PANI particles were synthesized using the conventional oxidation polymerization of aniline to produce a fine emeraldine hydrochloride form [ 5 ]. A solution of aniline monomer in 1 M HCl was chilled, and polymerization was initiated at 0°C by a prechilled solution of ammonium peroxysulfate in 1 M HCl. The reaction was maintained for additional 2 h to complete the reaction to produce the emeraldine hydrochloride form. A portion of the PANI particles was dedoped by reducing the pH of the aqueous medium, which contained the particles, to pH 9.0 using an aqueous NaOH solution, which allowed them to be used as dispersed particles for an ER fluid directly because of their high conductivity (≈ 1 S/cm). The rest of the PANI was capsulated by an insulating MF resin, which will be explained in the following section. The dedoped PANI particles were washed with ethanol to remove the unreacted monomer and oligomer. Products were subsequently dried for 1 day in vacuum. Synthesized PANI particles were used as the core material for the encapsulation with prepolymeric MF as the shell material [ 17 ]. The dried PANI particles were dispersed in an aqueous medium of pH 4 containing citric acid as a proton source. Then, both the melamine prepolymer and formaline were added to the solution by stirring at 40 45°C. As the temperature rose to approximately 65°C, the MF resins were cured [ 18,19 ], and microencapsulated polyaniline (MCPA) particles were produced. After completion of the encapsulation reaction, the particles were filtered and washed with distilled water several times to remove the unreacted chemicals. Finally, the products were dried in a vacuum oven [ 19 ]. The PANI concentration in MCPA was approx. 10% (w/w) based on the input quantity of these reactants. The MCPA did not require an additional dedoping process for ER applications due to the insulating MF shell. In the case of core/shell structures of PANI, initially, PMMA microparticles were prepared as core particles via a dispersion polymerization method [ 20 ]. The purified methyl methacrylate monomer and a radical initiator (azoisobutyronitrile) were dissolved in methanol containing poly(vinylpyrrolidone) as a stabilizer, and the polymerization was conducted at 55°C for 24 h. The monodisperse particles obtained were washed sufficiently with methanol, and then dispersed in water prior to the PANI polymerization.

Core/shell (PMMA/PANI) particles (PAPMMA) were obtained by following the protocol established by Barthet et al. [ 21 ]. The PMMA particles were first dispersed in water containing sodium dodecyl sulfate (SDS) and kept with mild stirring to allow SDS to contact and adsorb onto the PMMA surface. Aniline was then added to this suspension with a desired ratio to the amount of PMMA core, and the system was acidified to pH 0.7 using HCl, followed by the addition of a solution of initiator (ammonium persulfate) in aqueous HCl (pH 0.7). The polymerization was continued for 24 h. The PAPMMA products were washed and their conductivities were controlled within the range of 10-9 to 10-10 S/cm by the same above-mentioned procedure. After adjusting the conductivity, the final products were dried in a vacuum oven.

A Fourier transform infrared spectrometer (FT-IR, Perkin Elmer System 2000, Norwalk, CT) was used to identify the chemical structure of the particles and confirm their chemical reactions. Thermal stability of particles was also examined via a thermogravimetric analyzer (TGA, Polymer Laboratories, PL-TGA, Amherst, MA) at a heating rate of 20°C/min. The shape and the size of the particles were also observed using a scanning electron microscope (SEM, S-4200, Hitachi, Hitachnaka-shi, Japan). Conductivities of PANI, PAPMMA, and MCPA particles were measured from a two-probe method using compressed disks of the prepared samples. The resistance of the pellets was measured by a picoammeter (Keithley model 485, Cleveland, OH) with a conductivity cell [ 22 ].

ER fluids were prepared by dispersing the powders in dry silicone oil with 10% (v/v) of solid content. The density and kinematic viscosity of the silicone oil were 0.95 g/cm3 and 50 cSt, respectively. ER properties were examined via a rotational rheometer (Physica MC 120, Stuttgart, Germany) with a concentric cylinder geometry equipped with a high voltage generator [ 23 ]. Various DC electric field strengths were applied to the cylindrical cup, perpendicular to the flow direction. To obtain an equilibrium internal structure of the particles, an electric field was applied to the ER fluids for 3 min before the rheological measurements. All measurements were taken at 25°C unless specified. (c) PAPMMA (b) MCPA 100

80 ] % [ t igh 60 e w l a du 40 i s e R 20 0

0 4000 3500 3000 2500 2000 1500 1000

500

Wavenumber [cm-1]

Temperature [ oC]

Fig. 1 shows the FT-IR spectra for particles of PANI, MCPA, and PAPMMA. Note that PANI has peaks at 824, 1144, 1309, and at 1490 and 1586 cm-1, indicating aromatic C–H, aromatic amine and aromatic C–C stretching vibrations (spectrum (a)), respectively [ 24 ]. The peaks of MCPA (spectrum (b)) at 1480, 1550, and 3300, 1363, and 1660 cm-1 indicate –CH2–, N–H, C–N, and C=N, respectively, from MF resin in the shell. These also confirm that the MCPA was successfully synthesized using MF resins. Spectrum (c) is obtained from PAPMMA and there is a strong peak at 1740 cm-1 from carbonyl groups in the PMMA core.

TGA diagrams of the PANI, MCPA, and PAPMMA particles are shown in Fig. 2. The thermogram indicates that the weight loss around 100°C was due to evaporation of water or solvent used for washing, and decomposition of the PANI particles began at approximately 350°C. Weight loss of this sample was approximately 20% at 600°C, showing that the prepared PANI particles have a good thermal stability. Kulkarni et al. [ 25 ] investigated the thermal stability of PANI and suggested that the decomposition of the doped PANI could be described as a two-step process. The first step is the loss of the dopant and the second step corresponds to the decomposition of the backbone. As the doping state of the PANI particles increased, their thermal stability decreased. However, weight loss of MCPA above 100°C is due to uncured MF resin, especially oxidation degradation of the melamine rings at 350 - 400°C and degradation of the residual fraction at 450°C [ 17 ]. PAPMMA has little thermal degradation up to 200°C, and its weight loss increased abruptly from 300 to 400°C, which is due to the degradation of the PMMA core. The secondary plateau observed around 400°C is from the PANI shell that keeps more than 95 % of the initial weight.

Fig. 3(a) and (b) show SEM images of both PANI and MCPA particles. The shape of MCPA from the SEM image is grape-like. With respect to shape and size, MCPA was observed to be larger (average diameter of approx. 70 µm) and more spherically symmetric than PANI, which has an average particle diameter of 17.9 µm. MCPA also has a considerably smoother surface relative to PANI, which will be good for lowering the viscosity of the suspension. On the other hand, PMMA core (9 µm) and PAPMMA core/shell particles are spherical and have a narrow particle size distribution (Fig. 3(c) and (d)). The shape of PAPMMA particles (Fig. 3(d)) is very rough, and the uniformity of particle size was maintained without any aggregates, even after forming a PANI layer on the PMMA core.

The flow curves for three different ER fluids based on PANI, MCPA, and PAPMMA with a particle concentration of 10 vol.-% without an external electric field or under an applied electric field strength of 2 kV/mm are shown in Fig. 4. All ER fluids showed an increase of shear stress (or shear viscosity) by the electric field, and furthermore, the flow characteristics were largely changed from pseudo-Newtonian to Bingham fluid, which flows under an applied force larger than its yield stress. The change of flow properties is due to the transition of the ER fluid from the liquid-like to the solid-like state under the electric field. The typical trends for ER fluids [ 26,27 ], a decrease of shear stress (τ) with increasing shear rate ( γ& ) up to a critical shear rate ( γ& crit ), were found only for the PANI-based ER fluid (Fig. 4(a)). The MCPA-based ER fluid exhibits constant τ in the entire γ& range, while the PAPMMA-based ER fluid exhibits constant shear stress in the shear rate region that is smaller than γ& crit . γ& crit is a transition point of γ& at which the suspension begins to exhibit pseudo-Newtonian behaviour (τ increases with γ& ). Without an external electric field, the suspension shows shear thinning behaviour following a Newtonian plateau (i.e., constant viscosity with γ& ). Because of topological factors, PANI-based suspensions are two times more viscous than MCPA- or PAPMMA-based suspensions at the same particle concentration of 10% (v/v).