Suppression of Surface Charge on Micro- and Nano-structured Superhydrophobic Silicone Rubber

Zhipeng Yan1, Xidong Liang1

1Dept. of Electrical Engineering, State Key Lab of Power System, Tsinghua University, Beijing, 100084, China

and Ian Cotton2, Christopher Emersic2

2School of Electrical and Electronic Engineering, The University of Manchester, Manchester, UK

ABSTRACTLaser-ablated template and fluoroalkyl silane modified composite coatings are used to prepare a micro- and nano-structure on silicone rubber surfaces. By testing the surface potential accumulation and dissipation process, we assess the suppression of surface charge on superhydrophobic silicone rubber. The surface physicochemical properties and resistivity are examined. The electron and hole trap distribution are analysed using isothermal current decay theory. Peak density of traps associated with the superhydrophobic surfaces are found to be promoted to lower energy levels relative to unmodified silicone. Surface micro-structure increases trap density and reduce trap depth; nano coatings are shown to further reduce surface trap depth. The improvements shown to suppress surface charge accumulation is beneficial to a range of electrical industries and helps mitigate surface flashover on insulation.

Index Terms —Surface charge, micro-nano structure, silicone rubber, ICD

1 INTRODUCTIONIn modern power systems, the High Voltage Direct

Current (HVDC) system is considered the most efficient and economical technology for high-voltage, large-capacity, and long-distance transmission of electrical power. Polydimethylsiloxane (PDMS) based silicone rubber is widely used in HVDC as an outdoor insulation material because of its low surface energy, low density, and low surface and bulk conductivity over a wide temperature range [1]. Under DC voltages, the sustained electric field makes it easier for charge to accumulate on the insulator surface. The accumulated surface charge has a marked effect on insulation degradation and plays an important role during the development of surface flashover [2, 3].

Efforts have been made to suppress the surface charge accumulation by optimizing the insulator profile. In principle, increasing the surface conductivity of the insulator without influencing its bulk insulation would suppress the surface charge accumulation by allowing charge migration on the surface. Surface fluorination is an effective means to achieve this [4, 5]. The decrease in surface potential decay rate (increasing surface conductivity) with increasing fluorination time of epoxy resin insulators shows controllability of surface electrical properties [6, 7]. On silicone rubber, surface fluorination suppresses charge

accumulation and accelerates the decay of the surface charge [5, 8]. The presence of alumina trihydrate (ATH) fillers in silicone rubbers and ethylene propylene diene monomer (EPDM) affected the trap distribution and charging characteristics [9]. Other ways of modifying the trap depth have been achieved using dielectric barrier plasma discharges and chromium oxide (Cr2O3) coatings [10, 11]. Microscopic modification of surfaces is another possible method to prevent charge accumulation but no experimental work has so far shown this to be viable. Recent research, however, has investigated surface modification methods to reduce wettability for technological applications such as self-cleaning [12-16] and anti-icing [17, 18]. This has been achieved by modifying the geometric structure on both micro- and nano- scales [12].

In this paper, we investigate whether micro- and nano-surface structure can be used to suppress the accumulation of surface charge on insulators. Section 2 outlines the experimental equipment and process, section 3 provides a discussion of the results, and section 4 concludes.

Manuscript received on 5 September 2017, in final form 14 January 2018, accepted 19 January 2018. Corresponding author: Z. Yan.

2 EXPERIMENTAL METHOD2.1 PREPARATION OF SUPERHYDROPHOBIC

SILICONE RUBBER2.1.1 Micro-structured coating

Silicone rubber samples with micro structures were formed using a picosecond laser-ablated template. Micro craters were fabricated on an H13 mould steel using a high-power picosecond laser (Edgewave, Germany) incorporated with a high-speed scanning mirror (Scanlab HurrySCAN II 14). The laser energy density was 2 J cm-2 and the pulse frequency at a given location was set to 700 [16]. The laser beam passed through a scanning mirror and focused by a field lens producing a focal spot diameter of 30 µm. This template was used in the direct replication process of silicone samples in which the surface morphology of silicone rubber could be changed during vulcanization. Silicone rubber (purchased from Bluestar) with a molecular weight of 700000 g/mol was used. 40 phr (parts per hundreds of rubber) nano-SiO2 (purchased from Degussa) and 100 phr Al(OH)3 (purchased from Macklin) were added as fillers to improve the mechanical properties and flame retardance. The silicone rubber was placed on the H13 steel template within 24 h after mixing. The vulcanization temperature was 165°C, and the vulcanization pressure was 6 MPa [16, 19].2.1.2 Nano-structured coating

Hydrophobic nano-silica particles were modified by fluorinated alkyl silane (FAS). After freeze-drying, 0.2 g of silica particles were dispersed in 10 mL of tetrahydrofuran (THF) with the aid of ultrasonication. FAS (0.01 g) and PDMS-A (Bluestar elastomer, 0.25 g) were then added to the solution. The solution was ultrasonicated for 1 hour to form what we refer to henceforth as ‘Solution A’. PDMS-B (Bluestar elastomer, 0.25 g) was dissolved into THF (10 ml) to form ‘Solution B’. Prior to coating, Solutions A and B were mixed together at room temperature to form a final coating solution. Flat silicone rubber specimens (30 × 30 mm2) were coated with the final coating solution (400 µl) and then cured at 165°C for 10 min [20].2.1.3 Micro-nano-structured coatings

The same coating solutions were used to fabricate micro-nano hierarchical structured surfaces. Solutions A and B used for the nano-structured coating (section 2.1.2) were mixed and then 400 µl was applied to the micro structured silicone rubber (30 × 30 mm2) prepared by the template method (section 2.1.1). The sample was cured at 165°C for 10 minutes, similarly to the fabrication of the nano-structured coating.

2.2 SURFACE CHARGE DECAY MEASUREMENTSurface charging and decay measurements were

performed in a grounded stainless-steel chamber. The ambient temperature was 25°C, and a temperature of 45°C was used to measure attempt-to-escape frequency of electrons. The relative humidity was 30%. The equipment was divided into two parts: corona charge and potential

measurement (Figure 1). The specimen was charged with a needle electrode and charge evenly distribute on the surfaces using a grid plane electrode. The specimen and electrodes were separated by polyethylene. Two different grid voltage of ±1.5 kV and ±3 kV were used and the charging time was 60 s. The specimen and ground electrode were translated immediately after charging to measure its surface potential with a non-contacting probe. The measurement was conducted by a Kelvin-type probe (Trek 6000B-5C, USA) linked to an electrostatic voltmeter (Trek 347, USA).

Figure 1. Diagram of experimental setup for corona charging and surface potential measurement.

2.3 CHARACTERIZATION OF PHYSICAL AND CHEMICAL PROPERTIES

Drop static contact angles were measured using a DataPhysics OCA15 Pro drop dispenser and goniometry instrument. 4 µl drops of deionized water were used and static contact angle was measured using the supplied goniometry software’s auto elliptical fitting method. Drops on eight locations were tested for each surface, and the average static contact angle was calculated. Sliding angle was measured by slowly tilting the sample until the 4 µl drop slid off.

The surface microtopography of the silicone rubber was analyzed by scanning electron microscopy (TESCAN MIRA3, German). Surface chemical composition of the superhydrophobic surfaces, together with the original specimen as a reference, were conducted by a Fourier Transform Infrared (FTIR) spectrometer (Nicolet 6700, USA) with an attenuated total reflection infrared (ATR-IR) analysis.

3 RESULTS AND DISCUSSION3.1 SUPERHYDROPHOBIC PROPERTIES OF

SILICONE RUBBERMicro- and nano-structures on the surface of

silicone samples were prepared to measure and compare the charge accumulation and dissipation characteristics. Scanning electron microscope (SEM) images of the sample surfaces were obtained. An unmodified sample

is shown in Figure 2a and is shown to have a very low surface roughness. The nano-structured sample (Figure 2b) shows the nano-scale surface roughness is achieved by the dense packing of the silica particles due to the crosslinking and adhesive effect of polydimethylsiloxane (PDMS). The agglomeration of nanoparticle causes the increase in size of the fluoroalkyl silane modified nano-silica of the coatings. Most nanoparticles are distributed in size range from 100 to 200 nm. The micro papilla structure prepared by the template method (section 2.1.1) shows a typical conical shape (Figure 2c, d), similar to the morphology of a lotus leaf surface [21]. The conical base diameter ranges from 34 to 36 µm. The interval between each cone is approximately 40 µm and their height is approximately 60 µm. For the micro-nano structured surfaces, the nanoparticles are densely distributed on the micro-scale conical surface structure. The combination of nanoparticles and micro pillars form a uniform hierarchical structure (Figure 2e, f).

Figure 2. SEM of the original and superhydrophobic specimens, (a) Original, (b) nano, (c, d) micro, (e, f) micro-nano silicone rubber.

Figure 3a shows the result of static contact angle and sliding angle on the different micro, nano, and micro-nano hierarchically textured surfaces. The static contact angle of water drops on the unaltered silicone sample was 115±0.7° (classified as hydrophobic). For micro structured silicone surfaces, this increases to 151.1±1.7°, with a sliding angle of 4.1°. Nano-structured surfaces offered a static contact angle of 148.2° and sliding angle of 4.9°. The micro-nano structured samples offered the highest static contact angle of 153.3° and very low sliding angle of 2.7°.

The difference in surface structure results in different surface contact angles [16]. Figure 3b, c shows water does not appear to be penetrating into the surface papilla microstructure due to the presence of air pockets between the solid-liquid interfaces, as observed by other researchers [22]. At the same time, increasing surface roughness by micro and nano homogeneous interface results in the increasing of actual contact area and the projected area [23].

Figure 3. Contact angle and sliding angle of the original and superhydrophobic specimens. (a) contact angle and sliding angle, (b)

contact angle of specimen micro-nano, (c) air pockets between water and specimen micro-nano.

3.2 SURFACE CHARGE MEASUREMENTSFigure 4 shows the results of the absolute

value of surface potential |Us| on both the original and superhydrophobic silicone samples as a function of grid voltages. At negative voltages, the surface charge carriers are electrons; conversely, at positive voltages, the absence of electrons (‘holes’) are the principal carrier. The presence of the grid electrode distributes the corona discharge evenly across samples allowing effective dissipation of electrons and holes [24]. The removal of the external electric field allows the field from the surface charge to form, resulting in directional migration of charge [2, 25, 26]. The initial

surface potentials can be regarded as the equilibrium surface potentials of the specimens during the charging process.

The decay time of surface potential on the untreated silicone surface decays much more slowly than with treatment (Figure 4a). Positive surface charge was observed to decay more rapidly than negative and this appears to be independent of initial magnitude. The micro-structured sample showed approximately 2–3 times faster rate of surface charge decay (Figure 4b). The nano, and particularly, the micro-nano-structured surfaces had decay times over an order of magnitude quicker than the untreated sample (Figure 4c, d), indicating the surface microstructure offers significant surface charge suppression capability. The

surface patterning does not affect the bulk or surface conductivity via chemical changes and is purely a geometric alteration and the primary source of this behaviour.

The dissipation process for positive and negative surface charge is normalized in Figure 5 to compare the contribution of the surface structure to the surface potential decay. For both positive and negative charge, the rate of decay for micro-nano-structured surfaces is greatest for the first 600 s. However, after this time, the rate of decay reduces and is lower than the nano-structured surface. This indicates that multiple factors are likely operating at different geometric scales, and this is considered in more detail in section 3.4.

Figure 4. Surface potential decay plots; (a) original, (b) micro, (c) nano, (d) micro-nano silicone rubber.

3.3 SURFACE CHARACTERISTICSThe chemical composition of the original and

superhydrophobic specimens has been investigated by attenuated total reflection infrared (ATR-IR) analysis. The position and peak of the unmodified silicone sample is identical to the micro-structured sample (Figure 6a, b). Absorption occurs around 2962 cm-1 and is characteristic of an aliphatic C-H stretch in

CH3. Other absorptions between 1390–1440 cm-1 are due to an asymmetric CH3 deformation of Si-CH3. The absorption at 1259 cm-1 is cause by a symmetric CH3 deformation of Si-CH3, at 850–870 cm-1 due to a Si-alkyl group such as Si-(CH3)3, and the intense absorption around 1007 cm-1 is characteristic of asymmetric Si-O-Si stretching vibration [27].

Figure 5. Surface charge decay of electron and hole. Electron with grid voltage (a)-1.5kV, (b) -3kV, hole with grid voltage (c)+1.5kV, (d) +3kV.

The measurements show that producing a micro-structure does not modify the chemical composition of the surface. After modification with the nanoparticles, new peaks at 1201 cm-1

and 1147 cm-1 occurred (Fig. 6c, d), which are assigned to the C-F stretching vibrations [20], as the introduction of a silica/PDMS/FAS coating solution lead to the formation of C-F on the coating surface, and the Al(OH)3 filler in the rubber bulk is simultaneously shielded due to the formation of the coating. The three absorption peaks between 3454 and 3618 cm-1

in the original and micro-structured samples are due to the O-H group in the alumina trihydrate (ATH) fillers [28]. These groups are subsequently lost in the nano-structured surfaces. Figure 6. Fourier Transform Infrared (FTIR) analyses; (a) original, (b)

micro, (c) nano, (d) micro-nano silicone.

Surface charge on an insulating material can dissipate in three principal ways: along the surface, through the bulk of the material, and into the surrounding air. Dissipation along the surface is related to the surface resistivity of the material and dissipation through the bulk of the material is related to the volume resistivity. All surface variations of test sample showed a similar volume resistivity of approximately 2×1015 Ω cm (Figure 7), but variability in surface resistivity was greater, ranging from 1.28×1013 to 1.65×1016 Ω. The unmodified surface has the largest surface resistivity and the nano-structured surface has the lowest in

accordance with surface potential decay laws. The formation of superhydrophobic structures does not change surface chemical properties and has no influence on the bulk conductivity; charge conduction through the bulk does not occur or contribute to the difference in charge dissipation rates for each surface type. The lower surface resistivity of the nano-structured surface can account for the shorter decay times.

3.4 ELECTRON AND HOLE TRAP ENERGY DISTRIBUTION

In general, the accumulation and dissipation of macromolecules are affected by surface traps. Charge traps inside a polymer are formed by physical defects due to entanglements, kinks, and chemical defects such as impurities, oxidation products, and additives of macromolecular chains. These have significant influence on charge carrier trapping, de-trapping, transport, recombination and space charge formation etc. [29].

Figure 7. Surface and volume resistivity of samples

Based on isothermal current decay (ICD) theory [30, 31], the trap energy distribution can be produced in the isothermal current decay process. Solid in an ideal excitation state have a portion of the traps above the equilibrium Fermi level EF0 containing electrons and a portion below EF0 containing holes [31]. If the electric field is sufficiently high that an electron and hole can de-trap after the excitation source is removed, the rate of emission of electrons to the conduction band

can be determined:

(1)

(2)

where N(E) is the energy distribution of the trap levels. en(E,T) is the probability per unit time of

a trapped electron being emitted to the conduction band from a trap level at energy E. f0(E) is the initial excited state of the traps by electrons which can be simplified with a nominal value of 0.5 [31]. The value of en is directly proportional to the attempt-to-escape frequency v in Eq. (2). Ec is the energy of the conduction band edge and k is the Boltzmann constant.

By integrating from Ei to the conduction band Ec and value band Ev separately, the external current of electrons In and holes Ip can be determined by the contribution of all the traps in half of the band gap:

(3)

(4)

where L is the thickness of the specimen. In order to simplify, the weighted contribution of any electron-emitting level to the current G(E,t) is defined by [31]:

(5)

Em can be obtained by at E=Em. At different test temperatures, the attempt-to-escape frequency of electrons v can be determined from Eq. (7).

(6)

(7)

By approximating Gn(E,t) as , substituting the

simplified Gn(E,t) into Eq. (3) and Eq. (4), the relationship between the external current of electron In and hole Ip and the energy distribution of the trap levels can be obtained.

(8)

(9)

Considering the relationship between the de-trapping process and surface charge decay; double injection takes place under a high DC electric field based on the profile of space charges on the material. According to this phenomenon, a simplified charge distribution model inside a corona-charged sample is presented in [24, 32] and the following equation can be derived:

(5)

where δ is the thickness of the top charge layer, usually set as 2 µm [24]. By substituting Eq. (10) to Eq. (8) and (9), the electron trap density can be calculated by the surface potential measurement.

Carries trap energy distributions for the four different samples can be determined from isothermal current decay theory. The electron and hole trap energy distributions are shown in Figure 8; the traps of each sample are similar in shape and contain energy density peaks within the measurement range.

The energy levels of both electron and hole peak trap depth of the superhydrophobic samples are lower than that of the unmodified sample. For electron traps, the peak density is located at 0.840, 0.774 and 0.756 eV for micro-, nano-, and micro-nano-structured samples, respectively, and above 0.879 eV for the unmodified sample (Table 2). Hole traps show a similar trend; the deepest hole trap at the highest energy levels is on the unmodified sample and the shallowest is on the micro-nano-structured sample. The energy level of the peak hole trap density is lower than that of the electron traps for the micro-nano-structured and unmodified samples but similar for the nano- and micro-structured samples. The electron trap density on the micro-structured sample is greater than hole trap density, but the other sample types are similar.

Trap depth and density are closely related to the surface structure. The micro structuring appears to make the trap density increase but reduce the peak’s energy level. Two possible explanations may account for this. Firstly, the surface structure formed by the template method shown in Figure 2 increases the surface area of the samples. This is equivalent to introducing a large number of physical interfaces which generates shallow traps and increases the surface trap density [10]. Secondly, the formation of surface micro-structures increases the local surface electric field. This makes charge removal more energetically favourable so that the depth of the charge traps becomes shallow and the charge dissipation rate increases.

Figure 8. Carries trap energy distributions for the samples; (a) electron trap, (b) hole trap.

Nano-structured coatings result in reduced energy of peak trap density. It can be seen in the FTIR analysis in section 3.3 that although new peaks appear at 1201 cm-1 for C-F stretching vibrations, the formation of the coating cause the loss of peaks around 3400 cm-1 for O-H group. The O-H group has high molecular polarity and is the important component of chemical defects and a likely contribution to the reduction of deep traps [33].

Table 2. Carries trap energy parameters of the original and superhydrophobic specimens.

Carries trap energy parameters Original Micro Nano Micro-nano

Electron Trap

Peak Energy Ep

(eV) ~0.879 0.840 0.774 0.756

Trap Density N(E) (1021 eV-1 m3) ~4.564 5.171 4.085 3.059

Hole Trap

Peak Energy Ep

(eV) 0.871 0.838 0.776 0.756

Trap Density N(E) (1021 eV-1 m3) 3.443 4.311 3.329 3.009

The coatings with inorganic nanoparticle cause the formation of amorphous regions on the surfaces. For macromolecules, electron conduction states have an inter-chain characteristic rather than intra-chain [34, 35]. In other words, an electron added to composites will be repelled by macromolecular chains and localized between the chains in the amorphous regions. More amorphous space and less condensed molecular conformation on

the surface can form more shallow traps.4 CONCLUSIONS

This work has shown a significant suppression of accumulated surface charges on micro- and nano-structured superhydrophobic silicone rubber surfaces. The results of isothermal surface potential decay measurements indicate that deposited corona charge will be suppressed by the micro structured surfaces and cannot be stored on the nano structured surfaces. The SEM, ATR-IR, and resistivity measurements show that the surface modification only changes the surface structure and chemical composition rather than the bulk.

Using isothermal current decay theory, the electron and hole traps are examined, revealing the traps in superhydrophobic surfaces are shallower than those in unmodified silicone. Micro-structured surfaces make the trap density increase and the energy of peak trap density reduce, whereas the nano-structure results in further lowering of the energy of peak trap density.

Micro- and nano-structuring surfaces to achieve superhydrophobicity is shown here to reduce surface charge accumulation on insulators. Simultaneously, the superhydrophobic surfaces offer other advantages, including liquid pollution-repellence and anti-icing properties beneficial to a range of power industries.

APPENDIXThis study was supported in part by the Independent

Project of State Key Laboratory.

REFERENCES[1] R. Hackam, "Outdoor HV composite polymeric insulators," IEEE

Trans. Dielectr. Electr. Insul., vol. 6, pp. 557-585, Oct, 1999.[2] A. Winter and J. Kindersberger, "Stationary resistive field

distribution along epoxy resin insulators in air under DC voltage," IEEE Trans. Dielectr. Electr. Insul., vol. 19, pp. 1732-1739, Oct,, 2012.

[3] X. Jun and I. D. Chalmers, "The influence of surface charge upon flash-over of particle-contaminated insulators in under impulse-voltage conditions," Journal of Physics D: Applied Physics, vol. 30, p. 1055, 1997.

[4] B. Zhang, G. Zhang, Q. Wang, C. Li, J. He, and Z. An, "Suppression of surface charge accumulation on Al2O3-filled epoxy resin insulator under dc voltage by direct fluorination," AIP Advances, vol. 5, p. 127207, 2015.

[5] B. X. Du, Z. L. Li and J. Li, "Surface charge accumulation and decay of direct-fluorinated RTV silicone rubber," IEEE Trans. Dielectr. Electr. Insul., vol. 21, pp. 2338-2342, Oct, 2014.

[6] Y. Liu, Z. An, J. Cang, Y. Zhang, and F. Zheng, "Significant suppression of surface charge accumulation on epoxy resin by direct fluorination," IEEE Trans. Dielectr. Electr. Insul., vol. 19, Aug, 2012.

[7] Y. Liu, Z. An, Q. Yin, F. Zheng, Y. Zhang, and Q. Lei, "Rapid potential decay on surface fluorinated epoxy resin samples," Journal of Applied Physics, vol. 113, p. 164105, 2013.

[8] B. X. Du and Z. L. Li, "Hydrophobicity, surface charge and DC flashover characteristics of direct-fluorinated RTV silicone rubber," IEEE Trans. Dielectr. Electr. Insul., vol. 22, pp. 934-940, Apr, 2015.

[9] H. Sjostedt, S. M. Gubanski and Y. V. Serdyuk, "Charging characteristics of EPDM and silicone rubbers deduced from surface

potential measurements," IEEE Trans. Dielectr. Electr. Insul., vol. 16, pp. 696 – 703,Jun, 2009.

[10] C. Li, J. Hu, C. Lin, and J. He, "The control mechanism of surface traps on surface charge behavior in alumina-filled epoxy composites," Journal of Physics D: Applied Physics, vol. 49, p. 445304, 2016.

[11] C. Li, J. Hu, C. Lin, B. Zhang, G. Zhang, and J. He, "Surface charge migration and dc surface flashover of surface-modified epoxy-based insulators," Journal of Physics D: Applied Physics, vol. 50, p. 065301, 2017.

[12] X. Zhang, F. Shi, J. Niu, Y. Jiang, and Z. Wang, "Superhydrophobic surfaces: from structural control to functional application," J. Mater. Chem., vol. 18, pp. 621-633, 2008.

[13] P. Zylka, "On the surface performance of superhydrophobic silicone rubber specimens fabricated by direct replica method," IEEE Trans. Dielectr. Electr. Insul., vol. 21, pp. 1183-1188, Jun, 2014.

[14] J. Li, Y. Wei, Z. Huang, F. Wang, and X. Yan, "Investigation of the electric field driven self-propelled motion of water droplets on a super-hydrophobic surface," IEEE Trans. Dielectr. Electr. Insul., vol. 23, pp. 3007-3015, Nov, 2016.

[15] J. Li, Y. Wei, Z. Huang, F. Wang, X. Yan, and Z. Wu, "Electrohydrodynamic behavior of water droplets on a horizontal super hydrophobic surface and its self-cleaning application," Applied Surface Science, vol. 403, pp. 133-140, 2017.

[16] Z. Yan, X. Liang, H. Shen, and Y. Liu, "Preparation and basic properties of superhydrophobic silicone rubber with micro-nano hierarchical structures formed by picosecond laser-ablated template," IEEE Trans. Dielectr. Electr. Insul., vol. 24, pp. 1743-1750, 2017.

[17] S. Farhadi, M. Farzaneh and S. A. Kulinich, "Anti-icing performance of superhydrophobic surfaces," Applied Surface Science, vol. 257, pp. 6264-6269, 2011.

[18] F. Arianpour, M. Farzaneh and S. A. Kulinich, "Hydrophobic and ice-retarding properties of doped silicone rubber coatings," Applied Surface Science, vol. 265, pp. 546-552, 2013.

[19] Z. Yan, X. Liang, Y. Gao, and Y. Liu, "Aging and self-healing properties of superhydrophobic silicone rubber," IEEE Trans. Dielectr. Electr. Insul., vol. 23, pp. 3531-3538, Dec, 2016.

[20] H. Zhou, H. Wang, H. Niu, A. Gestos, X. Wang, and T. Lin, "Fluoroalkyl silane modified silicone rubber/nanoparticle composite: A super durable, robust superhydrophobic fabric coating," Advanced Materials, vol. 24, pp. 2409-2412, 2012.

[21] L. Chen, Z. Xiao, P. C. Chan, Y. Lee, and Z. Li, "A comparative study of droplet impact dynamics on a dual-scaled superhydrophobic surface and lotus leaf," Applied surface science, vol. 257, pp. 8857-8863, 2011.

[22] T. Deng, K. K. Varanasi, M. Hsu, N. Bhate, C. Keimel, J. Stein, and M. Blohm, "Nonwetting of impinging droplets on textured surfaces," Applied Physics Letters, vol. 94, p. 133109, 2009.

[23] Y. Su, B. Ji, K. Zhang, H. Gao, Y. Huang, and K. Hwang, "Nano to micro structural hierarchy is crucial for stable superhydrophobic and water-repellent surfaces," Langmuir, vol. 26, pp. 4984-4989, 2010.

[24] W. Shen, H. Mu, G. Zhang, J. Deng, and D. Tu, "Identification of electron and hole trap based on isothermal surface potential decay model," Journal of Applied Physics, vol. 113, p. 083706, 2013.

[25] J. Kindersberger and C. Lederle, "Surface charge decay on insulators in air and sulfurhexafluorid-Part I: Simulation," IEEE Trans. Dielectr. Electr. Insul., vol. 15, pp. 941-948, Aug, 2008.

[26] A. Winter and J. Kindersberger, "Transient field distribution in gas-solid insulation systems under DC voltages," IEEE Trans. Dielectr. Electr. Insul., vol. 21, pp. 116-128, Feb, 2014.

[27] A. Ghanbari-Siahkali, S. Mitra, P. Kingshott, K. Almdal, C. Bloch, and H. K. Rehmeier, "Investigation of the hydrothermal stability of cross-linked liquid silicone rubber (LSR)," Polymer degradation and Stability, vol. 90, pp. 471-480, 2005.

[28] Y. Gao, J. Wang, X. Liang, Z. Yan, Y. Liu, and Y. Cai, "Investigation on permeation properties of liquids into HTV silicone rubber materials," IEEE Trans. Dielectr. Electr. Insul., vol. 21, pp. 2428-2437, 2014.

[29] T. J. Lewis, "Charge transport, charge injection and breakdown in polymeric insulators," Journal of Physics D: Applied Physics, vol. 23, p. 1469, 1990.

[30] J. G. Simmons and G. W. Taylor, "High-field isothermal currents and thermally stimulated currents in insulators having discrete trapping levels," Physical Review B, vol. 5, p. 1619, 1972.

[31] J. G. Simmons and M. C. Tam, "Theory of isothermal currents and the direct determination of trap parameters in semiconductors and insulators containing arbitrary trap distributions," Physical Review B, vol. 7, p. 3706, 1973.

[32] P. Molinié, "Measuring and modeling transient insulator response to charging: the contribution of surface potential studies," IEEE Trans. Dielectr. Electr. Insul., vol. 12, pp. 939-950, Oct, 2005.

[33] G. Teyssedre and C. Laurent, "Charge transport modeling in insulating polymers: from molecular to macroscopic scale," IEEE Trans. Dielectr. Electr. Insul., vol. 12, pp. 857-875, Oct, 2005.

[34] M. Meunier and N. Quirke, "Molecular modeling of electron trapping in polymer insulators," The Journal of Chemical Physics, vol. 113, pp. 369-376, 2000.

[35] S. Serra, E. Tosatti, S. Iarlori, S. Scandolo, and G. Santoro, "Interchain electron states in polyethylene," Physical Review B, vol. 62, p. 4389, 2000.

Zhipeng Yan was born in China on 4 July 1990. He received his B.S. degree in 2013 from the Dept. of Electrical Engineering at Tsinghua University in Beijing. He is going to work toward the Ph.D. degree in high voltage engineering and insulation at Tsinghua University. His major research directions are superhydrophobic silicone rubber materials and aging properties of silicone rubber.

Prof. Xidong Liang was born in China, on 17 December 1962. He received the B.E., M.E., and Ph.D. degrees in high voltage engineering from Tsinghua University, Beijing, China in 1984, 1987 and 1991, respectively. He did his research in UMIST, England as a visiting scholar from 1991 to 1992. Then he was employed in Tsinghua University as an associate professor from 1993 and as a full

professor from 1997. He is a member of working groups of IEC and CIGRÉ dealing with composite insulators. His special fields of interest include outdoor insulation, composite insulator and its materials.

Prof Ian Cotton received a Class I B. Eng. (Hons.) degree in electrical engineering from the University of Sheffield, Sheffield, U.K., in 1995 and the Ph.D. degree in electrical engineering from the University of Manchester, Institute of Technology (UMIST), Manchester, U.K., in 1998. He is currently a Professor of High Voltage Technology at the University of Manchester and the Director of Manchester Energy. His main research interests include power systems transients, the use of higher voltage systems in aerospace applications and power system induced corrosion.

Dr Christopher Emersic was born in the United Kingdom in 1981. He received the MPhys degree from The University of Manchester Institute of Science and Technology (UMIST), UK, in 2003, and Ph.D. from The University of Manchester, UK in 2006. He has worked as a research associate at the University of Oklahoma, USA, New Mexico Tech, USA, and The University of Manchester, UK. His research interests include thunderstorm electrification and atmospheric electricity, cloud physics, and power electronics.