Comparing the Electrorheological Effect of Polyhedral Silsesquioxane Cage Structures with Different Numbers of Cyanopropyl Functional Groups.

Previous research has shown that polyhedral oligomeric silsesquioxane (POSS) particles in silicone oil show electrorheological (ER) activity. The effect of the number of same functional groups attached to POSS cage structures on the ER activity of these structures is discussed in this article. Two compounds of the octahedral geometry (T8) of cyanopropyl POSS (cPOSS) were used for this investigation. One of the compounds (monofunctionalized cPOSS) was commercially available, and the other (octafunctionalized cPOSS) was synthesized. The effects of the number of cyanopropyl functional groups attached to the inorganic silicon-oxygen core structure of the POSS compounds on the rheological properties are demonstrated through steady flow and oscillatory tests. Particular attention is paid to how the number of cyano functional groups affects the behavior of these suspensions of cPOSS compounds in silicone oil under increasing electric field strength. The research also contributed answers to the effects of changing the concentration of the cPOSS particles in the suspension. The flow curve was described by the Herschel-Bulkley model, and the yield stress values were ascertained from the model. The dielectric characterization was also done to support the ER response results, which showed that the octafunctionalized compound gave a better response. The differences in the ER properties of these compounds have also been discussed with the tests.

Introduction

Smart materials are materials designed so that one or more of their properties change as a result of external stimuli. Electrorheological fluids (ER fluids) are a kind of smart materials, which undergo a change in the properties by the application of the electric field. They are suspensions of polarizable particles in nonpolarizable solvents. The rheological properties of the ER fluid, such as viscosity, stress, and shear modulus, undergo a reversible change by orders of magnitude when an electric field is applied.[3] The suspended particles in the ER fluid are polarized by the electric field application and form chain-like structures and instantaneously return to their original state when the field is removed. The formation of the fibrillated structures causes an increase in resistance to flow and develops yield stress that must be overcome for flow to occur. ER fluids have received attention over the years and find application in shock absorbers, clutches, dampers, actuators, and in ER fluid-assisted polishing in semiconductors.

Diverse particle–solvent suspension systems exhibit ER behavior. The dispersed phase is commonly inorganic particles or organic materials, while the dispersing phase is normally nonpolar solvents with low conductivity. The internal parameters and factors that influence particle polarization have been widely investigated by researchers.

ER suspensions containing high concentrations of particles that possess permanent dipoles show larger ER effects in suspensions than conventional ER fluids.[1,3] Several systems have been used to show this phenomenon. In their comparison of electrorheological performance between urea-coated and graphene oxide-wrapped TiO2 nanoparticles and bare TiO2 nanoparticles, Dong et al.[5] showed that bare titanium nanoparticles had improved ER characteristics as a result of having more polar functional groups attached to it as shell materials. Specifically, for the graphene oxide (GO), there are lots of hydroxyl and carboxylic groups on the basal planes and its edges. These functional groups enhance the electrostatic interaction between the particles, reduce its conductivity, and influence the graphene oxide to become negatively charged. Kawai and Ikazaki[10] studied the electrorheology of three fluids consisting of dispersoids with different functional groups and the same alkyl chain of C16H33, which are dispersed in dimethyl silicone oil. The functional groups are the −OH from n-hexadecanol, −COOH from n-heptadecanoic acid, and −CN from n-heptadecanonitrile. They introduced temperature as a variable for their experiments and monitored the ER effect and complex dielectric constants of the three different compounds as they were heated to their melting points and past it. Through crystallography and infrared spectroscopies, they made observations in the structural changes of these compounds, and after the experiments, they concluded that the ER effect of hexadecanol is highest and that of the heptadecanonitrile is lowest. Cheng et al.[2] synthesized uniform titania microspheres with an enhanced ER effect using an acetic acid-assisted sol–gel method for synthesis. The enhancement in ER properties is attributed to the combined effect of the increased polar molecule content and reduced particle size. Plachy et al.[6] designed an ER suspension (urea-coated lithium titanate nanoparticle-based suspension), which has higher yield stress in two orders of magnitude in comparison to a bare lithium titanate nanoparticle/silicone oil combination. The ER behavior of the bare lithium titanate nanoparticles was improved by coating with urea.

The main attraction for polyhedral oligomeric silsesquioxane (POSS) variants is that they offer multiple functionalities because of their three-dimensional frameworks, which make it possible to add different functional groups or to use them as fillers for different end results.[16] POSS has been used as fillers to mitigate particle sedimentation, which is a major drawback in ER application. Li et al.[17] adjusted the hydrophilicity of graphene oxide (GO) nanosheets so that it can be compatible with silicone oil and overcome phase separation or sedimentation. This has been achieved by decorating the active amine groups of octameric aminopropylisobutyl POSS with the oxygen-containing groups of GO, thereby preventing the reaggregation of GO and reducing the total thickness of the final POSS-GO structure. High yield stress ER fluid that are comparable to that of conventional ER fluids has also been identified by McIntyre et al.[3] in a sulfonated-POSS/PDMS ER fluid system.

There are limited studies on the electrorheology of POSS and the effect of different or number of functional groups attached to the core of the POSS. McIntyre et al.[3] discovered that less than 10% (7% wt) by weight sulfonated polyhedral silsesquioxane cage structures (s-POSS or TSAE-POSS, which means the Tri sulfonic acid ethyl POSS) in poly(dimethylsiloxane) (PDMS or silicone oil) showed an increase of viscosity by a factor of 1 × 102 for an electric field strength of 2 kV/mm. This was in contrast to another POSS particle known as isotropic octa-isobutyl POSS, which exhibited virtually no ER behavior at the same weight percentage. Even when increased to 10% wt, the octa-isobutyl POSS particles still showed no ER effect. Upon investigating the dielectric properties, it was discovered that the octa-isobutyl POSS/silicone oil system is not dielectrically active, while that of TSAE-POSS is active and it was concluded that the presence of the sulfonated groups is responsible for the dipolar behavior. McIntyre et al.[1] did another study, which was an improvement in the previous work in ref (3). This study showed that when mixed with a small (1 wt %) amount of nanocage s-POSS, micrometer-sized polystyrene (PS) particles (10 wt %) in PDMS (89 wt %) liquid gave a significant ER effect. They associated this improved behavior to the formation of a thin adsorbed layer of s-POSS onto the PS surfaces to give a new core/shell particle with improved dielectric and conductive properties. The s-POSS/PS/PDMS suspension exhibited an increase in ER activity by an order of magnitude beyond that of the s-POSS/PDMS suspension. McIntyre et al.[7] improved on their work using sulfonated polystyrene particles (s-PS) instead of PS. The s-PS with s-POSS and PDMS combined to give an even better ER effect. The s-POSS/s-PS/PDMS system exhibited a further increase in yield stress of over 200% when compared with the s-POSS/PS/PDMS mixture.

We recently showed that a new class of ER fluid (octafunctionalized POSS variant in PDMS) exhibited an ER effect. This new POSS cage structure for ER applications was synthesized in our previous work[4] using the hydrolyzation route and was characterized. The dielectric study was done, and rheological properties were examined. In this research, the effects of the number of cyano functional groups attached to the POSS cage structure are shown. The two compounds of the octahedral geometry (T8) of cPOSS were used for this investigation. One of the compounds, octafunctionalized cyanopropylsilsesquioxane (8 CN-POSS), was synthesized and characterized in our previous work[4] and the other compound monofunctionalized cyanopropylisobutylsilsesquioxane (1 CN-POSS) was commercially available. The effects of the functional group attached to the inorganic silicon–oxygen core structure of the POSS compounds on the rheological properties are demonstrated through steady flow and oscillatory tests. Particular attention is paid to how the number of cyano functional groups affects the behavior of these suspensions of cPOSS compounds in silicone oil under increasing electric field strength.

Results and Discussion

Dielectric Properties

The dielectric test was carried out to gain further insight into the differences in the magnitude of the polarizability of the suspensions. Systems with high dielectric constants give enhanced particle polarization, and this leads to strong and stable mesostructure formation within the suspension spanning the electrodes.[3] The height of the ε′ curve shown in Figure reflects the size of the induced dipole moment, which indicates the presence of polarization. The data in Figure show that the 8 CN-POSS exhibited a higher permittivity than the 1 CN-POSS. The introduction of the cyano group to the POSS cage structure led to increased polarization and dielectric activity. The relative permittivity of both suspensions ascends in the low-frequency region, while the dielectric loss shows descent from its maximum value in the low-frequency region. This is an unusual occurrence. This anomalous increase in the dielectric constant before the decrease is attributed to low-frequency conductance. The conductivity peak can result at lower frequencies when the field is not alternating as quickly, and this conductivity is that of the particles and not the suspension. The mechanism of the ER effect had earlier been investigated and understood to be governed by interfacial polarization[11] through a slow polarization process (<1 × 105 Hz). There is no obvious dielectric relaxation observed for the two suspensions between the 1 × 102 and 1 × 105 Hz frequency region at the concentration tested as recommended in the work done by Hao et al.[20]

Figure 1

Frequency dependence of relative permittivity and dielectric loss for 2% wt 8 CN-POSS and 1 CN-POSS in PDMS suspension.

Steady Shear Electrorheological Behavior of 8 CN-POSS and 1 CN-POSS

The dependence of the shear stress on the shear rate for the different concentrations is shown in Figure . The steady-state flow curves were measured with and without applying the electric field to show the ER activity of the suspensions of both ER particles. The non-Newtonian behavior of both suspensions can be seen in Figure . An apparent yield stress is developed even at the 1% wt concentration without the electric field applied. The non-Newtonian behavior of the concentrations of both suspensions at no applied electric field in Figure is unusual, especially for the lowest concentration of 1%. Previous work of Liu et al.[18] showed a Newtonian behavior for silica microspheres and for particle concentration that was as high as 10% by volume. Conversely, an optically transparent ER fluid of urea-modified silica nanoparticles, which have been fabricated by Liu et al.,[19] have been found to behave like a non-Newtonian fluid, but at a higher concentration of 10% by volume at no electric field applied. They attributed this behavior to the dispersion state of the particles and their wettability to the dispersing oil medium. At 0 kV/mm in Figure , the flow curve data appear noisy and remain relatively the same for 0 kV/mm measurements as concentration increases for both 8 CN-POSS and 1 CN-POSS. At 4 kV/mm, it is seen that the shear stress increases as concentration increases. The increased amount of polarizable particles in the suspensions leads to a stiffer solid network being formed, and thus, a higher ER effect. This increase in shear stress is pronounced in the 8 CN-POSS suspension as the concentration is increased in the on-state of 4 kV/mm. The ER effect is higher in the 8 CN-POSS more than the 1 CN-POSS because of the polarity of the eight cyanopropyl functional groups attached to it. The higher particle concentrations in the suspensions are able to form stiffer and stronger chain-like structures than the suspensions with lower particle loading. The highest concentration of the 8 CN-POSS at 8% showed the highest ER response at the 4 kV/mm electric field strength. The 1 CN-POSS suspension only showed a slight increase in shear stress, even in the on-state of 4 kV/mm.

Figure 2

Shear stress as a function of shear rate at 0 and 4 kV/mm for all concentrations.

The electric field-induced ER effect is shown in Figure a, which shows a plot of the 8% wt concentration of both ER suspensions. The shear stress of the 8 CN-POSS is seen to increase as the electric field is increased from 1 to 4 kV/mm. It is important to note the sharp decrease in the shear stress values of the 8 CN-POSS in the lower shear rate region of <2 (1/s) at the highest electric field of 4 kV/mm. The drop in the shear rate may be associated with an electric saturation of the suspension. This occurrence is mainly in the lower shear rate region where the hydrodynamic forces are weak, and there is sufficient particle-to-particle interaction and contact for electrostatic discharge to occur. The dielectric particles within the suspension, therefore, breakdown as the electric field strength is increased beyond its breakdown voltage, and it acts as an electrical conductor with electrostatic discharge. Thus, the yield stress for the suspensions at increased concentrations go through a maximum at 3 kV/mm and then drops.[4,13]

Figure 3

(a) Shear stress vs shear rate curves for varying electric fields of 8% wt suspensions and Herschel–Bulkley fit of curves. (b) Herschel–Bulkley yield stress calculated from flow curves at different electric fields for 8% wt suspensions.

The apparent yield stress of the suspensions of the ER particles from Figure b suggests that even in the absence of the electric field (0 kV/mm), the ER particles in the suspensions will arrange into a network-like structure with the fluid, resulting in gel-like viscoelastic behavior.[12] The yield stress is the stress required for flow to occur, and it depicts the strength of the solid network of particles in the suspension. The Herschel–Bulkley yield stress for the off-state and on-state of the varying electric field applied has been shown in Figure b and is calculated from the flow curves of the ER suspensions. The nonlinear Herschel–Bulkley model was used to fit the flow curve data.

The log–log dependence of the yield stress on the electric field strength obeys the power law from the conduction and polarization models.where q is the rigidity of the internal structures formed when the electric field is applied, and the value α should be within the range of 1.5–2 for well-developed structures for the conduction model and 2 for the polarization model.[10,21] The conduction model considers particle interaction only as the factor for the ER effect when the gap between the conducting particles in the fluid decreases and the α value approaches 1.5.[21] As shown in Figure b, τ scales as ∼E0.7 for the 8 CN-POSS and as small as E0.15 for 1 CN-POSS. The polarization model on the other hand relates the material parameters such as the dielectric constants of the liquid and solid particles to the rheological properties using an idealized ER system of uniform hard dielectric spheres dispersed in a Newtonian fluid medium and the yield stress is found to be proportional to the square of the applied electric field (i.e., α = 2).

Figure shows the ER efficiency dependence on the concentration at shear rates of 10 (1/s) and 100 (1/s). The ER efficiency measures the electroviscosity difference between the off-state when no field is applied and the on-state with the applied electric field.[14] It is given as efficiencyIn this case, the data presented in Figure is the difference in viscosity at 0 kV/mm, when no field was applied, and 4 kV/mm. This is an important parameter that can be used to analyze the performance of the ER fluid across the shear rate band.[6] The ER efficiency increases as the concentration increases for the 8 CN-POSS compared to the 1 CN-POSS, which is basically a flat line as shown in Figure for the 10 (1/s) and 100 (1/s). This is an indication that the 8 CN-POSS has a higher ER effect than the 1 CN-POSS as a result of the increased number of cyano functional groups attached to its cage structure. The increase in viscosity for the 8 CN-POSS at higher shear rates of 100 (1/s) for instance is not as great as the lower shear rates because the structures at high shear rates have been broken apart.

Figure 4

ER efficiency of 8 CN-POSS and 1 CN-POSS suspensions at shear rates of 10 (1/s) and 100 (1/s).

Linear Viscoelastic (LVE) Behavior of 8 CN-POSS and 1 CN-POSS

Research has shown that ER fluids behave as viscoelastic materials.[12] Oscillatory tests carried out to determine the viscoelastic properties of the ER suspension help provide a better understanding of the microstructural changes in the linear and nonlinear viscoelastic regions.

The complex shear moduli G′ and G″ were measured at varying electric field strengths at an increment of 1–4 kV/mm as shown in Figure . These measurements allow for the comparison of the structure within the suspensions. The viscoelastic network of the ER suspensions of 8 CN-POSS and 1 CN-POSS is shown in Figure . The storage and loss moduli increase as the electric field is increased with 8 CN-POSS showing a higher increase in the storage and loss modulus values across the strain tested. The linear viscoelastic (LVE) region is the range in which the test can be carried out without destroying the structure of the sample. The plateau in the lower strain area that characterizes this LVE region (seen in (top) of Figure at 1 and 2 kV/mm) can be seen to become shorter as the electric field is increased. The storage modulus becomes increasingly nonlinear as the electric field increases, and it can be assumed that the deformation occurring to the structures is not reversible. The behavior of the loss modulus at lower electric fields (Figure (bottom)) shows an internal structure that is characterized by a weak strain overshoot[15] called the “Type III behavior”, where we see a strain hardening followed by a strain thinning. Normalization curves of G′ and G″ will show a transition from Type III to Type I (strain thinning) where the loss modulus decreases as the electric field strength increases.

Figure 5

(Top) Storage modulus G′ vs strain and (bottom) Loss modulus vs strain. At increasing electric fields from 1 to 4 kV/mm for 8% wt concentration.

Figure shows the frequency sweep of the ER fluid with both the storage modulus in Figure a and the loss modulus in Figure b at 1% fixed strain. The frequency sweep was used to investigate the time-dependent deformation behavior of the suspension in the linear viscoelastic region. Both the storage and loss modulus showed an increase as the strength of the applied electric field is increased, indicating that the ER fluids show solid behavior.[22] The 8 CN-POSS showed a higher storage modulus than the 1 CN-POSS. There is an increase of the storage modulus at higher frequencies. At higher frequencies, the structures in the suspension exhibit less flexibility and higher rigidity at a faster motion, and the storage modulus values could increase up to five to ten times higher than the plateau region at a low frequency.[23]

Figure 6

Frequency sweep of 8% wt of 8 CN-POSS and 1 CN-POSS at 0 and 4 kV/mm for Storage modulus (a) and loss modulus; (b) at a fixed strain amplitude of 1%.

Conclusions

Our previous experiments established the fact that the 8 CN-POSS is electrorheologically responsive. The effect of the number of functional groups attached to the octahedral POSS geometry on the ER response has been shown in this work by comparing the 8 CN-POSS with a commercial variant 1 CN-POSS, through steady flow tests and oscillatory tests. Concentrations of 1, 4, 6, and 8% by weight were prepared for the ER suspensions. The electrorheological characterization and analysis such as the steady flow curves, Herschel–Bulkley yield stress, and ER efficiency indicate that the ER effect of the 8 CN-POSS is greater with the increasing concentration and electric field strength than the 1 CN-POSS. The power-law with respect to conduction and polarization models has been used to quantify the ER effect. The viscoelastic analysis of both ER suspensions also shows that the 8 CN-POSS has a greater ER effect. The polarizability of the particles in the suspension was investigated, and the 8 CN-POSS was found to have a higher dielectric constant. Adjusting the functional groups attached on cage structures could improve the ER properties of ER fluid designs, which can be used for various electrorheological applications.

Experimental Section

In our previous work,[4] the hydrolyzation route used by Dare et al.[8] was adopted to synthesize the 8 CN-POSS cage structure. The suspensions of both 1 CN-POSS, which is commercially available, and 8 CN-POSS have been compared in this work. Figure shows the structure of both 8 CN-POSS and 1 CN-POSS. The 500 cSt silicone oil was purchased from Sigma-Aldrich, and the 1 CN-POSS particles were purchased from Hybrid plastics.

Figure 7

Octafunctionalized and monofunctionalized cyanopropyl POSS structure.

Characterization of 8 CN-POSS and 1 CN-POSS

The microstructure and morphology of the synthesized 8 CN-POSS and commercially available 1 CN-POSS were observed using a scanning electron microscope (SEM).

The SEM images of 8 CN-POSS in Figure a show that the particles are micrometer-sized and are well dispersed in size and range from 1 to 200 μm, while the SEM images of 1 CN-POSS in Figure b show that the particles are also micrometer-sized and range in size from 5 to 100 μm and appear smoother compared to 8 CN-POSS.

Figure 8

(a) SEM images of 8 CN-POSS in the increasing order of magnification. (b) SEM images of 1 CN-POSS in the increasing order of magnification.

The attenuated total reflectance Fourier transform infrared (FTIR) spectra method via a Cary 630 FTIR instrument was used to identify the absorption bands and assign them to the bonds within the compound. According to FTIR spectra database and the literature,[9] for 8 CN-POSS; stretching vibrations for the −CN group was seen at 2245.8 cm–1, stretching vibration for the C–H group at 2939.8 cm–1 and bending vibration at 1406.0 cm–1, stretching vibration for the Si–C group at 747.4 cm–1, while 1096 cm–1 corresponds to the Si–O stretch. For 1 CN-POSS; the −CN group stretch was barely seen, C–H stretching was seen at 2952.3 cm–1, the Si–C stretch was seen at 739.4 cm–1, and the Si–O stretch at 1079.8 cm–1.

Preparation of Electrorheological Suspensions

The required quantities of 8 CN-POSS and 1 CN-POSS and silicone oil of 500 cSt viscosity were weighed and vacuum-dried at 130 °C for 24 h to ensure that traces of moisture that would influence the outcome of the results are eliminated. After drying, the POSS compounds were crushed to ensure a uniform particle size distribution. Contact with air was minimized by capping the containers, and concentrations of 1, 4, 6, and 8% by weight were prepared in separate vials. A homogenous mixture between the particles and the silicone oil was obtained using a vortex mixer.

Electrorheological Measurements

Rheological properties were measured with a strain-controlled Anton Paar MCR 302 parallel plate rheometer. A plate size of 25 mm was used and 0.3 mm spacing between plates in such a way that to achieve a 1 kV/mm electric field strength, a setting of 0.3 kV will have to be entered. The electric field was applied using the FuG DC power supply HCP 14-12500. The sample from the vial is slowly and uniformly dispersed onto the top of the bottom plate, and the required experimental settings are entered in the PC rheoplus software before the start of each test. Each test (rotational, oscillatory tests) was carried out with fresh samples because once a sample is used, the structure is broken down and takes time to rebuild, therefore the chances of having an accurate result with the same sample reduce. A shear rate sweep of 0.1–100 s–1 was done at fields of 0–4 kV/mm in 1 kV/mm incremental steps for all four concentrations prepared. This was done at a constant temperature (25 °C) to measure the flow curves. The yield stress of the suspensions was ascertained using the Herschel–Bulkley model. Oscillatory tests for strain amplitude were conducted at variable amplitudes or strain from a lower to a higher strain, while the frequency is held constant. Amplitude sweep was done from 0.1 to 1000% at a constant frequency of 10 rad/s at electric fields ranging from 0 to 4 kV/mm in steps of 1 kV/mm. The storage and loss moduli are derived from this experiment

Dielectric Spectroscopy of Suspension

The dielectric test was conducted at 25 °C over a range of 20 Hz to 1 MHz frequency measured by an IET 1920 precision LCR meter equipped with a liquid measurement cell, which has two metal electrodes separated by a Teflon spacer. The test was carried out with a 2% wt concentration of both compounds in silicone oil. The aluminum measuring cell was filled with the suspension, and the terminals from the LCR meter was connected to the ends of the cell. The capacitance and dissipation factor were recorded remotely on the PC. The parallel plate capacitor equationwas used to calculate the dielectric constant. The dielectric loss was also calculated.