Temperature-dependent electrorheology of a suspension based on copolymeric P(NIPAM-co-[AMIm]Cl) colloidal particles

N-isopropylacrylamide (NIPAM) (TCI, 98%) was recrystallised in a mixture of hexane and toluene (volume ratio of hexane to toluene = 5:1). 2, 2'-azobis(2-methylpropionitrile) (TCI, 98%) was recrystallised in methanol. The ionic liquid (IL) 1-allyl-3-methyl imidazolium chloride ([AMIm]Cl) was purchased from Lanzhou Greenchem ILs and used directly. Other chemicals including divinylbenzene (DVB), sodium dodecylsulfate (SDS) (Sigma Aldrich, 99%), dehydrated tetrahydrofuran (THF) (Macklin, 99%), N,N-dimethylformamide (DMF) (TCI, 99%), and dimethyl silicone oil (KF-96, 100 cSt, Shin-Etsu Chemical Co., Ltd) were used as received without further purification.

NIPAM of 2.96 g (26 mmol), [AMIm]Cl of 0.41 g (2.6 mmol), and AIBN of 0.205 g were added to 10 ml THF in a 100 ml one-necked, round-bottomed flask. The solution was magnetically stirred under N₂ purging to remove oxygen. Thereafter, the flask was placed in an oil bath and heated at 65 °C for 12 h. The obtained product was purified by several re-precipitation cycles in cold diethyl ether. Then, the solid product was dried under vacuum for 5 h at 50 °C [42]. The dried copolymer was characterised by ¹H NMR (Avance III 500 Bruker A&T Center BNU 500 MHz) as shown in figure 1. The two monomers were also individually polymerised for comparative study. P[AMIm]Cl: 1.18 g [AMIm]Cl and 1.3 g DVB were added in 4.00 ml DMF in a 100 ml one-necked, round-bottomed flask. After removing oxygen by purging N₂, the system was heated at 70 °C under mechanical agitation with a speed of 180 rpm. After 12 h, the product was wahsed with ethanol and de-ionized water for several times, then freeze-dried for 24 h. PNIPAM was synthesised using the similar process for the copolymer without adding the IL monomer.

Zoom In Zoom Out Reset image size

P(NIPAM-[AMIm]Cl). ¹H NMR (500 MHz, CDCl₃, δH, ppm): 10.46 (1H, NCHN), 7.41 (1H, NHCO), 7.32 (1H, NCHCHN), 7.26 (1H, NCHCHN), 3.97 (1H, CH(CH₃)₂N), 1.70 (1H, CH₂CHCH₂), 1.68 (2H, CH₂CHCH₂), 1.61 (2H, CHCH₂CH), 1.52 (1H, CHCH₂CH), 1.32 (2H, CHCH₂N), 1.17 (3H, N(CH)₂CH₃), 1.12 (6H, CH(CH₃)₂).

The synthesised solid PNIPAM-[AMIm]Cl was milled and sieved (325 mesh) to form uniform powders with particle sizes in the range of 5 to 20 μm. An oil suspension, i.e. electrorheological (ER) fluid, was prepared by mixing the copolymer particles with dimethyl silicone oil by mechanical stirring and ultrasonication. The concentration of the ER fluid is 20 wt.%.

¹H nuclear magnetic resonance (¹H NMR) measurement of the sample was conducted at room temperature using a Bruker Avance III 500 NMR spectrometer (500 MHz). CDCl3 was used as a solvent.

The rheological properties of the suspensions were measured using a rotational rheometer (MCR 502, Anton Paar, Austria) with a cylindrical cone-plate geometry (CC17E, gap distance: 0.71 mm) and a DC high-voltage generator. The dielectric spectra of the ER fluid were detected using a broadband dielectric spectrometer (Concept 80, Novocontrol) under AC electric field conditions with frequencies ranging from 0.01 to 10⁷ Hz at different temperatures. A bias electrical potential of 1 V was applied during the measurement to avoid formation of fibrous structures in the suspension.

The thermo-sensitivity of PNIPAM in water required the investigation of the temperature dependence of the copolymer PNIPAM-[AMIm]Cl in silicone oil: the leakage current density at different temperatures was also identified. Figure 3 shows the flow curves of the ER fluid measured at 20 °C and 50 °C, respectively, in which figures 3(a)–(b) are the shear stress curves, and figures 3(c)–(d) are the shear viscosity curves. Without stimulus from an applied external electric

Zoom In Zoom Out Reset image size

field, the ER fluid is non-Newtonian with shear thinning observed at both 20 °C and 50 °C. The shear viscosity at 50 °C is slightly higher than that at 20 °C in the low-shear-rate region, however, this changes as the shear rate exceeds 100 s⁻¹( the reason for this will be explained later). When an electric field is applied, Bingham-plastic behaviour is observed with an electric-field-dependent yield stress. As the electric field strength increases, the yield stress increases. The shear stress at 50 °C is significantly higher than that at 20 °C when measured at the same applied electric field strength: the temperature has a remarkable positive effect on the ER performance of the PNIPAM-[AMIm]Cl-based ER fluid.

To demonstrate the temperature-dependence of the ER performance clearly, flow curves of the PNIPAM-[AMIm]Cl-based ER fluid were measured at different temperatures ranging from 5 °C to 50 °C. The result is shown in figure S1 available online at stacks.iop.org/SMS/29/124001/mmedia. At 5 °C, yield stresses of the ER fluid are very weak at about 1 to 10 Pa as the electric field strength increases from 0.5 to 4.0 kV mm⁻¹. The zero-field shear stress is a typical Newtonian fluid flow curve rather than a non-Newtonian fluid as shown by the ER fluid measured at other higher temperatures. When the temperature is above 10 °C, relatively stable and electric-field-dependent shear stress curves are observed. In addition, it seems that as the shear rate increases, the shear stress undergoes a slight decrease until the shear rate reaches a critical value, which could be explained as the balance between break-up and reformation of the chain structures by the suspended copolymer particles [43]. To describe the flow behaviour of the ER fluid in detail, a modified Bingham model called Cho-Choi-Jhon (CCJ) with six parameters is adopted to fit the shear stress curves [44]:

$$\begin{equation}\tau = \frac{{{\tau _{\text{y}}}}}{{1 + {{\left( {{t_2}\dot \gamma } \right)}^\alpha }}} + {\eta _\infty }\left( {1 + \frac{1}{{{{({t_3}\dot \gamma )}^\beta }}}} \right)\dot \gamma {\text{ }}\end{equation} \tag{ 1 }$$

where τ_y is the dynamic yield stress that can be obtained by extrapolating the shear stress curves to the zero shear rate direction, η_∞ represents the viscosity at an infinite high shear rate, t₂ and t₃ are time constants, α and β are both between 0 and 1 and are related to the change of shear stress in low and high-shear-rate regimes, respectively. The solid lines in figure 3 and figure S1 are obtained from fitting by equation (1), which are able to describe the shear stress curves over the whole shear rate range.

To compare the dynamic yield stresses of the ER fluid at different temperatures, the fitted values of τ_y are plotted in figure 4(a) and S2, respectively, as a function of temperature and electric field strength (E). As shown in figure 4(a) clearly, the dynamic yield stress increases with temperature in the measured temperature range. There is an inflection point near 30 °C, beyond which the slope of the curve becomes larger. It means when the temperature is higher than 30 °C, the temperature dependence of τ_y becomes stronger. In figure S2, the dynamic yield stress increases following an increment in the electric field strength. Generally, the dependence of the yield stress on the electric field strength follows a power-law relationship as follows:

$$\begin{equation}{\text{ }}{\tau _{\text{y}}} \propto {E^m}\end{equation} \tag{ 2 }$$

Zoom In Zoom Out Reset image size

In figure S2, the solid lines are fitted by a best-fit method using equation (2). It seems that the slope of the fitting line (m) is also related to temperature. Therefore, the fitting parameter m is plotted as a function of temperature in figure 4(b). There is a significant change in the value of m when the temperature is increased from 5 °C to 50 °C. In the low temperature region, m increases with temperature; however, it reaches a maximum at about 30 °C, then decreases as the temperature continues to rise. This result indicates that polarisation of the copolymer particles at an electric field is enhanced by increasing temperature, but there is a critical temperature above which the facilitating effect of the rising temperature diminishes.

The current density (I) in the ER fluid when explored in an electric field is also detected and shown in figure 4(b). It increases slightly with temperature when the temperature is below 30 °C. As the temperature increases, the current density increases rapidly to much higher values, which is similar with the phenomenon that is shown by the dynamic yield stress. Even though, the value of current density remains at an acceptably low level (μA cm⁻²) compared with other ER fluids. The same critical temperature of 30 °C for parameter m and current density I implies that high-temperature-induced current leakage might be the main reason for the decrease in m.

The temperature effect on ER properties has been reported in a previous study. It has been confirmed that, within an appropriate temperature range, the ER response could be improved by increasing temperature because of the facilitated polarisation at higher temperatures [45]. Yin et al[ 46] reported that for a carbonaceous-polyaniline-nanotubes (CPN)-based ER fluid, the yield stress increased as the operating temperature was increased. Similar results were also found in TiO₂- and the recently reported PIL-based ER fluids [28, 47] however, in these studies, the increment in yield stress or shear stress with temperature is significantly lower than that found in our study. To observe the difference more clearly, a relative increment in yield stress (Δτ_r) with temperature is defined as follows:

$$\begin{equation}\Delta {\tau _r} = \frac{{{\tau _{y2}} - {\tau _{y1}}}}{{{\tau _{y1}}\left( {{T_2} - {T_1}} \right)}}\end{equation} \tag{ 3 }$$

where τ_y1 and τ_y2 are the yield stresses at temperatures T₁ and T₂, respectively. For the aforementioned TiO₂ ER fluid, Δτ_r is 0.009 1 °C⁻¹( E = 3 kV mm⁻¹) when the temperature is increased from 20 °C to 50 °C. It is even lower for the CPN and pure PILs ER fluids, whereas it is 0.116 1 °C⁻¹( E = 3 kV mm⁻¹) for the copolymer PNIPAM-[AMIm]Cl-based ER fluid when the temperature increases from 20 °C to 50 °C. The relative increment in yield stress of this PNIPAM-[AMIm]Cl-based ER is much higher than that of previously reported ER fluids. This means the ER effect of the copolymer PNIPAM-[AMIm]Cl is more sensitive to temperature. Fortunately, the influence of temperature is towards a positive intensification of the ER effect.

Rotational rheological measurements at a constant shear rate (1 s⁻¹) with the application of a square voltage pulse (50 s) were used to observe the response sensitivity of the ER fluid of PNIPAM-[AMIm]Cl to electric field. In figure 5, it is also found that the electric-field-responsibility of the ER fluid depends on temperature. At the same electric field, the shear stress measured at 50 °C is significantly higher than that measured at 20 °C. In addition, the shear stress changes more rapidly to a stable value after the electric field is switched on or switched off. The obvious distinction on both shear stress value and response time between the two temperature situations can be attributed to the temperature-sensitive properties of the PNIPAM segment. As the temperature exceeds some critical value, the swelling of PNIPAM segment leads to an increase in the particle volume which weakens the binding of the polymer skeleton to the anion. The effect facilitates the polarisation of the copolymer particle which is aroused by the migration of anions in the particle, and enhances the interaction between particles [48].

Zoom In Zoom Out Reset image size

Dynamic oscillation testing is an important tool used when examining the viscoelastic characteristics of the PNIPAM-[AMIm]Cl ER fluid. At first, a strain sweep test was investigated at a fixed angular frequency of 6.28 rad s⁻¹ to determine the linear viscoelastic region (γ_LVE), as shown in figure 6. In the low-strain region, the storage modulus G'( the formation of energy storage in the sample during the shear process) is always higher than that of the loss modulus G'' (the formation of energy dissipated during the shear process) [49]. In addition, the values of G' and G'' are independent within the region under strain (the so-called linear viscoelastic region) [50]. In this region, elasticity predominates (as opposed to viscosity); the chain structures are formed under the applied electric field and remain stable. Notably, as the temperature increases from 20 °C to 50 °C, the linear viscoelastic region broadens because the increasing temperature not only causes the volumetric expansion of the particles, but also facilitates the stronger polarisation forces between the particles by the migration of counter ions within the particles [51]. As the strain amplitude increased, going beyond the LVE region, both G' and G'' decrease gradually due to an irreversible change in the chain structures caused by the oscillatory shear deformation. In addition, the value of G'' exceeds that of G', showing that the structures begin to be destroyed over a certain range of deformation, and the elasticity of the ER fluid disappears at a particular strain amplitude.

Zoom In Zoom Out Reset image size

The dynamic moduli as a function of the angular frequency at 0.01% strain in the linear viscoelastic region were measured and the results are shown in figure 7. In the absence of an electric field, both G' and G'' are dependent on the angular frequency: however, after the application of the electric field, both G' and G'' increase with the increase of electric field strength, so G' is always greater than G'', showing the predominance of elastic behaviour in the structure of PNIPAM-[AMIm]Cl ER fluids. Over the entire frequency range, G' remains stable, showing that the ER fluids exhibit behaviour typical of cross-linked rubber. Additionally, similar to the temperature sweep, it shows that at zero field, G' < G'' at 20 °C showing liquid-like behaviour. When the temperature is 50 °C, G' > G'' showing solid-like behaviour, thus verifying the accuracy of the temperature-scan data.

Zoom In Zoom Out Reset image size

To determine further the reason for dramatic increase in ER effect with temperature, we polymerised the monomer [AMIm]Cl containing DVB as a cross-linking agent, and also prepared pure PNIAM as another comparison sample. Figure 8 shows that, under an external electric field, the ER fluid of P[AMIm]Cl exhibits similar shear stress values at both 20 and 50 °C. The difference in shear stress at zero field may be caused by the dispersion state of the particles. In figure S3, for the PNIPAM-based ER fluid, as the electric field increases from 0.5 to 2 kV mm⁻¹, there is no change in shear stress, indicating no ER effect in this ER fluid. Comparing the zero and on-field shear stress curves, the increment of shear stress in the low shear rate region is due to the pressure of the pin (connected to the power supply) on the rotating bob. Thus we can conclude that PNIPMA itself has no ER effect, the ER response of the copolymer is originated from the PIL part. The interesting point is the addition of PNIPAM gives the copolymer significant temperature-dependent ER properties.

Zoom In Zoom Out Reset image size

Then, dynamic temperature sweeps and calorimetric analyses were conducted on these polymer-based ER fluids. As mentioned, PNIPAM is a temperature-sensitive polymer with an LCST of about 32 °C when dispersed in water, however, in this study, the copolymer particles are dispersed in silicone oil, of which the temperature effect has not been considered. To visualise the temperature dependence of the ER fluid, we also conducted dynamic temperature sweeps for the ER fluid, and for the suspension of P[AMIm]Cl and PNIPAM in silicone oil with the same mass fraction for sake of comparison. As shown in figure 9, both storage modulus (G') and loss modulus (G'') of the PNIPAM-[AMIm]Cl ER fluid increase with temperature. There is an intersection of G' and G'' plots near 32 °C, before which G'' exceeds G' indicating the liquid-like nature of the ER fluid, and after which G' exceeds G'' implying that elastic interaction is dominant in the ER fluid. It seems that the polymer chains are swollen by silicone oil at higher temperature, which enhances the volume fraction and then the interaction between polymeric particles. In the temperature sweep curve of the PNIPAM suspension, a similar trend is observed wherein both G' and G'' increase gradually with temperature; however, the values of G' and G'' for the PNIPAM suspension are lower than those of the PNIPAM-[AMIm]Cl ER fluid. By contrast, the suspension of pure P[AMIm]Cl shows slight decrease in G' and G'' with increasing temperature. This implies that the swelling of the copolymer mainly occurs in the PNIPAM segment rather than in the P[AMIm]Cl part. Thus, the temperature effect for the pure PNIPAM or copolymer PNIPAM-[AMIm]Cl in silicone oil is very different from that of the PNIPAM hydrogel, in which G' and G'' decrease or increase suddenly when the temperature is higher than LCST because of the phase transition occurring in the PNIPAM hydrogel [52, 53]. Similar phenomenon was also observed in the aqueous suspension of PNIPAM-[AMIm]Cl (20 wt.%) as shown in figure S4.

Zoom In Zoom Out Reset image size

To confirm the predicted swelling of PNIAPM in silicone oil, the heating capacity of suspensions containing different polymers PNIPAM-[AMIm]Cl, PNIPAM, and P[AMIm]Cl were respectively measured using a differential scanning calorimeter (DSC) in the temperature range of 10 to 60 °C. As shown in figure 10, broad endothermic peaks are observed in the C_p curves of PNIPAM/silicone oil and PNIPAM-[AMIm]Cl/silicone oil suspensions. There is no endothermic effect in the C_p curve of P[AMIm]Cl/silicone oil suspension. This confirms that the endothermic peak in PNIPAM-[AMIm]Cl/silicone oil is induced by the PNIPAM segment, however, the C_p curves of both the PNIPAM/silicone oil and PNIPAM-[AMIm]Cl/silicone oil are very different from that of aqueous suspension of PNIPAM, which shows a narrow endothermic peak because of the hydrophilic-hydrophobic transition at the temperature of LCST. In addition, according to the optical image of the PNIPAM-[AMIm]Cl/silicone oil suspension (inset, figure 9), it is found that the suspension is non-transparent and brown whether at lower (25 °C) or higher (50 °C) temperature, therefore, we predict that the endothermic effects in both PNIPAM/silicone oil and PNIPAM-[AMIm]Cl/silicone oil suspensions are attributed to the interaction between silicone oil and the PNIPAM chain, such as evinced by the swelling of the PNIPAM chain in silicone oil with increasing temperature.

Zoom In Zoom Out Reset image size

ER effect is generally considered to be aroused by the polarization induced electrostatic force between particles, the value of which is related to the dielectric properties of both the solid particles and the dispersion liquid. Therefore, dielectric analysis is a critical method to understand the ER effect of ER fluids. It has proved that a large dielectric strength and a dielectric relaxation in a proper range (10² ∼ 10⁵ Hz) are necessary for a good ER effect. From the study of dielectric spectra of ER fluids with different volume fractions or at different temperatures, one can achieve the information about the dielectric parameters [12], phase parameters [54], and even activation energy [13] for dielectric relaxation.

In this study, dielectric properties of the PNIPAM-[AMIm]Cl ER fluid were measured to correlate dielectric characteristics of the ER fluid with its rheological performance under different electric and temperature field conditions. To clarify the dielectric characteristics, the polarisation effect was investigated using two significant parameters: the dielectric constant (') and dielectric loss factor (''). Figure 11 shows the dielectric constant and dielectric loss factor as a function of frequency. The Havriliak–Negami equation is used to analysis the dielectric spectra of the ER fluid [55]:

$$\begin{equation}{\varepsilon ^ * }\left( \omega \right) = \varepsilon {\prime} + i\varepsilon {\prime\prime} = {\varepsilon _\infty } + {{{\varepsilon _0} - {\varepsilon _\infty }} \over {{{\left[ {1 + {{\left( {i\omega {\tau _{{\rm{HN}}}}} \right)}^\alpha }} \right]}^\gamma }}} + {{{\sigma _{{\rm{dc}}}}} \over {i{\varepsilon _0}\omega }} - A{\omega ^n}\end{equation} \tag{ 4 }$$

Zoom In Zoom Out Reset image size

where α and γ are the shape factors both in the range of 0 ∼ 1, ₀ is the dielectric constant when the frequency is close to 0 and _∞ is that at a high-frequency limit. Δ□ = ₀ − _∞ is the dielectric strength also considered achievable polarisability of ER fluids closely related the magnitude of the interparticle interaction [27, 56], σ is the DC-conductivity, and τ_HN represents the relaxation time expressed in terms of as the reciprocal of the frequency (f_max) where the maximum dielectric loss appears (i.e. τ_HN = 1/(2πf_max).

Parameters in equation (4) are listed in table 1. For the PNIPAM-[AMIm]Cl ER fluid, as the temperature increases, Δ also increases, indicating the enhancement of interparticle strength. The relaxation time of the ER fluid shows the opposite tendency and decreases as the temperature increases, which means the response of the particles becomes faster with the increasing of temperature. This shows that, as the temperature increases, the polarisability (Δ) of the particles increases and the relaxation time (τ_HN) decreases, resulting in better ER performance. Therefore, the analysis of dielectric spectra can confirm the results obtained from a comparison of their shear stress and storage modulus at different temperatures.

Table 1. The optimal values of the parameters in equation (4).

Temperature (°C)	Δ	α	γ	σdc	τHN( s)
20	3.23	0.86	0.66	0.02	8.13
25	3.59	0.83	0.70	0.02	2.86
30	3.65	0.82	0.71	0.03	0.92
35	3.52	0.85	0.64	0.03	0.30
40	2.77	0.89	0.69	0.25	0.08
45	2.44	0.99	0.51	1.12	0.02

In figure 11, it is clear that the relaxation peaks of the ER fluid at different temperatures all locate in the low frequency range referring to the interfacial polarisation, which is caused by the dissociation of ion pair and the local diffusion of dissociated ions. Herein, the cationic ion [AMIm]⁺ covalently bonded on the polymer backbone is almost immovable; therefore, the interfacial polarisation is only associated with the movement of Cl^-, however, the relaxation peak in this work is not in the above mentioned proper range (10²–10⁵Hz). To understand this, we calculated the activation energy (E_a) via fitting the experimental data by the Arrhenius equation:

$$\begin{equation}1/{\tau _{{\text{HN}}}}\propto{\text{ }}{e^{ - {E_{\text{a}}}/RT}}\end{equation} \tag{ 5 }$$

where R is the molar gas constant, T is the absolute temperature. The result is shown in figure 12 . Herein, E_a is related to the dynamics of ions in the interfacial polarization process. It is calculated to be 183.43 kJ mol⁻¹, which is much higher than that presented by the PILs with organic counter ions. The high activation energy may be the reason for the low current density of the PNIPAM-[AMIm]Cl ER fluid under electric field as well. Nevertheless, temperature-induced swelling of the backbone provides larger space for the ions to move through, which enhances the ER effect of the ER fluid.

Zoom In Zoom Out Reset image size