Electrochemical Properties and Electromechanical Analysis of a Stacked Polyvinyl Chloride (PVC) Gel Actuator

Abstract

We investigated the electrochemical properties of and conducted an electromechanical analysis on a stacked polyvinyl chloride (PVC) gel actuator, comprising a PVC gel plasticized with dibutyl adipate (DBA) sandwiched between a metal mesh and a metal disk electrode. In this study, we examined the electrochemical impedance, displacement, and electric current responses to square-wave voltage inputs. The linear motion of PVC gel actuators with and without ionic liquid (IL) additives was analyzed in relation to the mesh size and metal composition of the mesh electrode. The displacement increased with decreasing mesh numbers, indicating that displacement increases with increasing wire diameter and space length. The linear motion of the stacked PVC gel actuators with and without IL additives depended on the metal species of the mesh electrodes. The electrochemical impedance of the stacked PVC gel actuators under DC voltage application was analyzed with and without the IL. Based on the electromechanical and electrochemical results, a deformation model was developed to describe the linear motion of stacked PVC gel actuators in response to the applied voltage. The model attributed this motion to the deformation induced by Maxwell stress in the solvent-rich layer, successfully accounting for the experimental observations.

1. Introduction

Electroactive polymer (EAP) actuators are composed of soft, flexible, lightweight materials that exhibit mechanical output in response to electrical input. These actuators have recently received considerable attention due to their various potential applications in soft robotics, benefiting from advantages such as light weight, large actuation strain, and ease of processing [1,2].

EAP actuators can be classified into two main types: ionic actuators activated by the electrical transport of ions and/or solvents and nonionic actuators activated by an electric field [3]. A major advantage of ionic EAP actuators (e.g., ionic polymer metal composites (IPMCs)) is their low driving voltage (<3 V) compared with that of nonionic EAPs or other conventional actuators [4]. In contrast, nonionic EAPs (e.g., dielectric elastomers) have remarkable actuation properties such as an extremely high strain, response speed, and work density. However, their primary disadvantage is the requirement for a high driving voltage, often several kilovolts, which poses a significant challenge for commercial applications of dielectric elastomers [5].

Among the several types of existing nonionic EAP actuators, dielectric gels such as polyvinyl chloride (PVC) gels (PVC gels with a plasticizer) have shown great potential as soft actuators because of their large strain and high response with a relatively low driving voltage (200–600 V) [6].

Hirai et al. developed PVC gel actuators and proposed a process for electrical deformation [6,7,8,9,10,11,12,13]: When voltage is applied to the electrodes attached to a PVC gel, the induced anions move to the anode while dragging the molecules of the plasticizer, such as dibutyl adipate (DBA). This induces a space charge layer on the anode side, which can be measured using the pulsed electroacoustic method [8,9,10,11]. A solvent (plasticizer)-rich (S-R) layer near the anode was formed and characterized using Fourier-transform infrared (FT-IR) spectroscopy, in situ Raman spectroscopy, and a combined shear and compression test [10,11,12]. According to Hirai et al., the creep motion of the S-R layer, driven by an applied voltage, originated from the electrical deformation of the PVC gel [6,7,8,9,10,11,12,13].

Recently, other researchers have followed Hirai’s original study and examined the mechanism of the electric deformation of PVC gels. Cheng et al. examined the electrical deformation of PVC gels using in situ Raman spectroscopy [14]. Kim et al. examined the formation and creep motion of an S-R layer on the anode side of a PVC gel using the multiphysics finite element method [15]. Asaka and Hashimoto investigated the electrochemistry of a PVC gel and developed a deformation model for its S-R layer using Maxwell stress [16]. Wang et al. measured the voltage-dependent adhesion energy between a PVC gel and metal electrode, and they proposed an electromechanical model for the electrowetting behavior of the gel [17]. Huang et al. conducted an in situ characterization of plasticizer migration and charge transfer within PVC gels using in situ Raman spectroscopy, in situ broadband impedance spectroscopy, and electromechanical measurements, and they analyzed the changes in plasticizer content and mechanical and electrical properties [18].

PVC gel actuators are soft and lightweight; exhibit excellent electromechanical properties; and have the potential for a variety of applications, including mechanical, optical, and biomedical applications [19,20]. Several actuator designs have been developed for these applications, including a bending actuator [21] and a planar actuator using a stretchable carbon electrode [22].

Hashimoto et al. developed an excellent high-power soft actuator design for wearable robots using PVC gel actuators. They developed a contraction/stretch artificial muscle based on a stacked PVC gel actuator and investigated various properties of its linear motion [23,24,25,26]. The stacked PVC gel actuator consists of PVC gel between a stainless-steel mesh acting as an anode and a stainless-steel foil acting as a cathode. When a voltage is applied, creep deformation of the PVC gel occurs between the anode and the PVC gel, causing the latter to move into the holes in the mesh. As a result, the stacked PVC gel actuator shrinks. When the voltage is turned off, the stacked PVC gel actuator returns to its original shape due to the elasticity of the PVC gel. This type of artificial muscle has been reported to exhibit a generated strain of 10–15% and a response frequency of 3–7 Hz in the voltage range of 200–600 V [25]. Based on this excellent actuator design, other applications such as fish robots [27,28], wearable medical devices [29], and active vibration absorption systems [30] have been developed.

To use PVC gel actuators in wearable robots for various applications, the driving voltage must be considerably reduced. This implies that a higher strain and stress must be generated using lower driving voltages. However, there are no reports on the improvement of PVC gel materials, for example, via the addition of ionic liquid (IL) or the optimization of the mesh electrode (material and size), to lower the voltage of linear actuators using mesh electrodes.

In a previous study [16], based on extant studies of PVC gel actuators, we investigated the electromechanical response and electrochemical properties of a bending-type PVC gel actuator between two metal disk electrodes, including stainless-steel, Cu, Zn, Al, and Fe electrodes. The bending response and electrochemical impedance depended on the electrode materials used. We proposed an electromechanical model for PVC gel actuators based on an electrochemical analysis. In a subsequent study [31], we investigated the generated strain of bending-type PVC gel actuators with four types of ILs. We found that the actuator displacements of PVC gels with the addition of 0.01 wt% trihexyltetradecyl phosphonium chloride (THTDPCl) and tetradodecylammonium chloride (TDACl) were almost two times larger than those without the addition of ILs at low applied voltages. Based on the proposed electromechanical model, the improvement in the bending response of PVC gel actuators with IL additives can be clearly described.

In this study, we conducted an in-depth examination of the electrochemical and electromechanical properties of a contraction-and-expansion stacked PVC gel actuator composed of a PVC gel sandwiched between a metal mesh and a metal disk electrode, including the electrochemical impedance, displacement, and electric current response to a square-wave voltage input. We examined the dependence of the electromechanical and electrochemical properties of PVC gel actuators on the mesh size and metal species of the mesh electrode. Moreover, we investigated the electromechanical properties of PVC gel actuators with IL. Using these data, we developed an electromechanical model for the linear motion of a stacked PVC gel actuator based on the deformation model of the S-R layer of PVC gel due to Maxwell stress proposed in a previous study [16]. This study successfully provides the first clear material design guidelines for soft, high-power contraction-and-expansion stacked PVC gel actuators for various applications.

2. Materials and Methods

2.1. Preparation of PVC Gels

In this work, the polymer used was PVC powder (n = 3700, Sigma-Aldrich, Burlington, MA, USA), the plasticizer was DBA, and the solvent was tetrahydrofuran (THF). To prepare the PVC gel, PVC was dissolved in a mixed solution of THF and DBA, and the solution was cast in a polytetrafluoroethylene laboratory dish. THF was evaporated from the casting solution at ambient temperature for more than five days, producing a soft, transparent gel. The thickness of the prepared PVC gel film was approximately 1.0 mm. The final DBA content in the prepared PVC gel was 90 wt%. Small amounts of IL (0.01 wt%) were added to the PVC gels. The IL used was Trihexyltetradecyl phosphonium chloride (THTDPCl) [31].

2.2. Measurements of Linear Motion of the PVC Gel Actuator

As shown in Figure 1, the contraction–expansion of the PVC gel actuators was measured by applying a square-wave voltage to the PVC gel (1 cm × 1 cm) sandwiched between the metal mesh (upper) and plate electrodes (bottom). As shown in Figure 1, the mesh electrode was connected to an electric ground, and a voltage was applied to the metal disk electrode. When the applied voltage exceeded 0 V, the mesh electrode acted as a cathode. In the reverse case (less than 0 V), it acted as an anode. A metal disk was placed on the metal mesh electrode for electrode contact. A glass disk was placed on the metal disk for insulation. A 20 g load was placed on top of the glass/metal disk/metal mesh/PVC gel/metal disk stacked system (in this study, we term this stacked PVC gel system a PVC gel actuator). Figures S1 and S2 show the photos of the experimental setup.

The linear motion (contraction–expansion) of the PVC gel actuator in response to the applied square-wave voltage was measured using a laser displacement meter (Keyence model LK-G82, Osaka, Japan). Square-wave voltages from 0.2 to 1.0 kV at frequencies of 0.1 Hz and 1 Hz were applied to activate the PVC gel actuator by using a high-voltage amplifier (Matsusada Precision model HGR10-10P, Osaka, Japan) with a LabVIEW control system and a PCI-6251 digital acquisition board (National Instruments, Austin, TX, USA). The measurement frequency was chosen based on its application to power-assisted wearable robots [23,24]. The electrical current responses were measured simultaneously. The electrodes were stainless-steel, copper (Cu), and aluminum (Al). A displacement of 0.01 mm corresponded to a strain of 1%, as the thickness of the PVC gel film measured approximately 1.0 mm. Figure S3 shows photos of the linear motion of the PVC gel actuator without IL under 1.0 kV square-wave voltage at a frequency of 1.0 Hz, and Videos S1 and S2 show the motion of the same actuator. Figure S4 shows a schematic of the characteristics of the metal mesh. Table S1 summarizes the metal meshes used in this study.

2.3. Electrochemical Impedance Measurements

Electrochemical impedance measurements were carried out on the PVC gel actuators using a measurement system configured the same as that used for the electromechanical measurements, as shown in Figure 1, using a Solatron 1260/1296 frequency response analyzer (Farnborough, UK) with an applied voltage at an amplitude of 1 V rms in the frequency range of 1 Hz to 1 MHz. Bias voltage was applied to the PVC gel in the range of 0 to 400 Vdc using a high-voltage DC control system.

3. Results

3.1. Linear Motion of the PVC Gel Actuator

Figure 2 shows the electric current and displacement response of the linear motion of the PVC gel actuator (#20 stainless-steel mesh) with 0.01 wt% IL. As previously reported [16,31], the electric current continuously decreased during the application of the square-wave voltage until it reached a steady value. The displacement increased as the electric current decreased until it reached saturation. As stated in our previous papers [16,31], we expected the relationship between the changes in displacement and the electric current over time, when a square-wave voltage is applied, to be closely related to the electromechanical model of the PVC gel, wherein an S-R layer forms, and the PVC gel deforms as the S-R layer deforms under Maxwell stress. Similar responses were observed for the PVC gel actuators with Cu and Al meshes (Figure S5). The electromechanical model is discussed in detail in the Discussion section.

We obtained the saturated displacement values from the displacement response of the linear motion of the PVC gel actuators. The saturated displacement was plotted with respect to the applied voltage.

Figure 3 shows plots of the displacement of the PVC gel actuators without IL and with various stainless mesh numbers against the applied square-wave voltages at frequencies of 0.1 Hz and 1 Hz. It can be observed that the displacements at applied voltages of less than 0 V were larger than those at voltages greater than 0 V.

As mentioned in the Introduction section, we expected the PVC gel actuator investigated in this study to shrink significantly when the mesh electrode acts as an anode, as PVC gel creep deformation occurs between the anode and the PVC gel, and the latter moves into the holes in the mesh, causing the actuator to contract significantly [23,24,25,26].

In this study, the mesh electrode acted as an anode when the applied voltage was less than 0 V. Creep deformation of the PVC gel occurred between the anode and PVC gel. Therefore, larger displacements were observed at applied voltages of less than 0 V. Additionally, it could be clearly observed that the displacements increased with a decreasing mesh number, as shown in Figure 3.

Figure 4 shows plots of the displacement of the PVC gel actuators with and without IL additives and with various stainless-steel mesh numbers against the applied voltages. The results shown confirm that the displacements at applied voltages of less than 0 V were larger than those at voltages greater than 0 V. Additionally, we confirmed that the displacements increased with a decreasing mesh number.

In Figure 4, it is clear that the displacements of the PVC gel actuators with the addition of 0.01 wt% THTDPCl were two-thirds to one-third smaller than those of the PVC gel actuators without additives. This result contradicts that previously reported for the bending response of the PVC gel, in which the addition of 0.01 wt% THTDPCl increased the displacement by roughly a factor of 2 [31]. The factors underlying this trend are described in the Discussion section.

Figure 5 shows plots of the displacements of the PVC gel actuators with/without IL and with various sizes of copper mesh electrodes against the applied voltage. Consistent with the cases previously shown in Figure 3 and Figure 4, the displacements were larger at applied voltages of less than 0 V, at which point creep deformation occurred between the mesh electrode and PVC gel. Similarly, as shown in Figure 3 and Figure 4, the displacements increased as the mesh number decreased. Here, we determined that the displacements of the PVC gel actuators with the addition of 0.01 wt% THTDPCl were approximately 1.5 times larger than those of the PVC gel actuators without additives. Figure S6 shows plots of the displacements of the PVC gel actuators (Al mesh: PA18-323) with/without THTDPCl against the applied voltage. In this case, the displacements of the PVC gel actuators with the addition of 0.01 wt% THTDPCl were approximately 1.5 times larger than those of the PVC gel actuators without additives. These results are in agreement with those of the bending response of the PVC gels reported in the previous study. The details of the effects of these ILs, including the differences from the stainless-steel electrodes, are discussed later, along with the impedance results shown below.

3.2. Electrochemical Impedance Measurement Results

Figure 6 shows the impedance spectra of the PVC gel actuator with a stainless-steel mesh. A semicircle was observed in the complex plane plot in the absence of a bias voltage. In contrast, two semicircles were observed under bias voltages. Similar results were observed in the impedance spectra of the PVC gel actuator with a Cu mesh, which are shown in Figure 7, and in those of the PVC gel actuator with an Al mesh, which are shown in Figure S7. Hence, we analyzed the results using the equivalent circuit of a previous study, as shown in Figure S8 [16]. The impedance spectra shown as semicircles in the complex plane plot can be represented by one parallel capacitance and resistance element with a series of resistance elements, as shown in Figure S8a. The impedance spectra of the two semicircles can be represented by a series of parallel capacitance and resistance elements with a series resistance element, as shown in Figure S8b.

Plots of the impedance parameters according to the applied bias voltage are shown in Figure 8 for the PVC gel actuators with a stainless-steel mesh, in Figure 9 for those with a Cu mesh, and in Figure S9 for those with an Al mesh.

Capacitance C₁ was independent of the applied bias voltage and exhibited a value of 8–18 pF. C₁ corresponds to the capacitance of the bulk layer. Capacitance C₂ was much larger than C₁ and was considered to correspond to the capacitance of the S-R layer based on the previous analysis [16,31]. An increase in capacitance C₂ due to the addition of IL was observed for all electrodes. This was considered to be due to an increase in the adsorption capacitance [32], and this effect was considered to increase the amount of displacement [16,31].

Although the stainless-steel mesh electrode also showed an increase in C₂ due to the addition of IL, as discussed later, the resistance R₂ of the S-R layer was minimized, which was presumed to reduce the amount of deformation. This point is discussed in detail in the Discussion section. In all cases, C₂ was voltage-dependent and decreased as the absolute value of the voltage increased. This may be because the thickness of the S-R layer increased as the voltage increased.

Figure 10 and Figure 11 show plots of the applied square-wave voltage against the peak current of the PVC gel actuators with stainless-steel and copper electrodes, respectively. In these figures, it can be seen that the conductivity of both electrodes was about 100 μA/1 kV without IL; however, with the addition of IL, the conductivity of the stainless-steel electrode was 300 μA/1 kV at 0.1 Hz and 400 μA/1 kV at 1 Hz, while that of the copper electrode was 200 μA/1 kV at 0.1 Hz and 300 μA/1 kV at 1 Hz. In other words, the addition of IL increased the conductivity of both electrode systems, but the PVC gel actuators with the copper electrode exhibited a higher resistance than those with a stainless-steel electrode. Furthermore, as shown in Figures S10 and S11, there was almost no difference in the voltage–current characteristics between the Cu and Al electrode systems.

4. Discussion

In this study, we investigated the electromechanical response of a contraction-and-expansion-type PVC gel actuator composed of PVC gel sandwiched between a metal mesh and a metal disk electrode. We studied the dependence of the linear motion of PVC gel actuators on the mesh size and metal species of the mesh electrode. In our experimental system, the mesh electrode acted as the anode when the applied voltage was less than 0 V. Creep deformation of the PVC gel occurred between the anode and the PVC gel, which moved into the mesh holes. Therefore, we observed larger displacements at an applied voltage of less than 0 V.

Here, we discuss why the displacement increased with a decreasing mesh number. In our previous paper, we discussed the electromechanical model of a PVC gel actuator based on the deformation of the S-R layer due to Maxwell stress at the planar electrodes [16]. Similarly, here, we discuss the electromechanical model of a PVC gel actuator based on the deformation of the S-R layer on the surface of the anode mesh electrode due to Maxwell stress. Figure 12 shows a schematic of the formation process of the electric field-induced S-R layer structure based on the migration of plasticizer molecules in association with the electric current on the surface of the metal mesh wire used as the anode. Here, we focus on the movement of the PVC gel on a single mesh wire when voltage is applied. The S-R layer is considered to be a layer formed on a single mesh wire with a cross-section in the shape of an annular sector, as shown in Figure 12. When a voltage is applied between the PVC gel and mesh wire, the thickness of the S-R layer decreases, and it deforms in the circumferential direction, extending and attaching to the surface of the mesh wire, as shown in Figure 12. In the longitudinal direction of the wire, it does not stretch or shrink because it is constrained by the mesh space and maintains its length.

However, the capacitance C₂ on the negative voltage side in Figure 8c and Figure 9c may be due to the S-R layer on the mesh electrode shown in Figure 12. As shown in Figure 8c and Figure 9c, there are differences in these values with and without the addition of ionic liquid, but there are no significant differences in the mesh sizes. Considering that the capacitance C₂ of the S-R layer is almost constant with the mesh size, the capacitance of one mesh wire C_wire is proportional to the inverse of the square of the mesh number, as shown in Figure S12. That is, the smaller the mesh number, the larger the capacitance of the S-R layer per mesh wire.

However, for the same electrode material, the thickness of the S-R layer is considered to be almost equal; thus, it can be inferred from the value of C₂ that the size of the circumferential direction of the S-R layer covering one wire becomes larger. This supports the result that a smaller mesh number results in a larger displacement.

In Figure 3, the displacement of mesh size #500 is larger than that of #400, which does not follow the above discussion. As shown in Table S1, the difference between the wire diameter and space length of #400 and #500 is small, suggesting that the size of the S-R layer on the wire surface is affected by the mesh wire shape rather than the mesh density.

Here, we discuss the effects of the addition of IL on the linear motion of the PVC gel actuator in response to the applied voltages, including the differences in the Cu and Al electrodes from the stainless-steel electrodes. Without the addition of IL, we observed a maximum displacement of about 0.08 mm for the PVC gel actuators with stainless-steel, copper, and aluminum electrodes, which corresponded to a strain of about 8%. In contrast, when IL was added, the PVC gel actuator with the stainless-steel electrode showed a smaller strain value of approximately 6% (0.06 mm), whereas those with copper and aluminum electrodes showed a larger strain value of approximately 10% (0.1 mm). These strain values were compared to those reported in the literature [25].

As previously mentioned, C₂ was much larger than C₁ and was considered to correspond to the capacitance of the S-R layer based on the previous analysis. An increase in capacitance C₂ due to the addition of IL was observed for all electrodes. This was considered to be due to an increase in the adsorption capacitance [32], and this effect was thought to increase the displacement of the PVC gel actuators with Cu and Al electrodes. These experimental results can be explained using an electromechanical model based on Maxwell stress [31].

Although the stainless-steel mesh electrode also showed an increase in C₂ due to the addition of IL, the resistance R₂ of the S-R layer became very small, and this effect was expected to reduce the amount of deformation.

For instance, R₁ was in the range of 15 MΩ to 20 MΩ for the stainless-steel electrode, and it reduced to approximately 1 MΩ following the addition of IL. Furthermore, the copper and aluminum electrodes had an R₁ of 10 MΩ, and it reduced in the range of 1–2 MΩ, which is not a large difference. However, following the addition of IL, the R₂ value of the stainless electrode reduced from approximately 100 MΩ to 1 MΩ, while for the copper and aluminum electrodes, the R₂ values were one order of magnitude higher, being reduced from around 1000 MΩ to 10 MΩ. Hence, the R₂/R₁ ratio was approximately 10 for the PVC gels with Cu and Al electrodes with the addition of IL, whereas it was approximately 1 to 1/2 for those with stainless-steel electrodes with the addition of IL. This implies that the increase in C₂ due to the addition of IL was the cause of the increase in displacement in the Cu and Al mesh electrode systems, as the voltage was mostly applied to the S-R layer. However, the decrease in displacement in the stainless-steel mesh electrode due to the addition of IL was caused by the applied voltage not being sufficiently applied to the S-R layer because of the small R₂.

The physicochemical factors that caused the R₂ of the PVC gel with a stainless-steel mesh to be smaller than that of the Cu and Al mesh systems are currently not clear. These metals form a surface oxide layer via anodic polarization in ionic liquids, and it is presumed that the interaction between the surface oxide layer and the ionic liquid in the S-R layer is an important factor, but more detailed studies are needed.

This result did not depend on whether the electrode had a mesh or flat shape. In other words, even with a planar stainless-steel electrode, the addition of ionic liquid resulted in smaller displacement, which was thought to be due to the reduction in R₂. In the previous paper, it was found that bending deformation increased with ionic liquid addition, even for stainless-steel electrodes, and this was thought to be due to the effect of increased C₂, as the decrease in R₂ was not so large that it affected the electromechanical response [31]. This may be due to the different surface conditions of the stainless-steel electrodes used, which resulted in different contact conditions with the gel. In other words, it is assumed that the contact area between the electrode and PVC gel differs depending on the pressure applied when the gel is sandwiched between the electrodes due to the microroughness of the electrode surface and that R₂ is smaller when greater pressure is applied in the present case. However, these details are under further investigation.

We showed above that the change in electromechanical phenomena due to IL addition can be demonstrated using an electrochemical impedance model. The mechanism depends on the contact state at the interface between the electrode and the gel, but the details are under further investigation.

5. Conclusions

We conducted an in-depth examination of the electrochemical and electromechanical properties of a contraction-and-expansion stacked polyvinyl chloride (PVC) gel actuator composed of a PVC gel sandwiched between a metal mesh and a metal disk electrode, including the electrochemical impedance, displacement, and electric current response to a square-wave voltage input. We examined the dependence of the linear motion of the PVC gel actuators on the mesh size and metal species of the mesh electrode, including stainless steel, copper, and aluminum. In our experimental system, the mesh electrode acted as an anode when the applied voltage was less than 0 V. Creep deformation of the PVC gel occurred between the anode and PVC gel, which moved into the mesh holes. Therefore, we observed larger displacements at an applied voltage of less than 0 V.

The displacement increased with a decreasing mesh number, indicating that displacement increases with an increasing wire diameter and space length. Without IL additives, the maximum displacement observed was approximately 0.08 mm for the PVC gel actuators with stainless-steel, copper, and aluminum electrodes, corresponding to a strain of approximately 8%. In contrast, with IL additives, the PVC gel actuator with a stainless-steel electrode exhibited a reduced strain value of approximately 6%, whereas the PVC gel actuators with copper and aluminum electrodes displayed an increased strain of approximately 10%. The electrochemical impedance of the stacked PVC gel actuators under DC voltage application was analyzed with and without IL. Based on the electromechanical and electrochemical results, a deformation model was developed to describe the linear motion of the stacked PVC gel actuators in response to the applied voltage. The model attributed this motion to deformation induced by Maxwell stress in the solvent-rich layer, successfully accounting for the experimental observations. This study successfully provides the first clear material design guidelines for soft, high-power contraction-and-expansion stacked PVC gel actuators for various applications.