**3. Results and discussion**

318 Smart Actuation and Sensing Systems – Recent Advances and Future Challenges

The polymer solution (18 wt%) was poured into a 2.5 mL syringe. A potential of 10 kV was applied by connecting the power supply (GT80 GREEN TECHNO) to the syringe tip (Figure 5). In order to introduce anisotropic structure into the nanofiber gel, the 1.0 mL of the polymer solution in the syringe was sprayed at a flow rate of 2.0 mL/hour (sprayed for 30 minutes), and then the flow rate was changed to 1.0 mL/hour (sprayed for 60 minutes). The fibers were collected on the grounded glass substrate as a collector. The distance between the collector and the syringe tip was 15 cm. The temperature and humidity were 25 °C and 70%, respectively. After the electrospinning, the obtained sheet, with a thickness of about

*2.4.2. Electrospinning* 

200 μm, was dried overnight at 50 °C.

**Figure 5.** Schematic illustration of electrospinning set-up.

*2.4.2. Measurement of motion of the nanofiber gel actuator* 

controlled at 25 °C by utilizing the water bath equipment.

The open continuously stirred tank reactor (CSTR) (40 mL) was designed using an acrylic cell with a water jacket in order to control the solution temperature in the cell. Potassium bromated (0.26 mol/L), sodium sulfite (0.3 mol/L), potassium ferrocyanide (0.08 mol/L), and sulfuric acid (0.04 mol/L), solutions were pumped into the reactor at a flow rate of 50 mL/hour. The pH changes in the reactor were monitored continuously by utilizing a pH meter (F-55 HORIBA) held in the reactor, and its electronic output was directly recorded by a computer. The nanofiber gel (length 15 mm, width 3 mm) was set at the bottom of the water jacket. One end of the gel strip was sandwiched in the incision of the silicone rubber. Shape changes of the gel strip were recorded by a fixed microscope (Fortissimo Corp. WAT-250D) and a video recorder (Victor Corp. SR-DVM700). The temperature in the reactor was

#### **3.1. Self-oscillating behavior of poly(VP-***co***-Ru(bpy)3) solution**

The measurement of the solubility for the poly(VP-*co*-Ru(bpy)3) solution in the reduced and oxidized state was conducted by changing the temperature from 10 to 50 °C. As shown in Figure 6, there are no the lower critical solution temperature (LCST) for the polymer solutions in the reduced and oxidized state. The polymer solutions in the reduced and oxidized state have different transmittance values. That is, this result indicates that the polymer solution has different solubility in the reduced and oxidized state, respectively. The solubility of the polymer solution in the reduced state is higher than that in the oxidized state. Figure 7 shows self-oscillating behaviors of the poly(VP-*co*-Ru(bpy)3) solution in the different concentrations of malonic acid ([MA] = 0.04, 0.05, 0.06, 0.07, 0.08 and 0.09 M) under the fixed concentration of sodium bromate and nitric acid ([NaBrO3] = 0.3 M and [HNO3] = 0.3 M). As shown in Figure 7, the base line of the transmittance self-oscillation gradually decreased with time in all malonic acid concentrations. The damping behavior originates from the change in the ionic strength of the polymer solution when the transmittance measurement starts [47-49]. In order to cause the BZ reaction in the polymer solution, the self-oscillating polymer solution and the other solution of the BZ substrates are mixed just before the transmittance measurements. In general, the solubility of the polymer chain is significantly affected by the ionic strength of the solution. Therefore, when the ionic strength increased at the start point of the self-oscillation, the solubility of the polymer chain decreased. In the solution condition of this study, the ionic strength of the polymer solution was very high because the BZ reaction required a significant high concentration of the BZ substrates. Therefore, the damping behaviors occurred from the start point of the selfoscillation. In addition, as shown in Figure 7, the width of the waveform increased with decreasing the concentration of malonic acid. Basically, the width of the waveform of the transmittance self-oscillation depends on the rate of the BZ reaction because the selfoscillation was induced by the BZ reaction. As the concentration of the BZ substrates decreased, the rate of the BZ reaction decreased due to decrease in the collision rate among the BZ substrates. Therefore, the width of the waveform increased with decreasing the concentration of the BZ substrates. This tendency was observed in the transmittance selfoscillation of the AMPS-containing polymer solutions [48].

Figure 8 shows the transmittance self-oscillations of the novel polymer solution in the different concentration of sodium bromate ([NaBrO3] = 0.1, 0.2, 0.3 and 0.4 M) at 14 °C under the fixed concentration of malonic acid and nitric acid ([MA] = 0.1M and [HNO3] = 0.3 M). As shown in Figure 8, the amplitude of the self-oscillation gradually decreased with time in the same manner as in Figure 7. Moreover, the width of the waveform decreased with the increase in the concentration of sodium bromate due to the increase in the reaction rate of the BZ reaction. In addition, as shown in Figure 8, the amplitudes of the transmittance selfoscillations were hardly affected by the initial concentration of sodium bromate. In additon, Figure 9 showed the amplitude of the transmittance self-oscillation for the polymer solution under the different concentrations of the BZ substrates. As shown in Figure 9, all the BZ

substrate concentrations hardly influence the amplitude of the transmittance self-oscillation. That is, the amplitude values were almost the same in all BZ substrate conditions. In our previous investigations, we studied the effect of the concentration of the BZ substrates on the waveform of the transmittance self-oscillation for the AMPS-containing polymer solution. As a result, we clarified that the amplitude of the self-oscillation is significantly affected by the initial concentration of the BZ substrates [47-49]. This is because the AMPScontaining polymer chain caused damping, that is, the amplitude of the self-oscillation gradually decreased with time. The damping behavior of the polymer solution originates from the change in the size of the polymer aggregation with time. In the case of the NIPAAm-based polymer chains, the reduced Ru moiety in the polymer chain strongly interacts with the other reduced Ru one. Once the reduced Ru moiety strongly interacts with the other Ru one, the interaction hardly dissociates [48]. In the BZ reaction, the time in the reduced state is much longer than in the oxidized state. Therefore, the hydrophobic Ru(bpy)32+ moiety in the polymer chain dominantly behaved for the determination of the polymer aggregation state in the self-oscillating behavior. For this reason, the mole fraction of the Ru(bpy)32+ moiety in the polymer chain significantly affect the waveform of the transmittance self-oscillation. This influence can be explained by the overall process of the BZ reaction based on the Field-koros-Noyes (FKN) mechanism [33-36, 48]. On the other hand, in the case of the VP-based polymer chain, the bipyridine ligands interacted with the VP based main-chain. The strength of the interaction of the bipyridine ligands in the oxidized state is higher than in the reduced state. That is, the polymer aggregation increased in the oxidized state. However, the time in the oxidized state is much shorter than in the reduced state. Therefore, the size of the polymer aggregation changed very slowly compared to the AMPS-containing polymer solution. Consequently, the degree of the damping for the Vp-based polymer solution is considerably small. Therefore, the amplitude is hardly affected by the initial concentration of the BZ substrates.

**Figure 6.** Temperature dependence of transmittance for poly(VP-*co*-Ru(bpy)3) solutions under the different conditions of reduced Ru(II) (in Ce(III) solution) and Ru(III) (in Ce(IV) solution) states.

**Figure 7.** Oscillating profiles of transmittance at 14 °C for 0.5 wt% poly(VP-*co*-Ru(bpy)3) solution in the fixed nitric acid and sodium bromate conditions ([HNO3] = 0.3 M and [NaBrO3] = 0.3 M): (**A**) [MA] = 0.04 M, (**B**) [MA] = 0.05 M, (**C**) [MA] = 0.06 M, (**D**) [MA] = 0.07 M, (**E**) [MA] = 0.08 M, (**F**) [MA] = 0.09 M.

#### **3.2. Self-oscillating behavior of poly(VP-***co***-Ru(bpy)3) gel**

320 Smart Actuation and Sensing Systems – Recent Advances and Future Challenges

is hardly affected by the initial concentration of the BZ substrates.

**Figure 6.** Temperature dependence of transmittance for poly(VP-*co*-Ru(bpy)3) solutions under the different conditions of reduced Ru(II) (in Ce(III) solution) and Ru(III) (in Ce(IV) solution) states.

substrate concentrations hardly influence the amplitude of the transmittance self-oscillation. That is, the amplitude values were almost the same in all BZ substrate conditions. In our previous investigations, we studied the effect of the concentration of the BZ substrates on the waveform of the transmittance self-oscillation for the AMPS-containing polymer solution. As a result, we clarified that the amplitude of the self-oscillation is significantly affected by the initial concentration of the BZ substrates [47-49]. This is because the AMPScontaining polymer chain caused damping, that is, the amplitude of the self-oscillation gradually decreased with time. The damping behavior of the polymer solution originates from the change in the size of the polymer aggregation with time. In the case of the NIPAAm-based polymer chains, the reduced Ru moiety in the polymer chain strongly interacts with the other reduced Ru one. Once the reduced Ru moiety strongly interacts with the other Ru one, the interaction hardly dissociates [48]. In the BZ reaction, the time in the reduced state is much longer than in the oxidized state. Therefore, the hydrophobic Ru(bpy)32+ moiety in the polymer chain dominantly behaved for the determination of the polymer aggregation state in the self-oscillating behavior. For this reason, the mole fraction of the Ru(bpy)32+ moiety in the polymer chain significantly affect the waveform of the transmittance self-oscillation. This influence can be explained by the overall process of the BZ reaction based on the Field-koros-Noyes (FKN) mechanism [33-36, 48]. On the other hand, in the case of the VP-based polymer chain, the bipyridine ligands interacted with the VP based main-chain. The strength of the interaction of the bipyridine ligands in the oxidized state is higher than in the reduced state. That is, the polymer aggregation increased in the oxidized state. However, the time in the oxidized state is much shorter than in the reduced state. Therefore, the size of the polymer aggregation changed very slowly compared to the AMPS-containing polymer solution. Consequently, the degree of the damping for the Vp-based polymer solution is considerably small. Therefore, the amplitude

> Figure 10 shows the equilibrium swelling behaviors of the poly(VP-*co*-Ru(bpy)3) gels in the Ce(III) and Ce(IV) solutions under the same acidic condition. In the Ce(III) solution, the gel kept a tinge of orange, which indicated that the copolymerized Ru(bpy)3 moiety in the gel was in the reduced state. On the other hand, in the Ce(IV) solution, the gel quickly turned from orange to green, which showed the Ru(bpy)3 moiety in the gel had changed the reduced state to the oxidized state. In the oxidized state, the equilibrium volume of the gel was larger than that in the reduced state in all temperature condition. This is because the solubility of the Ru(bpy)3 moiety has significantly different properties in the oxidized and

**Figure 8.** Oscillating profiles of transmittance at 14 °C for 0.5 wt% poly(VP-*co*-Ru(bpy)3) solution in fixed nitric acid and malonic acid conditions ([HNO3] = 0.3 M, [MA] = 0.1 M) (**A**) [NaBrO3] = 0.1 M, (**B**) [NaBrO3] = 0.2 M, (**C**) [NaBrO3] = 0.3 M, (**D**) [NaBrO3] = 0.4 M.

**Figure 9.** Dependence of amplitude of transmittance self-oscillation for polymer solution at 14 °C in the change in one BZ substrate under fixed concentrations of the other two BZ substrates: **MA** ([MA] = 0.04, 0.05, 0.06, 0.07, 0.08 and 0.09 M, fixed [NaBrO3] = 0.3 M and [HNO3] = 0.3 M); **NaBrO3** ([NaBrO3] = 0.1, 0.2, 0.3, 0.4, 0.5 and 0.6 M, fixed [MA] = 0.1 M and [HNO3] = 0.3 M); **HNO3** ([HNO3] = 0.1, 0.2, 0.3, 0.4, 0.5 and 0.6 M, fixed [NaBrO3] = 0.1 M and [HNO3] = 0.3 M).

reduced state. The reduced Ru(bpy)3 moiety in the gel has an extreme hydrophobic property in the VP-based polymer gel. This property is attributed to the conformation of the bipyridine ligands surrounding the Ru ion, which induces the deswelling behavior. That is, in the VP-based gel, the bipyridine ligands surrounding the Ru ion exert a greater influence on the solubility of the polymer chain in the reduced state as compared with the ionization effect of the Ru ion.[47-52] On the contrary, the oxidized Ru(bpy)3 parts in the gel has a great hydrophilic property. The driving force of the swelling-deswelling self-oscillation is originated in the different solubility of the Ru(bpy)3 moiety in the reduced and oxidized states as shown in Figure 10. In the reduced and oxidized state, there is no observation of the volume phase transition because of the PVP main chain of the gel without LCST.

322 Smart Actuation and Sensing Systems – Recent Advances and Future Challenges

**Figure 8.** Oscillating profiles of transmittance at 14 °C for 0.5 wt% poly(VP-*co*-Ru(bpy)3) solution in fixed nitric acid and malonic acid conditions ([HNO3] = 0.3 M, [MA] = 0.1 M) (**A**) [NaBrO3] = 0.1 M, (**B**)

**Figure 9.** Dependence of amplitude of transmittance self-oscillation for polymer solution at 14 °C in the change in one BZ substrate under fixed concentrations of the other two BZ substrates: **MA** ([MA] = 0.04, 0.05, 0.06, 0.07, 0.08 and 0.09 M, fixed [NaBrO3] = 0.3 M and [HNO3] = 0.3 M); **NaBrO3** ([NaBrO3] = 0.1, 0.2, 0.3, 0.4, 0.5 and 0.6 M, fixed [MA] = 0.1 M and [HNO3] = 0.3 M); **HNO3** ([HNO3] = 0.1, 0.2, 0.3, 0.4, 0.5

[NaBrO3] = 0.2 M, (**C**) [NaBrO3] = 0.3 M, (**D**) [NaBrO3] = 0.4 M.

and 0.6 M, fixed [NaBrO3] = 0.1 M and [HNO3] = 0.3 M).

**Figure 10.** Equilibrium swelling ratio of poly(VP-*co*-Ru(bpy)3) gel in cerium sulfate solutions as a function of temperature. (●) [Ce2(SO4)3] = 0.001M and [HNO3] = 0.3M; (○) [Ce(SO4)2] = 0.001M and [HNO3] = 0.3M. The relative length is defined as the ratio of characteristic diameter at the initial state at 20C. (Reprinted ref. 57, Copyright American Chemical Society. Reproduced with permission.)

Figure 11 showed the logarithmic plots of the period against the initial concentration of one BZ substrate under fixed concentration of the other two BZ substrates at a constant temperature (*T*=20 C). As shown in Figure 11, all the logarithmic plots had a good linear relationship. Therefore, the period [T(s)] of the swelling-deswelling self-oscillation can be expressed as *a*[substrate]*b* where *a* and *b* are the experimental constants, and the brackets indicate the initial concentration. Moreover, as shown in Figure 11, the period of the selfoscillation have the saturation point at the following initial concentration: [MA] = 0.07M (Figure 11(a)), [NaBrO3] = 0.5M (Figure 11(b)) and [HNO3] = 0.7M (Figure 11(c)). The period at the saturated point in Figure 11(a) was significantly higher than that in Figure 11(b) and Figure 11(c). This tendency can be explained by considering the mole fraction of the reduced Ru(bpy)3 moiety in the gel. This is because the reduced Ru(bpy)3 moiety in the gel has the significantly high hydrophobic property. Therefore, the number of the hydrophobic reduced Ru(bpy)3 moiety in the gel exert influence of the self-oscillating behavior. The Field-koros-Noyes (FKN) mechanism explained the overall process of the BZ reaction. [33-36, 48] According to the FKN mechanism, the overall reaction is divided into the following three

**Figure 11.** Logarithmic plots of period (T in s) vs initial molar concentration of one BZ substrate at a constant temperature (T=20 C) under fixed concentrations of the other two BZ substrates; (a) [NaBrO3] = 0.084M and [HNO3] = 0.3M, (b) [MA] = 0.0625M and [HNO3] = 0.3M, (c) [MA] = 0.0625M and [NaBrO3] = 0.084M. (●) plots and (○) plots show the linear relation and the saturated line vs initial concentration of one BZ substrate, respectively. (Reprinted ref. 57, Copyright American Chemical Society. Reproduced with permission.)

main processes: consumption of Br ions (process A), autocatalytic formation of HBrO2 (process B), and formation of Br ions (process C).

A: BrO3- + 2Br- + 3H+ → 3HOBr B: BrO3- + HBrO2 + 2Mred + 3H+ → 2HBrO2 + 2Mox + H2O C: 2Mox + MA + BrMA → *f*Br- + 2Mred + other products

In the process of B and C, the Ru(bpy)3 moiety in the gel works as the catalyst: the reduced Ru(bpy)3 moiety is oxidized (process B), and the oxidized one is reduced (process C). Therefore, as the initial concentration of the MA increased, the mole fraction of the reduced Ru(bpy)3 moiety in the gel increased in accordance with the FKN mechanism. With increasing in the mole fraction of the reduced Ru(bpy)3 in the gel, the shrinking force originating in the hydrophobic reduced Ru(bpy)3 greatly increased as well. Generally, as for a polymer gel, deswelling speed is faster than the swelling one. Once the gel collapsed, it takes a lot of time for the aggregated polymer domain in the gel to recover the elongated state. This is because the polymer aggregation state is thermodynamically more stable in the polymer gel. Therefore, as the shrinking force increased, the swelling speed of the poly(VP*co*-Ru(bpy)3) gel significantly decreased. As a result, in the higher MA condition, the period at the saturated point was long (T=182.5) compared with the other condition ( T=33.7 (Figure 11(b)) and T=45.0 (Figure 11(c)). On the other hand, in the case of Figure 11(b), the period at the saturated point was greatly shorter than that in Figure 11(a). In the condition of Figure 11(b), the swelling force originating in the hydrophilic oxidized Ru(bpy)3 moiety increased due to the increase in the mole fraction of the oxidized Ru(bpy)3 moiety in the gel in accordance with the FKN mechanism. Therefore, the gel can cause the swelling-deswelling self-oscillation at the high speed due to the strong recovering force originating in the higher mole fraction of the hydrophilic oxidized Ru(bpy)3 moiety in the gel. Moreover, in the condition of Figures 11(c), the period for the poly(VP-*co*-Ru(bpy)3) gel had the different aspect from that of the conventional-type poly(NIPAAm-*co*-Ru(bpy)3) gel.[75] The period of the self-oscillation decreased with increasing the initial concentration of the BZ substrates because of the increase in the collision frequency among the BZ substrates. Therefore, we considered that the relationship between the period and the [HNO3] for the poly(VP-*co*-Ru(bpy)3) gel is of a more natural tendency. In addition, in the condition of Figure 11(c), the control range of the period by changing the initial concentration of the HNO3 for the gel was much wider than that for the poly(NIPAAm-*co*-Ru(bpy)3 gel. [75]

324 Smart Actuation and Sensing Systems – Recent Advances and Future Challenges

**Figure 11.** Logarithmic plots of period (T in s) vs initial molar concentration of one BZ substrate at a constant temperature (T=20 C) under fixed concentrations of the other two BZ substrates; (a) [NaBrO3] = 0.084M and [HNO3] = 0.3M, (b) [MA] = 0.0625M and [HNO3] = 0.3M, (c) [MA] = 0.0625M and [NaBrO3] = 0.084M. (●) plots and (○) plots show the linear relation and the saturated line vs initial concentration of one BZ substrate, respectively. (Reprinted ref. 57, Copyright American Chemical Society. Reproduced

(c)

(a) (b)


Period[s]

ions (process C).

+ HBrO2 + 2Mred + 3H+ → 2HBrO2 + 2Mox + H2O

C: 2Mox + MA + BrMA → *f*Br- + 2Mred + other products

ions (process A), autocatalytic formation of HBrO2

with permission.)

A: BrO3-

Period[s]

B: BrO3-

main processes: consumption of Br-

+ 3H+ → 3HOBr

(process B), and formation of Br-

+ 2Br-

Moreover, as shown in Figure 12, the period of the swelling-deswelling self-oscillation decreased with increasing the temperature because the temperature affects the BZ reaction rate in accordance with the Arrenius equation.30 The period of the swellingdeswelling self-oscillation to the temperature for the poly(VP-*co*-Ru(bpy)3) gel has the linear relationship. The period (2 second) reached the saturation at 46 C in the BZ condition ([MA] = 0.08M, [NaBrO3] = 0.48M and [HNO3] = 0.48M). This is because the swelling-deswelling speed of the gel is slower than the changing rate of the redox states of the Ru(bpy)3 in the gel. That is, the self-oscillating behavior of the gel cannot follow the changing the redox state of the Ru(bpy)3 moiety. The maximum frequency (0.5Hz) of the poly(VP-*co*-Ru(bpy)3) gel was 20 times as large as that of poly(NIPAAm-*co*-Ru(bpy)3 gel. [75] The self-oscillating behaviors of the poly(VP-*co*-Ru(bpy)3) gel at 20°C and 50°C were shown in the Figure 12(b) and 12(c), respectively. The displace of the volume change selfoscillation at 20°C and 50°C were about 10μm and 4μm, respectively. These results clarified that the displacement of the swelling-deswelling self-oscillation for the gel has the trade-off relationship against the period of the self-oscillation, that is, the length of the volume change decreased with increase in the period.

**Figure 12.** (a) Dependence of the self-oscillation period on the temperature. (●) plots and (○) plots show the linear relation and the saturated line vs temperature, respectively. (b) Self-oscillating profile of cubic poly(VP-*co*-Ru(bpy)3) gel at 50C (MA = 0.08M, NaBrO3 = 0.48M and HNO3 = 0.48M). (c) Self-oscillating profile of cubic poly(VP-*co*-Ru(bpy)3) gel at 20C (MA = 0.08M, NaBrO3 = 0.48M and HNO3 = 0.48M). Cubic gel (each side length is about 2mm and 20mm) was immersed in 8ml of the mixture solution of the BZ substrates. (Reprinted ref. 57, Copyright American Chemical Society. Reproduced with permission.)
