Understanding the Water Droplet Initiated Discharges on Gamma Irradiated Silicone Rubber Insulation.
Recently, silicone rubber is gaining importance for use in outdoor insulators and high energy radiation zones. They offer properties such as excellent electrical insulation, better chemical resistance, low toxicity, and so on. Also their low surface energy aids in maintaining a good hydrophobic surface, better pollution performance in outdoor service condition, low weight, easy handling, vandal resistance and improved seismic performance, better flexibility in design, and cost-effectiveness [1-3].
The high energy radiation can alter the crosslinking density, characteristic variation in surface properties, and partial discharge resistance of the material [4, 5]. The outdoor insulating material should be hydrophobic and it is essential to understand the effect of radiation dosage on variation in contact angle of the material . It has been reported that the dose rate in nuclear power plant varies widely from 10 [micro]Gy/h to 10 kGy/h with a total exposure of 1,000 kGy , The water droplets that form on the insulator surface due to rain or condensation of fog can cause field enhancement thereby generating corona followed by surface discharge activity. The inception of corona/surface discharges can cause reduction in hydrophobicity of the silicone rubber material, carbonization of surface, and finally, bridging of electrode gap leading to catastrophic failure of the insulation structure. Nagesh and Sarathi  studied water droplet-initiated discharges on epoxy nanocomposites adopting the UHF technique. They concluded that current pulses which are initiated by incipient discharges are formed due to water droplets under alternating voltages. These pulses have risen times of a few nanoseconds and radiate UHF signals. It is also essential to understand the characteristics of UHF signal formed due to multiple water droplets and to quantify the surface damage. Nazemi and Hinrichsen studied variation in water droplet waveshape under AC voltage stress and identified four modes of droplet movements . Recently Sarathi et al. studied the variation in droplet shape on epoxy nanocomposites under AC voltage stress and concluded that mode 1 type movement is predominant . It is vital to understand the variation in mode of water droplet movement with gamma irradiated silicone rubber specimen.
Murray et al. studied the influence of gamma ray irradiation and electron beam irradiation of polymer material, and observed that gamma radiation exposure leads to more discoloration and increased surface roughness .The surface degradation due to ageing/irradiation can allow charges to get deposited on the surface of the insulating material thereby causing local electrical field variation thereby causing surface discharges. Therefore, it is essential to understand the charge accumulation characteristics of unaged and gamma irradiated specimen. In addition, a fingerprint needs to be obtained to understand the phase at which the discharges occur. An attempt has been made in the present work to obtain phase resolved-partial discharge (PRPD) relationship using the UHF sensor-generated signal during discharges initiated by the water droplet as an input to spectrum analyzer (by operating it in zero span mode).
More importantly, to correlate the observed charge accumulation and discharge behavior with the dosage of gamma radiation to the silicone rubber sample, a thorough analysis of the variation in structural characteristics of the material caused by gamma radiation is vital. The objectives of this study are multifold: (i) to evaluate the variation in corona inception voltage due to aqueous N[H.sub.4]Cl droplets of different conductivity on gamma irradiated silicone rubber under AC voltages, (ii) to understand the characteristics of UHF signals radiated due to water droplet initiated discharges on gamma irradiated insulating material at the time of corona inception and arcing under AC voltages, (iii) to analyze surface charge variation on silicone rubber insulating material with water droplet, (iv) to obtain PRPD analysis at the time of corona inception and arcing, and (v) to evaluate the surface structural changes caused by gamma irradiation using analytical pyrolysis-gas chromatography/mass spectrometry (Py-GC/MS) technique and qualitatively assess their impacts on charge accumulation and discharge characteristics.
The experimental setup used to evaluate the corona discharge activity due to a water droplet on silicone rubber insulating material under AC voltages is shown in Fig. 1. The experimental setup contains three major parts: (i) high voltage source, (ii) test electrode arrangement, and (iii) UHF sensor connected to the high bandwidth digital storage oscilloscope and spectrum analyzer.
The high AC voltages were generated by a 100 kV, 5 kVA discharge-free test transformer with an internal PD level <5 pC at 50 kV. The voltage was increased at a rate of 300 V/s up to the required test voltage.
Test Electrode Arrangement
The electrode gap was made up of two angular stainless steel electrode tips cut at 45[degrees] (as per IEC 60112 ) and placed on silicone rubber. The electrodes were separated by a distance of 10 mm. One electrode was connected to the high voltage source through a series resistance of 10 M[OMEGA], and the other electrode was connected through a shunt resistance of 11 k[OMEGA] to the ground to measure the amplitude and shape of the injected current.
Silicone rubber premixed with aluminum trihydroxide was obtained in the form of 6-mm-thick slab via compression molding. Samples of size of 5 cm X 5 cm X 6 mm were cut from a long sheet, and were exposed to gamma-ray irradiation at a dose rate of 4.5 kGy/hr in air medium. In this study, the silicone rubber specimens were exposed to gamma ray irradiation from 100 to 1,000 kGy .The hydrophobic nature of gamma irradiated specimen was characterized using static contact angle measurement using goniometer. High-speed images of water droplet movement were obtained by using a Y series camera from Integrated Design Tools Inc., USA, operated at 500 frames per second. The 0.1 N N[H.sub.4]Cl solution having the volume of 10 [micro]L is used throughout for all the measurements.
UHF Sensor Details
The broadband UHF sensor used in the present study was placed at a distance of 20 cm from the test cell to avoid any discharges from the high-voltage electrode. The output of the UHF sensor was connected to a high bandwidth digital storage oscilloscope (LeCroyWavepro 7300A, 3 GHz bandwidth, operated at 20 GSa/s) with an input impedance of 50 [OMEGA] or to the spectrum analyzer (Agilent model, 6 GHz). Judd et al. have provided the frequency response of the sensor as measured using a UHF calibration system .
Surface Charge Measurement Studies
A needle-plane electrode configuration was used for corona generation. The needle was connected to high voltage and the sample surface to which the charge has to be deposited was placed above the bottom ground electrode. The supply voltage of 8 kV DC voltage for generating corona was provided by a Trek amplifier for 20 min. After specific time intervals of charge spraying by corona discharge process, the surface charge was measured (position-2) using an electrostatic voltmeter (Trek Model 34IB). The gap between the sensor and the test specimen was maintained at 2 mm, which could cover charge measurement up to a radius of 5 mm on the surface of the specimen .
The charge (Q) on the surface of the fresh and gamma irradiated silicone rubber was calculated as
Q = V [[[epsilon].sub.o] [[epsilon].sub.r]A]/d (1)
where [[epsilon].sub.o] is the electric permittivity of vacuum, er is the relative permittivity of the medium, A is the area of cross-section of the sensor, d is the distance between sensor and the sample surface, and V is the voltage measured by electrostatic voltmeter. A minimum of three samples were used for every experimental case to ascertain the repeatability.
Thermal Analysis Procedure
Representative samples were prepared by scraping the surface of virgin and aged silicone rubber, and analyzed using thermogravimetric analysis (TGA) and Py-GC/MS. TGA of the samples was carried out in a SDT Q600 (TA Instruments) TG analyzer in [N.sub.2] ambience (100 mL [min.sup.-1]). Typically 8 mg of the samples were pyrolyzed from ambient temperature to 700[degrees]C at a heating rate of 20CC [min.sup.-1]. The data of mass loss and differential mass loss with sample temperature were collected during the experiment. [T.sub.50%] (temperature corresponding to 50% conversion) and [T.sub.max] (temperature at maximum mass loss rate) were used as markers to quantify the thermal stability of the silicone rubber samples.
Fast pyrolysis of virgin and gamma ray aged silicone rubber samples were carried out in a single shot micropyrolyzer (PY-3030S, Frontier Laboratories, Japan) coupled with a GC/MS (Shimadzu GC-2010/QP2010). The pyrolysates were separated in a UA-5 capillary column (30 m X 0.25 mm i.d. X 0.25 [micro]m, stationary phase--5% diphenylpolysiloxane). Typically, 300 [+ or -] 30 [micro]g of the sample was loaded in a stainless steel cup and dropped into the quartz tube placed inside a temperature controlled furnace. The centerline temperature of the quartz tube was 500[degrees]C, which is also the sample temperature. The typical sample heating rate was >2,000[degrees]C [s.sup.-1]. The pyrolysis vapors were immediately swept by the helium purge gas (5.5 grade) to the GC column. The flow rate of helium through the GC column was 1.59 mL [s.sup.-1] with a split ratio of 5:1. GC column oven was programmed as follows: held at 40[degrees]C for 2 min followed by a ramp at 10[degrees]C [min.sup.-1] to 300[degrees]C, and finally held at this temperature for 15 min. The products were scanned in the mass range (m/z) of 40-600 Da with an electron ionization voltage of 70 eV. Injection and interface temperatures were maintained at 300[degrees]C, while the ion source temperature was set at 250[degrees]C. The identity of the pyrolysates was confirmed by comparing the mass spectra of the peaks in the total ion chromatogram with NIST library of mass spectra of organic compounds. All the identified compounds had a high match factor >85%. Area under the peaks was calculated, and the concentrations of the pyrolysates were semi-quantitatively reported using area%.
RESULTS AND DISCUSSION
Analysis of Corona Inception Generated Due to Water Droplets
Figure 2 shows the variation in corona discharge inception voltage initiated due to 0.1 N N[H.sub.4]Cl water droplet of different conductivity on gamma irradiated specimen. Studies were carried out with number of droplets ranging from 1 to 3 placed along the axis of the electrodes. The volume of each droplet was maintained constant at 10 [micro]L. The CIV was determined based on the first UHF signal generated by discharges initiated within the electrode gap due to the applied voltage. It can be seen that CIV reduces drastically in two and three droplet configuration as compared with single droplet configuration. Also only a marginal difference is observed in the inception voltage of two and three droplet configuration. On application of voltage to the electrode gap, deformation of water droplet occurs, thereby causing high electric field intensity near the edge of water droplet facing the high-voltage electrode initiating corona discharges at much lower voltages. Li et al.  have observed that the water droplets were flat when the electrical stress was maximum and were conical when the electrical stress was minimum.
From Fig. 2, it is evident that CIV is low for 5,000 (xS/cm conductivity N[H.sub.4]Cl solution as compared with deionized water solution .For the solution having conductivity in between deionized solution and 5,000 [micro]S/cm, only the marginal change in CIV was observed. The behavior is same under all the droplet condition. The values of the CIV in Fig. 2 are subjected to deviation of [+ or -] 0.2 kV. Importantly, from Figs. 2b and 4c, corresponding to gamma irradiated silicone rubber samples, it is evident that the CIV is decreased compared to virgin specimen. This can be partly attributed to the reduction in contact angle of the silicone rubber samples with increase in gamma radiation dosage as elucidated in Table 1. This shows the reduction in hydrophobicity of the sample with increase in radiation dosage. The trends in variation of CIV with increase in conductivity of aqueous N[H.sub.4]Cl droplet on gamma irradiated specimen are similar to that of the virgin specimen.
Table 2 depicts the variation in CIV with volume of the water droplet and the aqueous N[H.sub.4]Cl droplet (of 3,000 [micro]S/cm) on gamma irradiated silicone rubber. It is observed that with an increase in volume of water droplet (up to certain volume) in the electrode gap, a marginal reduction in CIV is observed and above which an increase in tendency of CIV is observed. The characteristics are the same with water droplet and with aqueous N[H.sub.4]Cl droplet. The cause for reduction in CIV is due to reduction in the mean distance between the electrodes and the droplet edge leading to increased electric field at the triple point junction. If the volume is increased, elongation of water droplet occurs in different direction (not in the axial direction of the electrode) and thus enhancing edge area, requiring higher voltage for corona inception. In general on application of AC voltage to the electrode gap, water droplet oscillations. Nazemi and Hinrichsen  carried out an extensive study on inception of discharges from a water droplet on a hydrophobic insulating material under high electric fields, and identified four oscillation modes. In this study, the different oscillation modes are significant, especially when the number of droplets in the electrode gap is increased with both virgin and gamma irradiated specimen.
Variation in Surface Charge Characteristics in Presence of Water Droplets
Figure 3 shows the surface charge decay pattern of the gamma-irradiated silicone rubber insulating material due to corona charging of surface under positive and negative DC voltages. The surface of the insulating material was charged using a needle plane configuration with 5 mm gap under 8 kV voltage (>CIV) for both polarities with and without the water droplet. The injected charges were found to decay exponentially on removal of charges by corona injection process. It is observed that with virgin and gamma irradiated specimen, corresponding to 500 kGy dosage, the charge decay characteristics are almost similar. The charge decay characteristics of the insulating material can be written as
Q(t) = [Q.sub.0] [e.sup.-[lambda]t] (2)
where [lambda] is the decay rate and the decay time constant "t" is calculated as = 1/[lambda]. Table 1 shows the decay time constant for surface charge decay of virgin and gamma irradiated specimens. It is observed that the time constant under positive corona charging is always lesser than that corresponding to negative charging.
The magnitude of charge accumulation due to corona injection is high under negative polarity. Initially, the accumulation of charges leads to local deformation, and hence, an increase in surface area of the water droplet. On continuous injection of charges, vaporization of the water droplet is observed. This phenomenon is predominant under negative DC voltage. The injected current pulse during corona injection process in presence of water droplet causes vaporization of water droplet. The three distinct stages in the evaporation process include (a) constant contact area mode, (b) constant contact angle mode, and (c) mixed mode. These modes occur regardless of variations in both initial quantity of water in the droplets and the hydrophobic properties of the silicone rubber surface . In the process, low volume of water droplet causes increase in magnitude of voltage, which leads to corona inception. Thus negatively charged water droplet can have higher CIV.
Figure 4 shows the distribution of charge along the length of the silicone rubber insulating material. It is observed that at the zone of corona injection (2.5 cm position from either side of silicone rubber), the charge accumulation is high and a marginal reduction in magnitude of charge is observed with a single water droplet. In the presence of two droplets, the charge gets accumulated near the water droplet than at the point of charge injection. A constant charge (flat plateau) is observed in between the two droplets. With three droplets, increase in charge accumulation occurs near the droplets at the ends than at the middle droplet to which the charges were injected by corona process. In all three configurations, the charge falls off rapidly with the time as shown in Fig. 4. This characteristic is similar for the charge injected due to positive/negative corona, and with the gamma irradiated specimens. Figure 5 shows charge decay characteristics in presence of water droplet in the electrode gap. The charge is measured exactly at the center of the electrode gap. It is observed that, irrespective of number of droplet, the rate of charge decay is nearly same, irrespective of polarity of corona charging.
Leakage Current Analysis Through UHF Signal
Under AC voltages, the water droplet moves towards the high voltage electrode, and the first discharge occurs between the water droplet and high voltage electrode (Fig. 6). Nagesh et al. have clearly measured the applied voltage and the injected current and have observed that discharges occurs at the peak of the applied AC voltage .Thus, field intensification occurs at the edge of the water droplet facing ground electrode thereby causing discharges between the water droplet and the ground electrode. Lopes et al.  have clearly indicated that the electrostatic forces tend to alter the shape of the water droplets, and spread them in the direction of the electric field. In the case of two/three droplets, the elongation of water droplet occurs and when the adjacent drop gets attached, the arc gets quenched and the current flow through the droplet leads to vaporization of water droplet (Fig. 6).
Figure 7i-a and ii-a shows, respectively, the typical UHF signal generated at the time of inception of corona and at the time of intense arc discharges that occur between the electrode and the droplet. Figure 7i-b and ii-b depicts the corresponding FFT analysis of the UHF sensor output, and proves that the corona initiated discharges from liquid droplet radiate UHF signals. It is observed that during corona inception, the frequency of the signal is predominantly in the range of 0.3-1.2 GHz.
Figure 8 shows the PRPD pattern corresponding to corona activity due to the water droplet. The phase resolved partial discharge studies were carried out with the UHF signal output by connecting it to the spectrum analyzer and by operating it in zero span mode with a center frequency of 1 GHz. It is observed that corona discharge activity always occurs around the peak of the applied AC voltage. Moreover, it is observed that at the point of corona discharge inception, the magnitude of PD occurrence is higher in the positive half cycle compared to the negative half cycle. When the applied voltage is increased during the occurrence of arc in the electrode gap (with different number of droplets), the PD activity is intense, and the zone of occurrence of discharge is very different. These details are provided in Table 3
Analysis of Thermal Stability and Structural Characteristics of Virgin and Aged Samples
Figure 9 depicts the TG mass loss and differential mass loss profiles of virgin and gamma radiation aged silicone rubber samples. Virgin silicone rubber exhibits two distinct decomposition regimes, viz. 225[degrees]C-350[degrees]C and 350[degrees]C-600[degrees]C, with maximum decomposition rates observed at 323[degrees]C ([T.sub.max(1)]) and 502[degrees]C ([T.sub.max(2)]) in the respective regimes. These temperature regimes correspond to endothermic depolymerization of silicone polymer to form oligomeric fragments. Importantly, with increasing dosage of gamma radiation, a significant shift in the mass loss profiles to lower temperatures is observed.
As evidenced in Table 4, [T.sub.max(1)], [T.sub.max(2)], and [T.sub.50%] decrease with increase in aging, which signifies the destabilization of silicone rubber with ageing. The overall mass of residue left at the end of 700[degrees]C was 44-45 wt% for all samples, which can be attributed to the filler present in the commercial silicone rubber sample. Interestingly, the shoulder observed in the differential mass loss profiles of virgin and 50 kGy samples at around 550[degrees]C (Fig. 9b) disappears when the radiation dosage is increased beyond 300 kGy. This shows that significant surface structural modifications occur during gamma radiation ageing, which tends to accelerate the high temperature thermal decomposition of silicone rubber by decreasing the activation energy of the process.
To understand the structural changes caused by gamma radiation ageing, Py-GC/MS analysis of the virgin and aged samples was carried out. Figure 10 depicts the GC/MS total ion chromatogram of the pyrolysates from virgin, 300 kGy aged, and 1200 kGy aged samples. Table 5 shows the various organics grouped according to major functional groups. Cyclic siloxane oligomers constitute a major fraction of the pyrolysate. For the virgin polymer, cyclic siloxane trimer was the major compound followed by tetramer, pentamer, hexamer, and heptamer. It is well reported in the literature that pyrolysis of poly(dimethyl siloxane) (PDMS) produces cyclic timer as the major product via intramolecular backbiting reactions of the chain segments [18-22]. The degradation is proposed to occur via a stepwise mechanism involving cyclization from the chain end. The observed relative composition of the cyclic siloxane oligomers for the virgin sample is in line with Grassie and Macfarlane . The total ion chromatogram is also in agreement with Tsuge et al. . From the literature, the effect of gamma irradiation on the pyrolysis product distribution from silicone polymers is not yet known. From Table 5, it is evident that the relative yield of cyclic siloxanes decreases with gamma radiation dosage. More importantly, the yield of cyclic trimer decreases and the formation of bigger cyclic molecules are observed. For the sample aged at 1,200 kGy, cyclic siloxanes with 9, 10, 11, and 13 repeat units are observed. The formation of cyclic siloxanes is attributed to the Si--O bond scission in a folded chain cyclic conformation involving the overlap of empty Si d-orbitals with orbitais of oxygen and carbon atoms .
The formation of cyclic siloxanes with more than 3 or 4 repeat units after gamma radiation can be due to the enhanced chain flexibility, which results in a lower ring strain energy for 1-X-cyclization reactions, where X [greater than or equal to] 5. Similar observation was also made when linear dimethyl silicone liquid was degraded in presence of catalysts like cerium ions . Besides cyclic siloxanes, the formation of other functional groups is also observed with gamma radiation aged samples. These include Sicontaining organics, linear and cyclic hydrocarbons (C8-C30) long-chain alcohols (C6-C16), benzene derivatives, and other oxygenated organics. As it is evident from Fig. 10, C[O.sub.2] was also produced. The formation of C[O.sub.2] and other linear Sicontaining organics can be attributed to the elimination of crosslinker residues .
It is important to note that during gamma irradiation, both scission and crosslinking of PDMS can occur . Hill et al.  showed that gamma radiolysis of PDMS under vacuum condition produced more crosslinking compared to that in presence of air at ambient conditions. As silicone rubber samples in this study were exposed to gamma irradiation under ambient air environment, crosslinking of silicone rubber is not expected. This is validated by the unchanging yield of solid residue from TG mass loss profile at the final temperature of 700[degrees]C (Fig. 9a). During gamma irradiation period, the diffusion of oxygen into the martix of silicone rubber controls the rate of aging , The presence of oxygen can lead to many reactions involving the formation of carbonyl, hydroxyl, and hydroperoxy groups in the polymer via free-radical mechanism .
The oxygenated organics in Table 5 is primarily composed of functional groups like carboxylic acids, esters, and carbonyl compounds with and without Si incorporation. The formation of long chain alcohols also stands as an evidence for the role played by gamma radiation and oxygen in disrupting the PDMS matrix and cleaving the crosslinks that are already present in the commercial rubber sample. Based on the thermal analyses, it can be ascertained that gamma radiation dosage causes significant surface rupture leading to (a) increased flexibility of siloxane backbone, (b) scission of the crosslinking bonds, and (c) formation of oxygen functionality involving polar groups such as hydroxyl, carbonyl and carboxylic acids. These surface changes can be expected to play a vital role in charge trapping during corona charging process. It was earlier observed in the section "Variation in Surface Charge Characteristics in Presence of Water Droplets" that the charge decay characteristics of virgin and 500 kGy dosed samples were similar with similar decay constants. This is in line with the higher yield of oxygenated organics, Si-based scission fragments, and higher order cyclic siloxane oligomers with m [greater than or equal to] 3, when the gamma radiation dosage is >500 kGy (e.g., 700 and 1200 kGy) (Table 5). This also justifies the low value of CIV for gamma radiation dosed samples compared to virgin sample (Fig. 2).
Corona inception voltage (CIV) due to water droplet on gamma irradiated silicone rubber reduces with increase in conductivity and number of droplets in the electrode gap along the axis of the electrode. The volume of droplet (with water and aqueous N[H.sub.4]Cl) has high impact on CIV. Increase in dosage of gamma irradiation reduces the CIV of water droplet. The UHF signal generated during corona inception due to water droplet has frequency bandwidth in the range of 0.3-1.2 GHz. The corona discharge occurs in the phase range of 30[degrees]-130[degrees] in the positive half cycle and about 210[degrees]-270[degrees] in the negative half cycle. Surface charge accumulation studies have shown a characteristic change with gamma irradiated specimen. In the presence of water droplet, the charge decay occurs rapidly with a single droplet and with three droplets. Py-GC/MS study reveals that higher cyclic siloxane oligomers, long-chain alcohols, hydrocarbons, aromatics, and oxygenated organics are also formed from gamma radiation dosed samples, which indicates that the siloxane backbone is rendered more flexible via cyclization, chain scission, and reactions of diffused oxygen into the matrix of siloxane.
One of the authors (RS) wishes to thank Dr M.D. Judd for his kind support to carry out this study. He also wishes to thank Department of Science and Technology, New Delhi, for sponsoring the project on understanding electrical properties of irradiated insulating materials. The authors also thank IGCAR, Kalpakkam, for irradiating the specimens.
[1.] L. Bokobza, J. Appl. Polym. Sci., 93, 2095 (2004).
[2.] J. Brandrup, E. H. Immergut, and E. A. Grulke, Eds., Polymer Handbook, 4th ed., 583 (1999).
[3.] K. Cao, Y. Ao, J. Chen, J. Peng, W. Huang, J. Li, and M. Zhai, J. Appl. Polym. Sci., 134, 454404 (2017).
[4.] I. Stevenson, L. David, C. Gauthier, L. Arambourg, J. Davenas, and G. Vigier, Polymer, 42, 9287 (2001).
[5.] S.G. Burnay, Nucl. Instrum. Methods Phys. Res. B, 185, 4 (2001).
[6.] H. Janssen, and U. Stietzel, 10th ISH, 149 (1997).
[7.] M. Kyoto, Y. Chigusa, M. Ohe, H. Go, M. Watanabe, T. Matsubara, T. Yamamoto, and S. Okamoto, J. Lightwave Teehnol., 10, 289 (1992).
[8.] G. Nagesh, and R. Sarathi, J. Phys. D Appl. Phys., 41, 155407 (2008).
[9.] M.H. Nazemi, and V. Hinrichsen, IEEE Intern. Conf. on Solid Dielectrics, 194 (2013).
[10.] R. Sarathi, V.S. Harsha, H. Griffiths, and A. Haddad, IEEE Trans. Dielectr. Electr. Insul., 21, 918 (2014).
[11.] K.A. Murray, J.E. Kennedy, B. McEvoy, O. Vrain, D. Ryan, R. Cowman, and C.L. Higginbotham, J. Mech. Behav. Biomed. Mater., 17, 252 (2013).
[12.] IEC publication, 60112, Recommended method for determining the comparative tracking index of solid insulating material under moist condition, 1972, 2nd edition.
[13.] M.D. Judd, and O. Farish, IEEE Trans. Instrum. Meas., 47, 875 (1998).
[14.] R. Sarathi, I. Merin Sheema, and J. Rajan, IEEE Trans. Dielectr. Electr. Insul., 21, 674 (2014).
[15.] Q. Li, R. Shuttleworth, I. Dupere, G. Zhang, S.M. Rowland, and R.S. Morris, IEEE Intern. Symp. on Electr. Insul., Puerto Rico, 2012.
[16.] J.-H. Kim, S.I. Ahn, J.H. Kim, and W.-C. Zin, Langmuir, 23, 6163 (2007).
[17.] U.S. Lopes, S.H. layaram, and E.A. Cherney, IEEE Trans. Dielectr. Electr. Insul., 8, 262 (2001).
[18.] N. Grassie, and I.G. Macfarlane, Eur. Polym. J., 14, 875 (1978).
[19.] G. Camino, S.M. Lomakin, and M. Lazzari, Polymer, 42, 2395 (2001).
[20.] G. Camino, S.M. Lomakin, and M. Lageard, Polymer, 43, 2011 (2002).
[21.] S. Tsuge, H. Ohrani, and C. Watanabe, Pyrolysis-GC/MS Handbook of Synthetic Polymers, Elsevier, Oxford, UK (2011).
[22.] J.P. Lewicki and R.S. Maxwell, Advances in gas chromatography, InTech, https://doi.org/10.5772/57509 (accessed 2014)
[23.] S. Yasufuku, IEEE Trans. Electr. Insul., EI-17, 338 (1982).
[24.] DJ.T. Hill, C.M.L. Preston, D.J. Salisbury, and A.K. Whittaker, Radiat. Phys. Chem., 62, 11 (2001).
[25.] J. Pfaendtner, and L.J. Broadbelt, Ind. Eng. Chem. Res., 47, 2886 (2008).
Palash Mishra, (1) Ribhu Gautam, (2) Ravikrishnan Vinu, (2) Ramanujam Sarathi (1)
(1) Department of Electrical Engineering, Indian Institute of Technology Madras, Chennai 600 036, India
(2) Department of Chemical Engineering, Indian Institute of Technology Madras, Chennai 600 036, India
Correspondence to: R. Sarathi; e-mail: email@example.com
Published online in Wiley Online Library (wileyonlinelibrary.com).
Caption: FIG. 1. Experimental setup. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 2. Variation in CIV for 1 droplet, 2 droplets, and 3 droplets configuration with different conductivity N[H.sub.4]Cl solution for (a) virgin sample, (b) 100 kGy, and (c) 1000 kGy. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 3. Surface charge decay pattern of gamma irradiated silicone rubber sample under positive DC and negative DC corona charging. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 4. Variation in surface charge distribution along the surface of silicone rubber insulating material with time under (a) positive DC and (b) negative DC corona charging for (i) 1 droplet, (ii) 2 droplets, and (iii) 3 droplets configuration. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 5. Surface charge decay characteristics of virgin silicone rubber insulating material under (i) 1 droplet, (ii) 2 droplets, and (iii) 3 droplets configuration. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 6. Water droplet movement on application of AC voltage with water droplet on silicone rubber insulating material with (a) 1 droplet, (b) 2 droplets, and (c) 3 droplets configuration. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 7. Typical (i) UHF signal in time domain and (ii) its corresponding FFT: (a) at point of corona inception and (b) during arcing. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 8. Phase resolved partial discharge pattern obtained due to discharges initiated due to water droplet on silicone rubber insulating material at the time of (a) corona inception and (b) arcing for (i) 1 droplet (ii) 2 droplets, and (iii) 3 droplets configurations. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 9. Thermogravimetric (a) mass loss and (b) differential mass loss profiles of the virgin and gamma ray aged silicone rubber samples. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 10. GC/MS total ion chromatogram of the pyrolysates obtained from pyrolysis of virgin silicone rubber and aged samples at 300 and 1,200 kGy gamma irradiation. The structure of the salient oligomers is also shown. [Color figure can be viewed at wileyonlinelibrary.com]
TABLE 1. Contact angle and the decay time constant for surface charge decay of virgin and gamma irradiated samples. Decay time constant ([tau]) (s) Sample Contact angle + DC -DC ([degrees]) Virgin 112.5 122.99 169.91 500 kGy 98.2 119.84 141.52 700 kGy 96.1 70.12 94.22 1,000 kGy 90 50.56 52.54 TABLE 2. Variation in corona Inception voltage due to water droplet and the aqueous N[H.sub.4]Cl droplet (of 3,000 ([micro]S/cm) of different volume on gamma irradiated silicone rubber. Corona inception voltage (kV) Volume Virgin 100 kGy 500 kGy 1,000 kGy ([micro]L) A B A B A B A B 20 3.08 3.08 2.95 2.94 2.34 2.33 2.12 2.09 40 3.03 2.99 2.90 2.85 2.19 2.15 2.00 2.01 60 2.97 2.81 2.71 2.69 2.08 2.08 2.01 2.00 80 3.04 2.99 2.86 2.80 2.27 2.18 2.09 2.07 100 3.09 3.08 2.97 2.94 2.44 2.30 2.25 2.19 TABLE 3. Number of peaks for PRPD at the time of corona inception and arcing and the zone of occurrence of discharge for 1 droplet, 2 droplets, and 3 droplets configuration. Number of peaks Configuration Inception Arcing Zone of discharge ([degrees]) 1 droplet 3 80 43.63-136.37 2 droplets 4 94 32.21-147.79 3 droplets 5 129 35.87-162.31 TABLE 4. Variation of [T.sub.50%], [T.sub.max(1)], and [T.sub.max(2)] for virgin and gamma ray aged silicone rubber samples. Sample name/ [T.sub.max(1)] [T.sub.max(2)] [T.sub.50%] gamma irradiation ([degrees]C) ([degrees]C) ([degrees]C) Virgin 323 502 484 50 kGy 317 500 477 300 kGy 308 494 466 700 kGy 300 482 448 1,200 kGy 305 482 449 TABLE 5. Composition of the pyrolysates (in GC/MS peak area%) obtained from pyrolysis of virgin and gamma ray aged silicone rubber samples at 500[degrees]C. Compounds/category Virgin 50 kGy 300 kGy Total cyclic siloxaneoligomers 96.36 81.96 81.32 Hexamethylcyclotrisiloxane 76.35 65.98 60.92 (n = 3, m = 1) Octamethylcyclotetrasiloxane 10.04 9.54 9.66 (n = 4, m = 2) Decamethylcyclopentasiloxane 3.05 3.40 3.85 (n = 5, m = 3) Dodecamethylcyclohexasiloxane 2.60 1.92 3.70 (n = 6, m = 4) Tetradecamethylcycloheptasiloxane 2.68 1.12 1.97 (n = 7, m = 5) Hexadecamethylcyclooctasiloxane 1.64 1.22 (n = 8, m = 6) Octadecamethylcyclonanasiloxane (n = 9, m = 7) Eicosamethylcyclodecasiloxane (n = 10, m = 8) Docosamethylcycloundecasiloxane (n = 11, m = 9) Hexacosamethylcyclotridecasiloxane (n = 13, m = 11) Other Si-based organic compounds 0 0.12 1.72 Linear chain hydrocarbons Alkane 0 4.70 1.55 Alkene 0 3.22 2.45 Alkyne 0 3.58 1.66 Benzene derivatives 0 0.10 1.20 Alcohols 0 0.80 1.48 Other oxygenated organics 0 0.97 1.92 Other unidentified compounds 3.64 4.55 6.70 Compounds/category 700 kGy 1,200 kGy Total cyclic siloxaneoligomers 80.55 73.80 Hexamethylcyclotrisiloxane 64.73 42.95 (n = 3, m = 1) Octamethylcyclotetrasiloxane 0 11.80 (n = 4, m = 2) Decamethylcyclopentasiloxane 0 2.95 (n = 5, m = 3) Dodecamethylcyclohexasiloxane 12.77 4.56 (n = 6, m = 4) Tetradecamethylcycloheptasiloxane 3.05 3.20 (n = 7, m = 5) Hexadecamethylcyclooctasiloxane (n = 8, m = 6) Octadecamethylcyclonanasiloxane 2.73 (n = 9, m = 7) Eicosamethylcyclodecasiloxane 1.07 (n = 10, m = 8) Docosamethylcycloundecasiloxane 2.07 (n = 11, m = 9) Hexacosamethylcyclotridecasiloxane 2.47 (n = 13, m = 11) Other Si-based organic compounds 3.38 7.76 Linear chain hydrocarbons Alkane 0.86 0.46 Alkene 1.52 4.68 Alkyne 0 0 Benzene derivatives 0.45 0.20 Alcohols 0.92 2.30 Other oxygenated organics 3.70 4.80 Other unidentified compounds 8.62 6.00 The basic structure for the cyclic siloxane oligomers is shown in Fig. 10. The general formula is given as [[(C[H.sub.3]).sub.2]--Si--O--].sub.n]. The value of n decides the number of repeat units, while the meaning of m is shown in Fig. 10.
|Printer friendly Cite/link Email Feedback|
|Author:||Mishra, Palash; Gautam, Ribhu; Vinu, Ravikrishnan; Sarathi, Ramanujam|
|Publication:||Polymer Engineering and Science|
|Date:||Jan 1, 2019|
|Previous Article:||[Poly(hexylacrylate).sub.Core]-[Poly(ethyleneglycol methacrylate).sub.Shell] Nanogels as Fillers for Poly(2-hydroxyethyl methacrylate) Nanocomposite...|
|Next Article:||Synthesis and Characterization of the Novel Nylon 12 6 Based on 1,12-Diaminododecane.|