Electro-responsive and dielectric characteristics of graphene sheets decorated with TiO₂ nanorods

Abstract

Titanium dioxide (TiO₂) nanorod-decorated graphene sheets have been synthesized by a simple non-hydrolytic sol–gel approach and demonstrated to be a highly effective dispersing material for electrorheological fluids. Electron microscopy and X-ray diffraction analysis indicated that the TiO₂ nanorods with a high crystallinity are well-dispersed and successfully anchored on the graphene sheet surface through the formation of covalent bonds between Ti and C atoms. Furthermore, electro-responsive properties of graphene sheets decorated with TiO₂ nanorods are investigated on the basis of the dielectric loss model. The dielectric property analysis based on the dielectric loss model clarifies that an introduction of TiO₂ nanorods has been coupled with the larger achievable polarizability and short relaxation time of interfacial polarization. Consequently, the TiO₂ nanorod-decorated graphene sheets exhibit consistently higher ER efficiency than that of GO sheets and show unprecedented electro-responsive performance in extremely low concentration ER materials (<0.2 wt%).

---

1 Introduction

Graphene, a two-dimensional single layer sheet of sp²-hybridized carbon, has attracted tremendous worldwide attention in recent years because of its unique properties, including a tunable bandgap, high carrier mobility, ambipolar field effect and quantum Hall effect.¹ In particular, there is now a growing interest in the use of graphene as an electrorheological (ER) material due to its several attractive features, such as its peculiar nanostructure, high surface area, and good mechanical strength.²

ER fluids, typically composed of polarizable solid particles in an insulating medium, undergo dramatic structural changes under an external electric field. Under the electric field strength, the dispersed particles are polarized and attracted to each other to form fibril structures along the external electric field direction, and these electrically induced structures vanish when the electric field strength is removed. ER fluids have received considerable attention due to their several attractive features, such as simple mechanics, low power consumption, rapid response time, and reversibility.

Recently, graphene has stimulated a great deal of interest as a good candidate for the reliable ER property. However, although there have been several recent reports describing the possible use of graphene as an ER material, many of these cases have resulted in insufficient shear strength and low colloidal stability, which could limit their practical or industrial applications.

For this reason, the need to modify the polarizability of graphene with a high relative dielectric constant has been highlighted. A convenient way to enhance the polarizability is to introduce a hybrid nanostructure that interconnects the graphene sheet with high dielectric materials. Recently, titanium dioxide (TiO₂) has been extensively studied because of its high dielectric constant, abundance, low cost, and structural stability against external stimulation.³ These features make TiO₂ particularly attractive for use in ER materials.

Herein, we have primarily focused on the fabrication of TiO₂ nanorod-decorated graphene sheets (TNGSs) by means of a simple non-hydrolytic sol–gel reaction and applied as a dispersing material for ER fluids. Graphene sheets with a large specific surface area could also potentially serve as a supporting material to anchor TiO₂ nanorods. The TiO₂ nanorods were introduced onto the graphene surface so as to decrease its current density and induce strong interfacial polarization. Furthermore, dielectric properties of the TNGSs and GO sheets were investigated to give deep insights into the primary role of TiO₂ nanorods that determine the electro-responsive characteristics.

2 Materials and methods

2.1 Materials

Graphite (flakes, <20 μm, synthetic), titanium(IV) chloride, oleylamine (70%) and toluene (99.5%) were purchased from Sigma-Aldrich. Other reagents including KMnO₄, HCl, H₂SO₄, H₂O₂ and ethanol were also obtained from Aldrich Chemical Co. All regents were used as received. For electrorheological (ER) fluid application, silicone oil (Aldrich, poly(methylphenyl-siloxane), viscosity = 100 cSt) was used as a dispersing medium.

2.2 Fabrication of graphene oxide (GO) sheets

GO was synthesized from graphite using a modified Hummers method. Typically, graphite (1.0 g) was added to 70 mL of H₂SO₄ in an ice bath, which was followed by the addition of KMnO₄ (3.0 g) and NaNO₃ (0.5 g). After stirring for 4 h, 70 mL of distilled water was slowly added and maintained at that temperature for 30 min. Subsequently, H₂O₂ solution was added to the solution until the color turned a brilliant brown indicating a fully oxidizing state. The as-prepared graphite oxide slurry was exfoliated to generate GO nanosheets by sonication using an ultrasonic generator (42 kHz, 100 W, Branson 3510, Branson Cleaning Equipment Co., Shelton, CT, USA) for 3 h. Finally, the mixture was separated by centrifugation, washed repeatedly with 5% HCl and distilled water, and dried in a vacuum oven at 40 °C for 24 h.

2.3 Synthesis of TiO₂ nanorod-decorated graphene sheets

TNGSs were fabricated via a non-hydrolytic sol–gel reaction. A variable amount of GO was dispersed in 30.5 mL of oleylamine solution. For a complete exfoliation of GO, the mixture was ultrasonicated for 3 h. Next, the exfoliated GO solution was transferred into a three-neck round-bottom flask and heated to 290 °C with nitrogen purging. A TiO₂ precursor (TiCl₄) was then injected into the solution under vigorous stirring. The weight ratio of GO/TiCl₄ was 5 wt%. The color of the solution changed from dark purple to yellow during the reaction, indicating that TiO₂ nanorods were formed by the non-hydrolytic sol–gel reaction. After 15 min, the reaction was quenched by the addition of 6 mL of toluene and then the reaction mixture was allowed to cool to room temperature. The resultant product was separated from the surfactants by centrifugation using an excess of acetone. The products were ultimately retrieved and allowed to dry in a vacuum oven for 12 h. Finally, the products were calcinated for 2 h at 500 °C under a N₂ atmosphere to achieve high crystallinity.

2.4 Characterization

The TEM images were taken with a JEOL EM-2000 EX II microscope. To observe TEM images, the TNGSs diluted in toluene were deposited on a carbon film coated copper grid. TNGSs were characterized by high-power X-ray diffraction (XRD, M18XHF-SRA (Mac Science Co.)) with a Cu Kα radiation source (λ = 1.5406 Å) at 40 kV and 300 mA (12 kW), and a field emission scanning electron microscope (FE-SEM, JEOL-6700) equipped with an energy dispersive X-ray spectrometer (EDS). X-ray photoelectron spectra (XPS) were obtained by a Sigma probe (Thermo).

2.5 Investigation of electrorheological properties

The ER properties were examined via a rheometer (AR 2000 Advanced Rheometer, TA Instruments) with a concentric cylinder conical geometry of 15 mm cup radius with a gap distance of 1.00 mm, a high-voltage generator (Trek 677B), and a temperature controller. To start a run, an ER fluid is placed between the cup and rotor, and DC voltage is applied. An electric field was applied for 3 min to obtain an equilibrium columnar structure before applying shear. All measurements were made at room temperature, and the shear rate was fixed to 0.1 s⁻¹.

3 Results and discussions

3.1 Fabrication of TiO₂ nanorod-decorated graphene sheets

The formation of TNGSs was confirmed by transmission electron microscopy (TEM) analysis. Fig. 1a and b showed large quantities of well-dispersed TiO₂ nanorods, with a dimension of 25 nm (length) × 4 nm (diameter) (L/D ratio ≈ 6), successfully decorated on the graphene surface. Interestingly, as-synthesized TiO₂ nanorods are highly anisotropic with the aspect ratio of ∼6. Under our experimental condition, the TiO₂ nano-rods have crystallographically anisotropic structures where surface energy can play a crucial factor. The TiO₂ nanorod has a {001} surface with higher energy. Because the growth rate is exponentially proportional to the surface energy under the kinetic growth process, the energy difference between the higher energy surface and other lower energy surfaces can promote preferential growth along the 〈001〉 directions of TiO₂.⁴

Fig. 1 (a) TEM image of overall TNGSs morphology. (b) Higher magnification of the boxed area in (a). (c) EDS mapping of carbon (C), titanium (Ti), and merged image (Ti + C) of TNGSs (scale bar: 1 μm). EDS analysis on the selected area of the TNGSs surface. (d) XRD patterns of TNGSs (red line indicates the crystalline phase of anatase TiO₂) (inset: deconvoluted XPS peak for TNGSs at the Ti2p core level).

In addition, decoration of the surface of graphene with TiO₂ nanorods was analyzed with elemental mapping by energy dispersive spectrometry (EDS) (Fig. 1c). The TNGSs were enriched with titanium (Ti) and carbon (C), indicating a uniform distribution of the TiO₂ nanorods throughout the whole graphene sheet. The EDS analysis also indicated the presence of C (5.57%), O (49.81%), and Ti (44.62%), which proved that the TiO₂ nanorods were successfully introduced on the carbonaceous graphene sheets. Interestingly, it is possible that the TiO₂ nanorods located at the surface of graphene could result in a relatively lower intensity of carbon content compared to the higher intensity of titanium content.

The TNGSs were further characterized by XRD and X-ray photoelectron spectroscopy (XPS). The TNGSs exhibited XRD peaks corresponding to the (101), (004), (200), (105), (211), (204), (116), (204) and (215) planes for the anatase structure of TiO₂ (Fig. 1d).⁵ The most predominant peak centered at 2θ = 26.5° was representative of the (101) anatase phase reflections with an interlayer spacing of 3.5 Å, namely, the diffraction pattern of TNGSs revealed the crystallinity and anatase phase of the TiO₂ nanorods. XPS analysis was utilized to investigate the hybridization of graphene and TiO₂ nanorods (Fig. 1d inset). In the case of the Ti2p XPS spectra in Fig. 1d, two bands were located at binding energies of 464.5 and 458.9 eV, which were assigned to Ti2p_1/2 and Ti2p_3/2, respectively. The deconvoluted Ti2p spectrum confirmed two low-intensity peaks centered at 465.8 and 460.2 eV, which were assigned to the Ti–C bond.⁶ Judging from these data, it can be concluded that TiO₂ nanorods with a high crystallinity were chemically anchored onto the graphene sheet surface through the non-hydrolytic sol–gel reaction.

3.2 Electrorheological activities of TiO₂ nanorod-decorated graphene sheets

Fig. 2a plots the flow curves of the shear stress versus shear rate for the ER fluid (both TNGSs and GO sheets 1 wt% in silicone oil). The shear stresses were measured as a function of shear rate (ŕ) under an applied electric field strength (3 kV mm⁻¹). Under an applied electric field, both TNGSs and GO sheet-based ER fluids show a shear stress plateau region for a broad range of shear rates, indicating typical Bingham plastic behavior. In this plateau region, the electrostatic interactions within ER materials and hydrodynamic force caused by shear flow are competing with each other.⁷ However, beyond the critical shear rate (ŕ_crit), the shear stress converged to the zero electric field strength value, giving a typical curve of Newtonian fluid behavior. This phenomenon can be attributed to the enhanced hydrodynamic force under high shear.⁸

Fig. 2 (a) Shear stress as a function of shear rate for 1 wt% of various graphene-based ER fluids under an applied electric field strength (3 kV mm⁻¹). (b) Dynamic yield stress as a function of weight fraction for various graphene-based ER fluids under 3 kV mm⁻¹ of electric field [inset: dynamic yield stress of various graphene-based ER fluids as a function of electric field strength (1 wt% in silicone oil)].

To gain an insight into the ER activity, the influence of the weight fraction (ω) of the ER material on the dynamic yield stress was evaluated by changing the weight fractions of ER fluids in the ω = 0–1 wt% range. Interestingly, the TNGSs represent typical Bingham fluid behavior in an extremely low concentration of ER materials (<0.2 wt%), which is an unprecedented result. The dynamic yield stress value increased with increasing weight fraction of ER materials at a fixed electric field strength. The dynamic yield stress value of TNGSs is about 40.6 Pa at 1 wt%, which is 3.9 times that of the same weight fraction of pristine GO sheets.

In addition, the relationship between the yield stress (τ_y) and the electric field strength is also investigated (Fig. 2b inset). If the ω value is constant, τ_y shows the dependence on the electric field strength. Similar to most ER fluids, the yield stress values of ER fluids are directly proportional to the increase in the applied electric field strength. In our experimental condition, both TNGSs and GO sheet-based ER fluid showed that τ_y is proportional to E₀² at low E₀ (<1 kV mm⁻¹) and typically approaches E₀^3/2 at high E₀ (>1 kV mm⁻¹). Even if the two different types of samples (TNGSs and GO sheets) had the same electro-responsive behavior, they displayed significantly different ER activity in terms of ER efficiency (defined as (τ_E − τ₀)/τ₀ × 100 or Δτ/τ₀ × 100, where τ_E is the shear stress with an applied electric field strength and τ₀ is the shear stress without an electric field strength). When the shear rate reached 10² s⁻¹, the TNGSs provided much higher ER efficiency than that of GO sheets and the corresponding ER efficiencies were about 554 and 113%, respectively.

To evaluate the real-time responses of the two different types of graphene-based ER fluids, an applied electric field (1 kV mm⁻¹) was alternately turned on and off, and their shear stress values were monitored in real time (Fig. 3a). When the electric field was applied, the shear stress values of both GO sheets and TNGSs increased instantaneously. In contrast, the shear stress values dropped rapidly back to their original level when the electric field was removed. This result indicates that the responses of the graphene-based ER fluids, upon the application of an electric field, are reversible and reproducible. However, the two different types of graphene-based ER fluids showed different response times (t_res) and recovery times (t_rec) (defined as the time required for the response or recovery from 0% to 90% of its final value). Under the electric field strength, the TNGSs-based ER fluid tends to react more quickly than GO sheets. It is noted that the TNGSs ensure very short response times closely related to their intrinsic characteristics of the decorated TiO₂ nanorods.

Fig. 3 (a) Effect of switching the applied electric field on the shear stress of various graphene-based ER fluids (1 wt% in silicone oil). (b) Microscope images of chain formation in a silicone oil suspension of TNGSs (1 wt% in silicone oil) without an electric field and with an applied electric field of 1 kV mm⁻¹. The gap distance between the two electrodes was 1.0 mm.

In addition, a microstructural transition of TNGSs-based ER fluid was observed using an optical microscope (OM) under an applied electric field as shown in Fig. 3b. Randomly dispersed TNGSs began to move rapidly toward the electrodes (within 100 milliseconds) and then formed a fibrillated structure along the applied electric field direction. This aligned fibrous structure, which is dominated by sufficient electrostatic interaction between TNGSs, provides rapid structure reformation under a shear force as well as a better resistance to the shear flow.

3.3 Dielectric properties of TiO₂ nanorod-decorated graphene sheets

A possible explanation for this difference in ER efficiency and response (recovery) time may reside with the dielectric properties of ER materials, which are closely related to their polarizability under an applied electric field. It has been reported that the ER property depends on the dielectric properties of the ER materials.⁹ In order to preliminarily understand the enhancement of ER efficiency, the dielectric properties of the two different types of samples are investigated. Fig. 4 presents the dielectric spectra (ε′ and ε′′) as a function of electric field frequency and Cole–Cole plots for the TNGSs and GO sheet-based ER fluids and their dielectric parameters are listed in Table 1. The Cole–Cole curve, which is a dielectric loss model for analyzing the dielectric characteristics of materials, has been introduced to describe the relationship between the ER activity and dielectric properties.¹⁰ Fitting lines in Fig. 4 are given by the following equation:

ε* = ε′ + iε′′ = ε_∞ + (ε₀ − ε_∞)/[1 + (iελ)1 − α], (0 ≤ α < 1) (1)

here, ε′ is the dielectric constant, ε′′ is the dielectric loss factor, ε₀ is the static permittivity (ƒ → 0), and ε_∞ is the fictitious permittivity (ƒ → ∞). It is widely accepted that good ER materials should have a large achievable polarizability (Δε = ε₀ − ε_∞) and short relaxation time of interfacial polarization (λ). The Δε values of GO sheets and TNGSs were 0.49 and 1.89, respectively. The Δε is related to the electrostatic interaction between ER materials. The higher Δε value of TNGSs indicates that it has an enhanced electrostatic interaction and can provide an improved ER property.

Fig. 4 (a) Permittivity (ε′) and loss factor (ε′′) as a function of frequency for various graphene-based ER fluids (1 wt%). Open and closed symbols indicate permittivity and loss factor, respectively. (b) Cole–Cole plots for various graphene-based ER fluids. The fitting lines are obtained from eqn (1) with parameters given in Table 1.

Table 1 Electrorheological values and dielectric parameters for various graphene-based ER fluids

Samples	Electrorheological value^a				Dielectric parameter^b
	τ y	τ E at 10^2 s^−1	τ 0	ER efficiency	ε 0	ε ∞	Δε	ƒ max	λ
a Electrorheological properties were acquired with a particle weight fraction of 1 wt% under 3 kV mm^−1 electric field strength. τE and τ0 values were obtained at a shear rate of 10^2 s^−1. b These values were measured by a Solartron SI 1260 Impedence/gain-phase analyzer with a Solartron 1296 dielectric interface. c The local frequency of the peak on the dielectric loss factor ε′′ and the fmax values were obtained by nonlinear regression analysis using OriginPro. d The relaxation time, denoted by λ = 1/2πfmax (fmax is the frequency of the loss peak).
TNGSs	40.62 Pa	65.72 Pa	10.05 Pa	553.93%	4.49	2.60	1.89	6.92 Hz	0.02 s
GO sheets	10.32 Pa	21.34 Pa	10.03 Pa	112.76%	3.01	2.52	0.49	0.89 Hz	0.18 s

---

Furthermore, the λ value is associated with the proper interfacial polarization response denoted by the relaxation time. The λ values, λ = 1/2πƒ_max, were 0.18 and 0.02 s for GO sheets and TNGSs based ER fluids (Table 1). In general, the relaxation time is connected with the proper interfacial polarization response.¹¹ The polarization rate of TNGSs might be faster than that of GO sheets under an electric field strength in terms of the dielectric loss model. Taking these results into account, it is concluded that the large achievable polarizability and the short relaxation time for interfacial polarization give a combined or synergistic contribution to superior ER activity of TNGSs.

4 Conclusion

In conclusion, titanium dioxide (TiO₂) nanorod-decorated graphene sheets have been synthesized by a simple non-hydrolytic sol–gel reaction and demonstrated to be a highly effective dispersing material for ER fluids. In particular, the TNGSs exhibited consistently higher ER efficiency than that of GO sheets and showed unprecedented electro-responsive performance in an extremely low concentration of ER materials (<0.2 wt%). Furthermore, the dielectric analysis based on the dielectric loss model clarified that the decoration of the graphene surface with TiO₂ nanorods has been coupled with the larger achievable polarizability and short relaxation time of interfacial polarization. Consequently, the synergistic contribution of both a large specific surface area of graphene sheets and high dielectric constant of TiO₂ nanorods has a strong influence on the ER activity, which provides an outstanding enhancement in the ER efficiency compared with the other research results of graphene-based ER materials.

Acknowledgements

This research was supported by the WCU (World Class University) program through the National Research Foundation of Korea funded by the Ministry of Education, Science and Technology (R31-10013).

Notes and references

1. (a) J.-Y. Hong and J. Jang, J. Mater. Chem., 2012, 22, 8179; (b) J.-Y. Hong, K.-Y. Shin, O. S. Kwon, H. Kang and J. Jang, Chem. Commun., 2011, 47, 7182; (c) D. Long, J.-Y. Hong, W. Li, J. Miyawaki, L. Ling, I. Mochida, S.-H. Yoon and J. Jang, ACS Nano, 2011, 5, 6254; (d) K.-Y. Shin, J.-Y. Hong, S. Lee and J. Jang, J. Mater. Chem., 2012, 22, 7871; (e) K.-Y. Shin, J.-Y. Hong and J. Jang, Chem. Commun., 2011, 47, 8527; (f) K.-Y. Shin, J.-Y. Hong and J. Jang, Adv. Mater., 2011, 23, 2113.
2. (a) J.-Y. Hong and J. Jang, Soft Matter, 2012, 8, 3348; (b) J. Yin, W. Wang, R. Chang and X. Zhao, Soft Matter, 2012, 8, 294; (c) W. L. Zhang, Y. D. Liu, H. J. Choi and S. G. Kim, ACS Appl. Mater. Interfaces, 2012, 4, 2267; (d) W. L. Zhang and H. J. Choi, Chem. Commun., 2011, 47, 12286; (e) H. Hu, X. Wang, J. Wang, F. Liu, M. Zhang and C. Xu, Appl. Surf. Sci., 2011, 257, 2637; (f) W. L. Zhang, Y. D. Liu and H. J. Choi, J. Mater. Chem., 2011, 21, 6916; (g) W. L. Zhang, B. J. Park and H. J. Choi, Chem. Commun., 2010, 46, 5596.
3. (a) J. Yin, X. Zhao, L. Xiang, X. Xia and Z. Zhang, Soft Matter, 2009, 5, 4687; (b) C. H. Hong, H. J. Choi and J. H. Kim, J. Mater. Sci., 2008, 43, 5702; (c) L. Xiang and X. Xhao, J. Colloid Interface Sci., 2006, 296, 131.
4. (a) B. Koo, J. Park, Y. Kim, S.-H. Choi, Y.-E. Sung and T. Hyeon, J. Phys. Chem. B, 2006, 110, 24318; (b) J.-H. Seo, Y.-W. Jun, S. J. Ko and J. Cheon, J. Phys. Chem. B, 2005, 109, 5389.
5. (a) E. Lee, J.-Y. Hong, H. Kang and J. Jang, J. Hazard. Mater., 2012, 219, 219; (b) S. H. Hwang, J. Song, Y. J. Jung, O. Y. Kweon, H. Song and J. Jang, Chem. Commun., 2011, 47, 9164; M. Choi, C. Kim, S. O. Jeon, K. S. Yook, J. Y. Lee and J. Jang, Chem. Commun., 2011, 47, 7092.
6. (a) O. Akhavan and E. Ghaderi, J. Phys. Chem., 2009, 47, 20214; (b) Y. Huang, W. Ho, S. Lee, L. Zhang, G. Li and J. C. Yu, Langmuir, 2008, 24, 3510.
7. (a) J.-Y. Hong and J. Jang, Soft Matter, 2010, 6, 4669; (b) J.-Y. Hong, E. Kwon and J. Jang, Soft Matter, 2009, 5, 951; (c) J.-Y. Hong, M. Choi, C. Kim and J. Jang, J. Colloid Interface Sci., 2010, 347, 177.
8. (a) S. Park, M. Cho, S. Lim, H. J. Choi and M. Jhon, Macromol. Chem. Phys., 2005, 26, 1563; (b) B. Wang and X. Zhao, Langmuir, 2005, 21, 6553.
9. (a) J. Yin, X. Zhao, L. Xiang, X. Xia and Z. Zhang, Soft Matter, 2009, 5, 4687; (b) P. Tan, J. Huang, D. Liu, W. Tian and L. Zhou, Soft Matter, 2010, 6, 4800.
10. (a) W. L. Zhang and H. J. Choi, Langmuir, 2012, 28, 7055; (b) W. Wen, X. Huang and P. Sheng, Soft Matter, 2008, 4, 200; (c) Y. G. Ko and U. S. Choi, Soft Matter, 2012, 8, 253.
11. (a) S. Park, M. Cho, S. Lim, H. J. Choi and M. Jhon, Macromol. Chem. Phys., 2005, 26, 1563; (b) B. Wang and X. Zhao, Langmuir, 2005, 21, 6553.

---