Radiation processing of polymer insulators: a method for improving their properties and performance.
Rod-shaped polymer insulators are increasingly used in power engineering, in place of traditional high-voltage porcelain and glass insulators, for electric apparatus and overhead power transmission lines (1-3) see Fig. 1. Their advantages are mainly because of polymer properties. They can be a dielectric, exhibit a low density ([approximately]2 g/[cm.sup.3]), an absence of brittleness, and a higher (1.5 to 2.0 times) flashover characteristics in wet and soiled condition. Moreover, their mass can be smaller (by [approximately]10 times) because metal details (caps and pins) are absent and plastic fiber glass rods with a high tensile strength can be used. The better impact characteristics, typical of polymers, are particularly important in the manufacture of seismically stable bus support insulators or contact circuit insulators for railways. High flashover characteristics of polymer insulators in the wet and soiled condition are because of their smaller equivalent diameter and optimized dimensions of shed and intershed spaces. Hence, it is possible to design high-voltage constructions of minimum height (vertical dimension determined by the electrical strength of the air gap) with optimum screening, e.g. for bus support insulators or interphase spaces of contact railways lines.
Moreover, the use of polymer insulators simplifies transport, mounting, and other operations during the construction and repair of high-voltage overhead lines and electric apparatus and with reduced cost.
Polymer insulators are used at substations and for overhead power transmission lines. They are applied in tensioning and in string suspensions of conductors and interphase spacers in compact lines, and also in contact circuits of electric railways: This makes it possible to increase train speed. It is particularly advisable to use these insulators in electrical engineering for constructions in areas of difficult access.
Polymer insulator coatings are made from polymer compositions based on silicone rubber, polytetrafluoroethylene, fluorine-containing copolymers, and other thermoplastics (1-4). However, all these polymers are relatively expensive.
We used an ethylene and vinyl acetate copolymer (EVA) as the base polymer composition for high-voltage insulators. It is relatively cheap and can be processed by molding. However, products made from these polymer compositions exhibit low shape stability at high temperatures, because their Vicat thermal stability is [approximately]85 [degrees] C (5). Moreover, these compositions are insufficiently stable to track and erode when insulators operate under the conditions of wet and dirty surfaces.
To eliminate these defects, radiation processing was carried out, which induces formation of a three-dimensional network (6). Radiation processing is preferable to thermochemical crosslinking. This is because the undesirable process of premature crosslinking, upon heating, during molding is eliminated, and the polymer composition does not creep. Moreover, initiator residues can decrease tracking, and surface erosion shortens their service lifetime.
Radiation processing increases the shape and heat stability of insulators. To optimize radiation crosslinking of EVA functional additives are proposed by us. Stability against tracking and erosion was also ensured by introducing aluminium hydroxide as a filler.
The composition contained EVA with a vinyl acetate unit content of 12 mole % plus a filler, aluminium hydroxide, in approximately equal molar ratios. This composition is known in the CIS countries(*) under the trademark SEVA-113-12.
The SEVA-113-12 composition also contains a light stabilizer: 4-alkoxy-2-hydroxy-benzophenone; alkyl, [TABULAR DATA FOR TABLE 1 OMITTED] [C.sub.7][H.sub.15]-[C.sub.9][H.sub.19] (Benzone OA), in an amount of 0.25 to 0.35 wt% and a thermostabilizer: 2,2[prime]-methylene bis (6-tert butyl-4-methylphenol) (Agidole-2) at 0.15 to 0.25 wt%. Benzone OA and Agidole-2 are trademarks in CIS countries.
The radiation processing of the EVA composition was carried out on K-300,000 radioisotope unit (7).
The gel fraction was determined by extracting the polymer portion soluble in xylene, from the irradiated samples, at 130 [degrees] C for 8 h. The data on the gel fraction were used to calculate the gelation dose ([D.sub.g]) by the Dole method (8). The [G.sub.s]/[G.sub.c](**) ratio was found from the Charlesby-Pinner equation (9).
Properties of Radiation-Processed Polymer Compositions and Insulators
DSC thermograms were recorded in the temperature range from 20 to 360 [degrees] C in air at a heating rate (V) of 8 [degrees] C/min with a DSC-D instrument. DSC curves showed the melting temperatures ([T.sub.m]) of the crystalline phase EVA, the beginning ([T.sub.b]), the maximum ([T.sub.max]), and the end ([T.sub.e]) of the first stage of the oxidative degradation, which probably occurs without mass loss. DSC thermograms were used to calculate the total degradation heat evolution ([Delta]H) in J/g.
Thermomechanical curves (TMC) were recorded with a Heppler consistometer by the method of stress relaxation at compression loads from 0.026 to 6.4 MPa and a heating rate of 1.5 to 2 [degrees] C/min.
Stability to Tacking and Erosion
This test was carried out by immersing the insulators with a length of the insulated part of 1 m into an aqueous solution of sodium chloride (10 wt%) for 35 s. Subsequently they were removed from the solution after 15 s, then maintained at a voltage 73 kV for 35 s, and then immersed into the solution again. The tests were performed in cycles consisting of 17 h tests in solution at 1500 Ohm cm, 9 h at 750 Ohm cm, and 3 h at 300 Ohm cm. The cycles were preceded by an initial stage, which consisted of a 20 h test in a 10% solution of NaCl at 1500 Ohm cm (10).
RESULTS AND DISCUSSION
Radiation Crosslinking of EVA Composition
EVA was used with a vinyl acetate unit content of 12 mole %. The introduction of vinyl acetate units into the polyethylene chain facilitates crosslinking by decreasing the crystallinity and thus increasing the macrochain mobility (11). Moreover, the presence of acetate groups in the macromolecule imparts adhesion properties to the polymer. This is important in the manufacture of insulators with a glass-reinforced plastic rod because the product acquires the properties of a whole. Further radiation processing of insulators increases these properties because of a "memory" effect exhibited by polyolefins (12).
Among the tested additives, sensitizers of radiation-induced crosslinking, N,N[prime]-m-phenylene dimaleimide (PDMI) exhibited the highest effectiveness; it was also effective with other compounds undergoing radiation processing: e.g. polyethylene, polyisoprene, poly(vinylidene fluoride) (12-16).
Gamma radiation not only ensures the uniform formation of a three-dimensional network throughout the polymer bulk, but also makes it possible to control easily the network density by varying the absorbed dose. To determine the effect of filler and sensitizer on the rate of radiation-induced EVA crosslinking, pure EVA, not containing these components, was used in control experiments (Table 1) (17, 18).
These data indicate that in the investigated range of absorbed doses, EVA and its formulations are highly crosslinked and the values of [G.sub.s]/[G.sub.c] are in agreement with the published data (19). Aluminium hydroxide and PDMI decrease the yield of the gel fraction and the effect of degradation. In all cases oxygen in air leads to greater degradation.
The results of the gel-sol analysis were also used to calculate the molecular weight between the network junction ([M.sub.c])(***) and the number of network chains ([v.sub.c]). For this purpose, the equilibrium swelling method was applied, with the calculations carried out according to Flory-Rehner's equation (12, 20). To account for the effect of the interaction between the polymer and solvent, the swelling coefficient of the crosslinked polymer was determined in two solvents: xylene and chloroform. Hence, it was possible to estimate the Huggins constant (17). The parameter [M.sub.c] was calculated on the assumption that the network is formed by tetrafunctional junction. The data obtained are in Table 2. Since the parameters [M.sub.c] and [v.sub.c] were determined for samples processed with absorbed doses of 100 to [TABULAR DATA FOR TABLE 2 OMITTED] 400 kGy, the medium has no molecular network density because of relatively high gelation (Table 1). Hence, the values of [M.sub.c] and [v.sub.c] for samples irradiated in argon and in air virtually do not differ.
The analysis of data in Table 2 shows that the presence of PDMI profoundly affects the network density: [M.sub.c] is decreased [approximately]2.0 to 2.5 times. Moreover, the presence of PDMI in the composition makes it possible to attain the same value of [v.sub.c] at an absorbed dose four times smaller than that for SEVA-113-12, which does [TABULAR DATA FOR TABLE 3 OMITTED] not contain this additive. Consequently, these experiments have shown that by increasing the network density PDMI plays the role of the sensitizer for the radiation-induced cross-linking of the EVA-based polymer.
Properties of Radiation-Processed Materials and Insulators Based on EVA
The effect of radiation processing of EVA, and compositions based on it, was followed by differential scanning calorimetry (DSC). Beginning from 190 [degrees] C, DSC thermograms [ILLUSTRATION FOR FIGURE 2 OMITTED] show the development of oxidation in a non-irradiated EVA. The introduction of aluminium hydroxide greatly increases the polymer resistance to oxidation, as [T.sub.b] rises to 263 [degrees] C. The addition of PDMI to the non-irradiated composition increases [Delta]H from 10 to 85 J/g, compared to EVA, which results from thermal crosslinking. A similar effect of this additive has been observed in the investigation of thermal crosslinking of polyethylene (21). This process occurs simultaneously with oxidation, and [Delta]H characterizes the overall exotherm. Radiation processing decreases the resistance of the composition to oxidation, and [T.sub.b] is displaced to lower temperatures (Table 3). Moreover, a distinct negative effect of low dose rates is observed. Thus, for SEVA-113-12 irradiated at a dose rate of 0.4 kGy/h, the [T.sub.b] of oxidation is 190 [degrees] C, whereas for the non-irradiated sample it is 263 [degrees] C. Irradiated samples of SEVA-113-12 both with the addition of PDMI, or without it, are characterized by a decrease in [T.sub.b], particularly at a dose rate of 0.4 kGy/h. However, in the presence of PDMI, [Delta]H is lower by almost one order of magnitude than that of the radiation-processed composition without the additive (Table 3). Hence, PDMI prevents the occurrence of thermo-oxidative degradation. This agrees with the data obtained for EVA without filler, but containing 2 to 5 wt% PDMI.
Consequently, this investigation established the following optimum conditions for radiation processing of insulators based on EVA in the presence of PDMI for the formation of a three-dimensional network: that is, an absorbed dose up to 200 kGy at a dose rate of 10 kGy/h in air. The variation in the dose rate showed that optimum gelation occurs at 2 kGy/h, and the required absorbed dose decreases to 100 to 150 kGy. At a dose rate [less than]2 kGy/h the exposure time is long and an inert gas medium should be used to suppress oxidative degradation, with radiation processing not being efficient.
The Analysis of TMC.
The main aim of radiation-processing of compositions based on EVA was to increase shape stability at high ambient temperatures. Therefore, the investigation of the corresponding properties was our primary focus. The analysis of TMC of filled SEVA-113-12 without a stabilizer showed that the dose rate and the absorbed dose do not essentially change the softening temperature ([T.sub.s]) as compared to that of non-irradiated samples ([T.sub.s] = 90 to 100 [degrees] C). However, a sample that absorbed a dose of 200 kGy at 0.4 kGy/h is more readily deformed ([T.sub.s] = 88 [degrees] C) than those irradiated with the same dose but at 2 and 10 kGy/h ([T.sub.s] = 96 [degrees] C). This is probably because of partial polymer degradation caused by ozonized air that accumulates in the irradiation chamber with prolonged exposure.
A filled EVA composition with 2 wt% of PDMI irradiated with different doses at different dose rates, TMC [ILLUSTRATION FOR FIGURE 3 OMITTED] (22) shows a region of equilibrium deformation ([Epsilon] = 3.5 and 4 mm; [T.sub.S] = 74 and 92 [degrees] C for dose rates 2 and 0.4 kGy/h, respectively, the absorbed dose was 200 kGy). Thermochemical measurements were used to calculate the elastic modulus and [M.sub.c] (23). The results agree with those determined by the equilibrium swelling method (Table 2).
The parameter Me depending on the dose rate at an absorbed dose of 200 kGy is as follows:
dose rate, kGy/h 0.4 2.0 [M.sub.c] 8670 6960
This parameter, just as the high elastic modulus, indicates the advantage of radiation processing at a dose rate of 2 kGy/h, because the sample is characterized by the highest network density, and degradation is less pronounced.
Evaluation of the yield stress, at which the sample begins to flow, i.e. to lose its primary shape at 100 [degrees] C, showed that SEVA-113-12 and its composition with PDMI flows at low loads before radiation processing [ILLUSTRATION FOR FIGURE 4 OMITTED]. Radiation processing increases the flow limit to 0.9 to 1.2 MPa depending on the dose rate. The addition of 2% PDMI increases it to [similar to]5 MPa [ILLUSTRATION FOR FIGURE 4 OMITTED]. This performance can guarantee high shape stability of products in service.
The parameters of shape stability and resistance to tracking erosion were also studied for finished insulators. For this purpose a SEVA-113-12 coating consisting of several sheds was applied to a fiber glass plastic rod by injection molding (pressure [similar to]150 MPa, temperature [similar to]140 [degrees] C, time [similar to]1 min). The insulator obtained [TABULAR DATA FOR TABLE 4 OMITTED] was subjected to radiation-processing at an optimum dose rate and absorbed dose.
The evaluation of shape stability consisted of maintaining polymer insulators in boiling water for 4 h. The change in the shed shape, which was initially conical, was followed visually. All irradiated insulators (100 to 200 kGy) passed these tests successfully, whereas non-irradiated insulators lost their shape after the first hour in boiling water and subsequently the sheds melted [ILLUSTRATION FOR FIGURE 5 OMITTED].
The tests for shape stability at [greater than] 100 [degrees] C in a hot air chamber showed that irradiated insulators retain their shape up to 200 [degrees] C, but with the addition of PDMI to [greater than] 200 [degrees] C (Table 4). This is probably because of the inclusion into the polymer network of rigid-chain maleimide fragments exhibiting high heat stability (24, 25).
Tests for Stability to Tracking and Erosion
Table 4 shows that radiation processing does not adversely affect the main insulator property: its stability to tracking and erosion, which is because of the presence of the filler, aluminium hydroxide. For the composition containing PDMI this parameter is even slightly higher at an absorbed dose of 100 to 200 kGy, which is probably caused by greater network density and the imide fragments present.
Types of Polymer Insulators Subjected to Radiation Processing
At present, in St. Petersburg Technical University, several types of high-voltage insulators based on EVA are manufactured with the application of radiation-chemical technology, and they are being successfully used in the field (22). Table 5 gives the main characteristics of these insulators. The electric characteristics of all insulators listed in Table 5 meet the requirements of the standard (State Standard 1516.1-76) with respect to the path length of leakage of external insulation (State Standard 9920-89) and specific requirements to long rod polymeric suspension insulators (State Standard 28856-90).
The trademarks of insulators listed in Table 5 contain the abbreviation "UKhL," which means that they are intended for use in "moderate and cold climate." For more than a decade several hundred insulators for 110 kV overhead lines have been in operation, mainly in the central region of Russia. Several hundred of [TABULAR DATA FOR TABLE 5 OMITTED] them were also used as parts of disconnectors for 10 and 35 kV. The experience with their operation is generally positive.
In our opinion, the characteristics of the shape-(heat) stability of radiation-processed polymer insulators make it possible to use them in regions of hot climate.
The radiation processing of polymer insulators based on the compositions of an ethylene-vinyl acetate copolymer proved to be highly suitable for the manufacture of insulators with high shape stability for high-voltage overhead power transmission lines (35 to 110 kV) and electric apparatus. The use of PDMI as a sensitizer of radiation-induced crosslinking decreases the required absorbed dose and increases the product shape stability as well as its resistance to oxidation.
The authors express their warm gratitude to Dr V.A. Goldin for the radiation processing of polymer samples and insulators and to A.V. Kuznetsov for the tests of polymer insulators.
* CIS countries = countries of Commonwealth of Independent States (former Soviet Union).
** [G.sub.s] = radiation-chemical yield of scission.
[G.sub.c] = radiation-chemical yield of crosslinking.
*** Relative values.
1. G. N. Aleksandrov and V. L. Ivanov, Insulation of Electric High-Voltage Apparatus, Energoatomizdat, Leningrad (1984).
2. G. N. Aleksandrov, Extra-High-Voltage Installations and Environment Protection, Energoatomizdat, Leningrad (1989).
3. H. Klockhaus and G. Wanser, Abschluss und Verbindungstechnik bei Starkstromkabeln, VWEW, Frankfurt/Main (1979).
4. Rheinisch-Westfaelische Isolatoren Werke GmbH, BE Patent 853,994 (1977).
5. J. Kosaka, Nippon Gomu Kyokaishi, 51, 702 (1978).
6. A. Charlesby, Atomic Radiation and Polymers, Pergamon Press, Oxford England (1960).
7. V. A. Goldin and E. D. Chistov, Installations and Instruments for Radiation Technology, Energoatomizdat, Moscow (1985).
8. M. Dole and V. M. Patel, Radiat. Phys. Chem., 9, 433 (1977).
9. A. Charlesby and S. H. Pinner, Proc. Roy. Soc., A249, 367 (1959).
10. Polymeric Long-rod Insulators. Specifications. Test Methods (OST 34-27-688-84). Center of Scientific and Technical Information on Power Engineering and Electrofication, Ministry of Power Engineering of the USSR, Informenergo, Moscow (1985).
11. N.M. Burns, Radiat. Phys. Chem., 14, 797 (1979).
12. V.S. Ivanov, Radiation Chemistry of Polymers, VSP, Utrecht, The Netherlands (1992).
13. S. M. Miller, R. Roberts, and R. L. Vale, J. Polym. Sci., 58, 737 (1962).
14. S. M. Miller, M. W. Spindler, and R. L. Vale, J. Polym. Sci., A1, 2537 (1963).
15. S. M. Miller, M. W. Spindler, and R. L. Vale, in Industrial Uses of Large Radiation Sources, Vol. 1, p. 329, IAEA, Vienna (1963).
16. V. S. Ivanov, I. I, Migunova, and A. I. Mikhailov, Radiat. Phys. Chem., 37, 119 (1991).
17. I. I. Migunova, T. E. Panina, A. V. Kuznetsov, G. N. Aleksandrov, and V.S. Ivanov, Vestnik Leningradskogo Univ., Ser. Fiz.-Khim., No. 4 (25), 70 (1991).
18. A.V. Kuznetsov, G. N. Aleksandrov, V. S. Ivanov, I, I. Migunova, V. A. Goldin, and N. P. Syrkus, in Proc. Third Annual Scientific Conference Nuclear Society International. Moscow. Nuclear Technology Tomorrow, S.V. Kryukov, ed., Transact. Amer. Nucl. Soc., 67 (Suppl. 1), 423 (1993).
19. A. G. Sirota, Modification of Structure and Properties of Polyolefins, Khimiya, Leningrad (1984).
20. R. A. Hayes, Rubber Chem. Technol., 59, 138 (1986).
21. M. D. Pukshansky, I. I. Migunova, T. N. Kondakova, and V. S. Ivanov, Vestnik Leningradskogo Univ., Ser. Fiz.-Khim., No. 3 (18), 121 (1987).
22. V. S. Ivanov, N. A. Kalinina, I. I. Migunova, G. N. Aleksandrov, V. I. Irkhin, A. V. Kuznetsov, A. I. Shchelokov, V. A. Goldin, F. Z. Raychuk, and N. P. Syrkus, Vestnik Radtech "Euroasia," No. 2 (4), 32 (1992).
23. L. R. G. Treloar, The Physics of Rubber Elasticity, 3d ed., Clarendon Press Oxford, England (1975).
24. V. S. Ivanov, I. P. Bezhan, and L. K. Levando, Vestnik Leningradskogo Univ., Ser. Fiz.-Khim., No. 10, 157 (1965).
25. V. S. Ivanov, Radiation-Induced Polymerization, Khimiya, Leningrad (1967).