0.1 Hz.The initial strain was adjusted for better view of the plots.