Anti-Tracking High Voltage Insulating Materials

This invention relates to high-voltage insulating materials.

While polymeric materials are widely used for insulating a wide variety of electrical apparatus, they are not suitable for high voltage applications in contaminated atmospheres where moisture or fog, together with salts, dust particles and ionic pollution causes leakage currents to flow across the surface of the insulation. This current causes a rise in temperature with consequent moisture evaporation and ultimately dry hand formation. The electrical stress across these dry bands often exceeds the breakdown stress of the air-insulation interface, so that discharge or spark scintillation takes place. The spark temperature is extremely high, often 2000.degree. C. or higher, and the heat produced may be sufficient to cause degradation of the insulation surface with the ultimate formation of carbonaceous spots. These carbonaceous spots usually link up in dendritic fashion and the organic insulation fails by progressive creepage tracking.

Over the years many solutions to these problems have been proposed of which perhaps the most effective has been the incorporation of hydrated alumina, preferably the trihydrate, in fairly substantial quantities to for example butyl rubber, epoxy resins, especially of the cycloaliphatic type, and, more recently, to ethylene-propylene rubbers as illustrated in U.S. Pat. Nos. 2,997,526; 2,997,527; and 2,997,528.

There have been several suggested modes of operation for the hydrated alumina, but whatever the correct mechanism, it is found in practice that polymeric materials containing large proportions of alumina trihydrate are substantially protected against tracking and usually fail only by progressive surface erosion. The amount of alumina hydrate required to produce the anti-tracking effect is very high, however, and is usually in the region of 50-90% by weight of the entire insulation. Especially in the case of polymers that are shaped by molding or extrusion as well as in the field of heat-recoverable articles, the high filler content is undesirable because the high temperature used and/or the radiation employed in cross-linking cause loss of the hydrated water with accompanying development of porosity and the formation of voids leading ultimately to failure of the insulation.

The present invention provides an electrically insulating material which comprises one or more polymers and an anti-tracking filler system comprising, (a) a hydrate of alumina have a specific surface area of at least about 2 m.sup.2 /g, and (b) a compound from the group consisting of oxides, mixed oxides and mixtures of oxides wherein said compound contains at least one element from the transition elements, the lanthanide series of the nontransuranic actinide series, as defined hereinafter. The resulting electrically insulating material has an initial tracking voltage of at least 2.5 kilovolts when tested according to ASTM-D 2303-68 liquid contaminant incline plane test, incorporated herein by reference.

By a "transition element" there is herein meant the elements of sub-groups IVa, Va, VIa, VIIa, and Group VIII of Mendeleef periodic table which are not also in the nontransuranic actinide series, e.g., titanium, zirconium, and hafnium, vanadium, niobium, and tantalum; chromium, molybdenum and tungsten; manganese, technetium, and rhenium; and iron, cobalt, nickel, ruthenium, rhodium, palladium, osmium, iridium and platinum.

By "lanthanide series" there is herein meant the elements cerium, praseodymium, neodymium, promethium, samarium, europium, gadolinium, terbium, dysprosium, holmium, erbium, thulium, ytterbium and lutetium.

By "nontransuranic actinide series" there is herein meant the elements thorium, protactinium, and uranium.

All references to "specific surface area" in this specification are to that measured by the Brunauer, Emmett and Teller (BET) nitrogen absorption method assuming the area covered by a nitrogen molecule to be 16.2 A.sup.2. (The BET method is referred to, for example, in "The Physics and Chemistry of Surfaces" by N. K. Adam published by Dover, and in "Solid Surfaces and the Gas-Solid Interface", Advances in Chemistry Series Vol. 33)

The alumina hydrate is preferably the trihydrate, Al.sub.2 O.sub.3.3H.sub.2 O, and preferably constitutes a major amount of the anti-tracking filler system.

It has surprisingly been found that the use of alumina hydrate of high surface area greatly enhances the properties of the insulating compositions of the present invention. It is preferable for the specific surface area of the alumina hydrate to be at least 3 m.sup.2 /g and advantageously greater than 6 m.sup.2 /g. Especially good results are obtained when the specific surface area lies in the range of from 8 to 20 m.sup.2 /g. Lower surface areas will still yield acceptable results. However, the surface area should not be below about 2.0 m.sup.2 /g to obtain the surprising improvement in insulating properties disclosed herein.

The hydrated alumina advantageously has a maximum particle size less than about 4 microns and preferably less than about 2 microns. Since the specific surface area of hydrated alumina is not directly dependent on particle size, it is understood that the invention is not limited to a particular particle size range.

The specific surface areas and particle size distributions of two forms of alumina trihydrate suitable for use in the present invention are given below.

Types A and B are sold by the Aluminum Company of America as "Hydral 705" and "Hydral 710", respectively. It is to be noted that the above surface areas are those claimed by Alcoa. Actual surface areas for samples of A have been found to vary from 6 to 21 m.sup.2 /g, averaging about 12 m.sup.2 /g.

Hydral 705 gives generally good results when used in the anti-tracking systems of the present invention, especially when the surface area is 10 m.sup.2 /g or higher.

Alumina hydrate of the desired specific surface area may be prepared by methods known per se, for example, by dissolving alumina in caustic soda and then reprecipitating it by bubbling carbon dioxide through the solution. According to the pH and the rate of bubbling, the necessary values of which can easily be determined by experiment, alumina hydrate of the desired specific surface area can be obtained.

Among oxides, mixed oxides and mixtures suitable as component (b) there may be mentioned, for example,

(i) Transition metal oxides, for example, TiO.sub.2, V.sub.2 O.sub.5, Cr.sub.2 O.sub.3, Fe.sub.2 O.sub.3, NiO.sub.2, Ni.sub.3 O.sub.4, Co.sub.2 O.sub.3, Co.sub.3 O.sub.4, MoO.sub.3, WO.sub.3, Nb.sub.2 O.sub.5, and mixtures thereof.

(ii) Lanthanide series oxides, for example, Pr.sub.6 O.sub.11, Yb.sub.2 O.sub.3, Ce.sub.2 O.sub.3, holmium oxide, erbium oxide, and mixtures thereof.

(iii) Nontransuranic actinide series oxides, for example, UO.sub.3, ThO.sub.2 and mixtures thereof.

(iv) Mixed oxides containing one or more transition metals, for example, nickel titanate, nickel molybdate and cobalt silicate.

(v) Any of the above oxides, mixed oxides or mixtures carried on an ahydrous alumina support, e.g., nickel cobalt molybdate on alumina and cobalt oxide on alumina CoO.Al.sub.2 O.sub.3 (blue cobalt aluminate).

(vi) Any of the above systems doped with small quantities of alkali metal or alkaline earth metal oxides, especially Li.sub.2 O and K.sub.2 O.

It is understood that the above listing is illustrative only, and is not intended to be a complete list of all the oxides which are operable in the invention.

The component (b), which is believed to react synergistically with the alumina hydrate in reducing tracking, may in some cases be used in quantities down to or below 0.5% by weight based on the total weight of the insulation material, but in general is preferably present in an amount in the range of from 2 to 10%, especially from 3 to 5%. Although amounts higher than 10% may be employed, little additional benefit to the tracking and erosion properties is gained thereby with most oxides. However, with some oxides of the invention e.g., ThO.sub.2, the oxide is preferably present in quantities of about 15% or more by weight.

The particle size and surface area of component (b) are not so important as that of the alumina hydrate but the best results appear to be obtained with materials having a particle size less than 75 microns (BS 200 mesh, U.S. Sieve No. 200) and, for ease of dispersion during preparation of the insulating material, it is preferable to use materials having a particle size less than 45 microns (B.S. 350 mesh, U.S. Sieve No. 325).

Compounds of iron, cobalt and chromium are especially preferred as component (b). Particularly good results are obtained when Co.sub.3 O.sub.4 is used as component (b). Ferric oxide, cobaltic oxide and chromic oxide are also very useful.

It has also been found that a given weight of some combinations of two or more substances as component (b) is more effective than the same weight of only one of the substances, i.e. a synergistic effect between the substances is noted. An especially good component is a mixture of Fe.sub.2 O.sub.3 and Co.sub.2 O.sub.3. Insulating materials incorporating a given weight of this mixture show substantially less erosion on testing than materials incorporating an equal amount of either Fe.sub.2 O.sub.3 or Co.sub.2 O.sub.3 alone.

One of the problems of the previously proposed anti-tracking compositions in which alumina hydrate is used alone has been the difficulty of obtaining good U.V. stability. Normally, for maximum weather and U.V. protection it is common practice to incorporate small quantities of carbon blacks, usually the channel type. Alternatively, organic U.V. absorbers such as, for example, substituted benzophenones or benzotriazoles are employed which are far less effective than carbon black, although they have the advantage that a variety of colors are obtainable using normal pigments. In the case of polymeric insulating materials for outdoor use, a lifetime exposure running into decades, typically 25-30 years, is required. Some polymeric insulating-materials are not sufficiently U.V. stable to last this length of time and it is therefore necessary to incorporate channel blacks into them. The effect of these channel blacks, even in small concentrations of the order of 0.3% or less, has, however, been catastrophic on the previously proposed insulation materials in that tracking failures have rapidly occurred.

One of the surprising features of the present invention is that carbon black fillers can be incorporated into the insulation, in amounts up to about 3% by weight, without failure by progressive creepage tracking occurring such materials failing only by erosion. This is obviously a very important advantage particularly where the base polymers do not have good U.V. resistance.

In general, the anti-tracking filler system will constitute from 20-75% of the total weight of the insulating material. However, owing to the synergistic effect between component (b) and the alumina hydrate it is possible to reduce the proportion to below 20% in some cases without loss of anti-tracking properties. This is especially useful in the formation of heat-shrinkable articles from the materials when proportions of from 20 to 35% are preferred to avoid the development of porosity during manufacture and to maintain the necessary elongation, modulus and tensile strength above the crystalline melting point. If the material is to be used in non-heat-shrinkable applications then it is preferred to increase the proportion of alumina hydrate since this effectively reduces erosion rates and also lessens the overall cost of the insulation. The preferred proportion of anti-tracking filler system (a) plus (b) thus falls within the range of from 30 to 60% of the total weight of the insulation in these cases.

In general any polymer normally used for high voltage insulation may be used in this invention. Among polymeric materials into which the anti-tracking system may be incorporated there may be mentioned polyolefins and other olefin polymers, obtained from two or more monomers especially terpolymers, polyacrylates, silicone polymers and epoxides, especially cycloaliphatic epoxides. Among epoxide resins of the cycloaliphatic type there may especially be mentioned those sold commercially by CIBA (A.R.L.) limited under the names CY 185 and CY 183. Particularly suitable polymers include polyethylene, ethylene/ethyl acrylate copolymers, ethylene/vinyl acetate copolymers, ethylene/propylene copolymers, ethylene/propylene non-conjugated-diene terpolymers, chlorosulphonated polyethylene, polypropylene, polydimethyl siloxane, diemthyl siloxane/methyl vinyl siloxane copolymers, fluoro silicones, e.g. those derived from 3,3,3-trifluoropropyl siloxane, carborane siloxanes, e.g. "Dexsil" polymers made by Olin Mathieson, polybutyl acrylate butyl/ethyl acrylate copolymers, butyl acrylate/acrylonitrile copolymers, butyl acrylate/glycidyl methacrylate copolymers, polybutene, butyl rubbers, ionomeric polymers, e.g. "Surlyn" materials sold by DuPont, or mixtures of any two or more of the above.

The present invention also provides a composition suitable for processing into the insulating material of the invention.

The present invention furthermore provides an anti-tracking filler system suitable for use in the insulating materials and compositions of the invention.

The present invention also provides an electrical component especially a cable, insulated by the insulating material of the invention.

The present invention also provides a shaped article made from an insulating material of the invention. The material may be cross-linked and, if desired, the property of heat-recoverability may be imparted to the article.

Among the many uses for the insulating materials of the present invention there may especially be mentioned the production of heat-shrinkable tubing, heat-shrinkable sheds and heat-shrinkable 3-core udders for cable connection and termination, especially at high voltages of up to 33 KV and higher (vide "Heat-shrinkable plastics for termination and jointing of power and auxiliary cables" D. Q. Ash, Ir Bagdzinski and R. J. T. Clabburn PROC.IEE Vol. 117, No. 7, July 1970 pp 1365-1372). These and other shaped parts are especially useful in the termination of high voltage cables to overhead lines, to transformers and to switchgear, especially in outdoor environments.

The insulating material of the present invention may also in some cases advantageously be applied to, for example, a termination in situ by application of the basic composition in the form of a lacquer in a suitable solvent, for example, toluene, xylene or carbon tetrachloride. In some cases, especially when the polymer component is a silicone, the composition may itself be sufficiently fluid for in situ application which will harden on standing.

It will be appreciated that although the primary purpose of the anti-tracking filler is to prevent creeptracking as outlined above, the materials will also be effective in stabilizing the insulation under arcing conditions, i.e. in cases where a direct arc passes between two parts of an electrical apparatus forming a carbonaceous track along its line. This phenomenon is slightly different from creep-tracking where, for example conducting contaminate and/or surface irregularities cause a leakage current and a dendritic carbonaceous path develops on the surface of the insulation.

The insulating material and compositions of the present invention may, if desired, contain other fillers, for example, flame retardants, reinforcing fillers, pigments and mixtures thereof.

The anti-tracking system can be incorporated into polymer(s) by any of the commonly used techniques, for example, in a twin-roll mill at elevated temperatures. Similarly the resulting compositions can readily be processed into sheets of material or other shaped articles by any of the usual methods.

The following examples illustrate the invention, parts and percentages being by weight unless otherwise stated. The surface area of the alumina trihydrate is 16.0 m.sup.2 /g unless otherwise indicated.

The following formulations were prepared by compounding on a twin-roll mill:

After thoroughly mixing all the ingredients in the order given, plaques 5.times.2.times.0.25 inches were pressed at 200.degree. C. for 10 mins. The cross-linked plaques were then tested according to the ASTM D2303-68 liquid contaminant inclined plane test which measures the tracking and erosion resistance of insulating materials. The test method used was the initial tracking voltage test in which the voltage is raised by 0.25 KV every hour and was identical to the ASTM test except that the voltage supply was 50 HZ and no silver paint was used in conjunction with the stainless steel electrodes. The silver paint is not necessary and its absence has no effect on the test result. The samples were used for testing approximately 24 hours after preparation and were stored at room temperature and humidity. They were not soaked in contaminant prior to test. The contaminant which was ammonium chloride had a resistivity of 380 ohm. cm. and contained 0.02% Triton X100 as wetting agent.

The results of the tests are given in Table 1.

Comparison of the results of the tests on Samples 1 to 7 clearly shows that whereas Samples 6 and 7 containing ferric oxide but no alumina hydrate track rapidly, and Samples 3, 4 and 5 containing alumina hydrate but no ferric oxide track after a relatively short period, Samples 1 and 2 containing both alumina hydrate and ferric oxide in accordance with the present invention are non-tracking and fail only by erosion after 200 and 270 minutes on test, respectively.

The results of the tests on Samples 3, 4 and 5 illustrate that in the absence of ferric oxide as component (b), an increased proportion of alumina hydrate does not prevent tracking.

The results of the tests on Samples 6 and 7, similarly show that, in the absence of alumina hydrate, an increased proportion of ferric oxide does not prevent tracking.

The results of the tests on Samples 1 and 2 show clearly the synergistic anti-tracking effect of a combination of alumina hydrate and ferric oxide.

A further experiment was carried out on Sample 1 to evaluate the rate of erosion. In this experiment, the voltage was maintained constant at 2 KV, and the contaminant contained 1% sugar in addition to the ammonium chloride. After 24 hrs. exposure to the scintillation a very small depression about 0.050 in. deep had eroded away, showing the extremely good erosion resistance. By comparison, Sample 3 tracked, under the same conditions after 150 minutes.

Test samples containing the following formulations were prepared as described in Example 1.

The following results were obtained using the test method described in Example 1.

Samples containing the following formulations were prepared and tested as in Example 1.

All samples contained 30 parts dimethyl silicone elastomer, 30 parts low density polyethylene, 30 parts ethylene/ethyl acrylate copolymer, 30 parts alumina trihydrate (Type A), 2 parts Agerite Resin D, 1 part triallyl cyanurate and 1 part 2,5-dimethyl-2, 5-di-t-butyl peroxy hexyne-3.

In Addition, Samples 10-19 contained 5 parts of the following components, respectively:

The results of the tests are shown in TABLE 2. It will be seen that Samples 13, 14, 16 and 20 gave especially good results and that all the samples show substantially better anti-tracking properties than Samples 3, 4 and 5, which contained only alumina hydrate.

Samples containing the following formulations were prepared and tested as described in Example 1.

The results are shown in Table 3.

These results show the improved anti-tracking effects obtained using a system of the present invention in two different types of polymeric insulation.

Samples containing the following formulations were prepared and tested as described in Example 1. The results are shown in Table 4.

These results show the improved effect obtained using ferric oxide as component (b) in another type of polymeric insulation.

Samples containing the following formulations were prepared and tested as described in Example 1, except that the ammonium chloride had a resistivity of 330 ohm cm. (thus making the test slightly more severe).

The results obtained were as follows:

Samples 28-32 prepared in accordance with the present invention, were non-tracking and failed only by erosion after 320, 384, 428, 395 and 420 minutes, respectively. These figures illustrate that, where component (b) is ferric oxide, increasing the proportion does not result in increased tracking protection and only has a small effect, if any, on the erosion of the materials. However, with other oxides disclosed as operative in the invention larger proportions of the oxide may be required to obtain the desired anti-tracking properties in the insulation.

Three further compositions containing cobalt silicate, cobalt aluminate, and brown iron oxide, respectively, as component (b) were prepared and tested as described in Example 1.

The starting voltage was 3 KV and the following results were obtained.

These experiments illustrate the importance of the surface area of the alumina hydrate. The following formulations were prepared as described in Example 1.

Test samples were tested as described in Example 1 but at a constant voltage of 3 KV, so that the time to track or erode at this voltage was determined. The results are presented in Table 6.

The above results show quite clearly that hydrates having a surface area below about 2.0 m.sup.2 /g do not prevent tracking, while, surprisingly, materials containing hydrates of about 2.0 m.sup.2 /g or higher surface area fail by erosion only. Although the mechanism by which alumina hydrates with a surface area above about 2.0 m.sup.2 /g combine synergistically with the disclosed metal oxides is unknown, Table 6 clearly discloses that where the surface area is about 2.0 m.sup.2 /g or over the hydrates confer increasing resistance to erosion as their surface area increases.

The surface areas were measured by the BET technique on a Micromeritics Surface Area Analyzer Model 2200 after initially degassing the samples for 40 minutes at 150.degree. C.

This Example illustrates compositions based on various other polymers.

The following formulations were prepared and tested as described in Example 1.

The following results were obtained:

This example illustrates the use of oxides of the lanthanide and actinide series.

The following formulations were prepared and tested as in Example 1.

The ASTM test was commenced at 3.0 kV and the following results were obtained.

The above figures show that, although in general the results obtained using these oxides are not so good as those obtained using transition metal oxides, there is a definite improvement in anti-tracking properties when compared to the use of alumina trihydrate alone.

This example illustrates the use of thorium oxide.

The formulations were prepared and tested as in Example 1, except that the ammonium chloride had a resistivity of 385 ohm cms., thus making the test slightly less severe.

The ASTM test was commenced at 2.5 KV, and the following results were obtained.

The test results in Table 9 demonstrate that thorium oxide performs according to the invention. Best results with thorium oxide are obtained by using alumina hydrate of a fairly high surface area and a moderate amount of the oxide. It is preferable to avoid use of the oxide in large quantities.

This example further illustrates the use of oxides and mixtures of oxides in the lanthanide series.

The following formulations were prepared and tested as in Example 1, except that the ammonium chloride had a resistivity of 385 ohm cms., thus, making the test slightly less severe.

Samples 61b-65b correspond to samples 61a-65a respectively and were prepared from the same components in the same amounts except that the alumina trihydrate had a surface area of 5.8 m.sup.2 /g.

The ASTM test was commenced at 2.5 KV. The results are presented in Table 10.

The test results in Table 10 demonstrate that the lanthanide oxides act synergistically, even at low surface areas, with alumina hydrate, giving increased resistance to tracking over alumina hydrate alone.

The test results in Table 10 may be advantageously compared with those presented in Example 8 to demonstrate that the surface area of the alumina trihydrate similarly affects the degree of anti-tracking resistance imparted to electrical insulators by all of the oxides and mixed oxides disclosed as operative in the invention. Thus, as shown in Example 8, with iron oxide, the time for erosion failure increases from 166 through 360 to 523 minutes as the surface area of the alumina hydrate is increased from 3.7 through 10.5 to 16 m.sup.2 /g. Similarly, in the case of ytterbium, the time for erosive failure increases from 147 to 253 minutes as the surface area is increased from 5.8 to 10.2 m.sup.2 /g.

In contrast, the effect of surface area variations in the alumina trihydrate when the latter is used alone appears to be minimal, as can be seen from the following Table drawn up from the comparative samples tested and reported elsewhere in this specification.

Table 10a, allowing for two apparently anomalous results, indicates that, in the absence of an additional filler in accordance with the present invention, the anti-tracking properties are virtually independent of the surface area of the alumina trihydrate and, implicitly, indicates a synergistic relationship between such fillers and the said surface area.

This example compares the effectiveness of a composition employing a metal oxide within the scope of the invention (calcined ferric oxide) with that of a metal oxide outside the scope of the invention (zinc oxide).

The following formulations were prepared and tested as described in Example 1.

The results obtained are reported in Table 11 below.

The above figures clearly show that the use of zinc oxide with alumina hydrate does not prevent tracking failure of the insulating material whereas the use of ferric oxide in the place of zinc oxide in an otherwise identical formulation completely eliminates tracking failure. Although sample 68, containing zinc oxide, took longer to fail by tracking than control sample 69, containing alumina hydrate alone, failure of sample 68 was still a tracking failure. This demonstrates that zinc oxide does not act synergistically with alumina trihydrate to give vastly improved resistance to tracking failure as do the oxides and mixed oxides of the invention.

This Example illustrates the use of the insulation materials of the present invention in heat-recoverable articles.

Using a formulation as specified in Example 1, Sample 1, a shed of internal diameter 0.49 inches was prepared by moulding at 190.degree. C. for 12 minutes. After moulding, the shed was cooled in water, trimmed and then heated in a glycerine bath at 170.degree. C. for 3 minutes. A PTFE mandrel of diameter 1.1 inches was then forced through the shed and then the mandrel plus shed was cooled in cold water for 5 minutes, whereupon the mandrel was removed. The shed then had an internal diameter of 1.05 inches. It was pore free. It was then reheated with a hot air gun to 170.degree. C. whereupon it shrank and completely recovered to an internal diameter of 0.49 inches.

Similar results were obtained with a larger shed of internal diameter 1.05 inches which was expanded to 2.2 inches, and then completely recovered.

Using the same formulation, but with the peroxide absent, tubing of the following dimensions was prepared by extrusion.

The tubing was irradiated using a 1.5 mev electron beam to a total dose of 12 megarads. The tubing had the following properties.

This tubing was then expanded in a Raychem tubing expander of the type described and claimed in British Pat. No. 990,235 at a temperature of 140.degree. C. The tubing obtained had an internal diameter of 1.10 inches. On heating, the tubing immediately recovered to its original internal diameter.

The tubing and sheds described above were successfully utilized to form a support insulator of the type described in British Pat. No. 1,292,276.

An epoxy polyamide/bisphenol A resin system was used as adhesive.