NO150376B - ELECTRICAL INSULATION MATERIAL BASED ON LD POLYETHYLENE AND ELECTRIC WIRE INSULATED WITH THE SAME - Google Patents

Description

This invention relates to a mixture based on non-cross-linked or cross-linked LD-polyethylene, for use as electrical insulation material, particularly for high-voltage cables. Furthermore, the invention relates to an electrical conductor which is insulated with such a mixture. The invention is particularly aimed at improving the resistance to electrical breakdown of insulating material consisting of or produced from the mixture.

Electrical breakdown of high-voltage insulation is often initiated by polluting particles. It is very difficult, if not impossible, to extrude solid organic insulation such as polyethylene onto a conductor without defects occurring. Although the polyethylene manufacturer is very careful about the purity requirements during production, contaminants can later be introduced in the subsequent handling of the resin before the final molding. Another cause of electrical breakdown in insulation is the presence of voids.

High voltage cables insulated with insulating polymers are subject to dielectric breakdown by a mechanism known in the art as electrical "wood growth". Wood growth is a relatively long-progressive breakdown of an insulation caused by electron-

and ion bombardment of the insulation, resulting in the formation of microchannels with a tree-like appearance, hence the name. A tree is initiated at points of contamination or voids that are foreign to the polymeric insulation, by ionization action (corona) under high voltage waves. Once a tree is formed, it usually grows, particularly under further high voltage waves, and after a certain time dielectric breakdown can occur.

To overcome this problem, various additives have been proposed, particularly in polyethylene or other polyolefins, which require an increase in the applied voltage to effect the initiation of a tree. This use of an additive is intended to prevent the breakdown of the insulation by preventing any formation of trees.

Maloney in US Patent 3,499,791 describes a coating for a high voltage electric cable comprising a polyethylene resin containing an inorganic ion salt of a strong acid and a strong Zwitter ion compound. An insulated cable provides resistance to electrical breakdown and voltage breakdown under the influence of corona.

Kato et al. describes in US Patent 3,956,420 insulation with improved electrical breakdown strength comprising polyolefin, a ferrocene compound and a substituted quinoline compound. In the patent there is also indicated the further use of a small amount of a polyhydric alcohol, dispersant, surfactant or unsaturated polymer or mixtures thereof to achieve another improvement in the electrical breakdown strength.

MacKenzie, Jr. describes in US patent document 3,795,646 an ethylene-containing polymer material which exhibits improved ionization resistance under high-voltage loading, by using a silicone liquid in a cross-linked polyethylene material.

Japanese patent document 14348/75 relates to wire cables with improved dielectric breakdown strength provided by an insulation of polyethylene containing 0.1% by weight of an aromatic ketone.

German patent document 2 147 684 indicates an increase in the electrical breakdown resistance of polymers, especially polyethylene, by modifying the free path for charge carriers (electrons) by incorporating additional scattering centers or by reducing the polymer's crystallinity.

Japanese patent application 7 201 988 describes insulated power cables with improved degradation resistance by providing an insulated layer of polyethylene, polypropylene, poly-carbamate or polyester containing mica particles coated with hydrophobic insulating material of silicone oil, stearic acid, palmitic acid or oleic acid.

Japanese patent document 49/119 937 describes electrical insulation resin materials which give an increase in the dielectric breakdown voltage by mixing in a resin material such as polyethylene a ferrocene-aldehyde (or ketone) polymer which has ferrocene groups or a mixture of the ferrocene-containing polymer and a higher alcohol.

It has now been shown that by incorporating smaller amounts of certain alcohols into LD-polyethylene, mixtures are obtained which, when used as insulating material, show markedly improved resistance to electrical breakdown.

According to the invention, a mixture based on non-crosslinked or crosslinked LD polyethylene is thus provided for use as electrical insulation material, particularly for high voltage cables, which polyethylene contains at least 85% polymerized ethylene units and at least 95% by weight polymerized olefin units. The mixture excels

characterized in that it consists of LD polyethylene and from 0.5 to 10% by weight, calculated on the polyethylene, of at least one straight-chain or non-straight-chain monovalent aliphatic alcohol with 6-24 carbon atoms.

Likewise, according to the invention, an electrical conductor is provided which is insulated with a mixture as indicated above.

An accelerated test procedure designated here as Test Method A shows that the presence of the alcohol inhibits electrical wood growth (but not wood formation) and results in at least a thousandfold increase in the electrical resistance of the polyethylene. The test is believed to give a direct correlation with long-term electrical resistance for the insulation on a conductor, i.e. an increased service life for the insulation - when it is used for its intended purpose.

As used herein, the term "polyethylene" is limited to a homopolymer or a copolymer containing not less than 85% by weight ethylene polymerized units, and not less than 95% by weight olefin polymerized units. These polymers will conform to the definition for "polyethylene plastics" as defined in the 1976 Annual Book of ASTM Standards, Part 36, page 70 as "plastics or resins prepared by the polymerization of not less than 85% ethylene and not less than 95% by weight total olefins ". A preferred polyethylene contains approx. 100% by weight ethylene polymerized units.

Suitable olefins which can be used as comonomers include propylene, butene-1, hexene-, octene-1 and decene-1. Other comonomers include norbornene, butadiene, styrene, methacrylic acid, vinyl acetate, ethyl acrylate, isobutyl acrylate, and methyl vinyl ether. LD polyethylene is suitable for use in the mixture according to the present invention. HD polyethylene is not suitable, because when HD polyethylene is used, little or no improvement, and no thousand-fold improvement at all, is achieved in the electrical resistance after the addition of an alcohol as specified here. "LD" (low density) indicates a polyethylene that has a density of up to 0.92 g/cm 3. Reference is made to ASTM

D 1248-74 for the expression "LD".

In addition, the stiffness of an insulation is a factor that must be taken into account when choosing polyethylene for certain applications, e.g. flexibility is required in a power line cable.

The necessary additive to the LD polyethylene is an aliphatic, monohydric alcohol with 6-24 carbon atoms, and preferably 8-12 carbon atoms. The alcohols can be straight chain or branched. Suitable examples are n-hexyl, n-heptyl, n-octyl, n-decyl, n-dodecyl, n-tetradecyl, stearyl and eicosyl alcohols, 2-decanol, 4-decanol, cyclohexanol, 3-methylhepta-nol-3 , 2-methyloctanol-2 and the like. The alcohols which are useful are also referred to in this description as "wood growth inhibitors".

The alcohol may be added in any conventional manner, including mixing with the solid polyethylene prior to extrusion, injection into molten polyethylene, and diffusion into solid polyethylene of alcohol applied by spraying, immersion, or vapor contact.

One test method, test method A, for determining whether an additive is suitable to increase the electrical resistivity at least a thousand times, makes use of non-crosslinked LD polyethylene and excludes the use of a peroxide crosslinking agent. Since conversion of non-crosslinked polyethylene to crosslinked polyethylene using a peroxide crosslinker can result in an increase in electrical resistivity, the addition of a peroxide crosslinker can mask to some extent the improvement provided by the wood growth inhibitor. However, a modification of Test Method A to make use of cross-linked polyethylene has been found to be satisfactory as a selection technique to show whether any improvement in electrical resistivity is achieved by the addition of an additive. However, such a modification of test method A is not sufficient in all cases to determine whether at least a thousandfold improvement in the electrical resistivity takes place.

In test method A, initialization of a tree is required i

an insulation test. Cross-linked polyethylene allows visual inspection of a tree. In contrast, non-crosslinked polyethylene is opaque and does not allow visual determination of a tree without cutting into the polyethylene. Therefore, in test method A, where non-crosslinked polyethylene is used, it is usually necessary to use several samples and destroy one of the samples to determine whether a tree has formed under the initial stress conditions.

The most preferred mixture according to the invention contains non-crosslinked LD-polyethylene, a peroxide crosslinking agent and a wood growth inhibitor consisting of at least one monovalent, aliphatic alcohol with 8-12 carbon atoms. This most preferred mixture is a precursor for an insulating material containing cross-linked polyethylene.

Conventional peroxide crosslinking agents well known in the art for crosslinking polyethylene can be used herein and include di-α-cumyl peroxide, 2,5-bis-(t-butylperoxy)-2,5-dimethylhexane, 2,5-dimethyl- 2,5-di-(t-butylperoxy)-hexyne-3, etc.

If a peroxide cross-linking agent is soluble in the alcohol, this agent can be dissolved in the alcohol and both added to the polyethylene. The polyethylene containing these additives is not normally cross-linked until after application to an electrical conductor.

For the purpose of giving an explanation, though without being. tied to some theory regarding the way in which the alcohol acts as a wood growth inhibitor, it is to say that the alcohol has the ability to diffuse through the LD polyethylene and into cavities in the insulation. Wood growth in the insulation follows the formation of microchannels and gives a wood-like appearance. Wood growth will usually continue until dielectric breakdown of the insulation takes place. In the present case, however, it is assumed that the alcohol diffuses into the cavities and prevents electron and ion bombardment. Inhibition of wood growth after initiation causes an increase in the insulation's electrical resistance and durability.

The underlying purpose of the test procedure is to mimic a mechanism that causes dielectric breakdown.

In practice, three are usually initiated in power cables under high-voltage waves, e.g. due to switching transients, lightning strikes, etc. Then breakdown of the insulation can occur at normal operating voltage, or especially during further high-voltage waves. In the mixture according to the invention, an alcohol which has an inhibitory effect on tree growth will extend the useful life of the insulation by inhibiting the growth of trees after their initiation and preventing premature breakdown of the insulation.

In contrast to conventional test procedures in the field, which determine the dielectric strength of an insulation, test method A is considered to give a correlation to the service life of the insulation. In this latter test, a high alternating voltage which is initially applied across electrodes placed in the insulation, will cause tree initiation without resulting breakdown of the insulation. This tree initiation is followed by a rest period during which no voltage is applied for at least approx. 24 hours. Then the time until breakdown of the insulation is measured with 12,000 volts applied between electrodes separated by 2 mm, i.e. an average electric voltage, if the field was uniform, of 6,000 volts/mm (as described in test method A). However, the field is intensified as a result of the small diameters of the shaped ends of the electrodes, to a value greater than 6,000 volts/mm.

The accelerated test for determining the electrical resistance is believed to provide a useful correlation to the extended lifetime of insulation used for a longer time, e.g. at least 30 years. It is naturally impractical to carry out such a long-term test. Only relatively few samples will actually fail in long-term testing, and a statistical investigation would be necessary. With the presence of the alcohol in a mixture consisting mainly of LD polyethylene, it is assumed that in normal use no failure based on dielectric breakdown of the polyethylene will occur.

The test for determining an increase in the electrical resistance is specified as test method A and includes the following:

Test method A

Polyethylene for testing according to this method is first cast in a block indicated here as a "SPING" (which is an abbreviation for "solid phase internal needle gap specimen"). A SPING is square with a side edge of 25 mm and is 6 mm thick, and it contains two electrodes embedded in it lengthwise and in line, equidistant from the faces and from opposite edges, with the tips 2 mm apart at the center of the block. Each electrode is approx. 30 mm long and approx. 0.6 mm in "diameter. One of the electrodes has a conical tip with an angle of 30° and a radius of 5 ym and is the high voltage electrode; the other electrode has a hemisphere of radius 0.3 mm grounded in one end and is the grounding electrode.

At least five SPINGS are used in the test at the same time. Each ■ SPING is placed under silicone oil, which prevents surface spillage. The high voltage electrode connects to a high voltage rail, while the ground electrode connects to a separate pair of 6.25 cm balls connected to ground through a 1 megohm resistor. A sufficiently wide gap is arranged between the balls to achieve a voltage sufficient to initiate a tree in the SPING. With e.g. the spheres arranged with a gap of 0.76 2 cm is applied a voltage (60 Hz) increasing at a rate of 500 volts/sec until a discharge takes place between the two spheres. Before this penetration takes place, the stress on the test piece is essentially equal to zero. However, at the moment the air gap breaks down, the applied voltage plus a radio frequency signal produced by the arc is applied across the test piece electrodes and maintained for 1-2 seconds, so that a tree develops. The voltage required to initiate a tree will vary with the mixture being tested. For a mixture of LD polyethylene and an alcohol according to the present invention, a voltage of 35-40 kV is necessary. For LD polyethylene containing other additives, the required tension may be higher or lower, but the tension to be applied can be easily determined by visual examination of each test piece to see if a tree has been initiated.

After the tree has been initiated, the SPING is held without applied voltage for approx. 24 hours before placing it in a high voltage of 12,000 volts applied between the electrodes (an average applied voltage of 6,000 volts/mm). The time in hours before the average sample fails (e.g. the third out of five, the fifth out of nine, counted in time until failure) is measured and referred to as the electrical resistance.

Failure is indicated by dielectric breakdown. When failure occurs, a tree will connect the two electrodes and result in a sudden increase in amperage (which can be measured on a recording ammeter) which terminates the test for that specimen.

The wood growth inhibitor used according to the present invention works in a different way than known additives when it comes to achieving an improved result. The previously known additives are generally designated as compounds which, when incorporated into polyethylene or other suitable insulating materials, will require a higher characteristic voltage to initiate a tree at a needle tip. (These methods generally use a sharp needle tip embedded in the sample. The manner in which another electrode is present varies.) The characteristic voltage is the voltage at which half of the samples in the test will initiate a tree within one hour. This is determined by examining several groups of test pieces at several different voltages. The test ends when the characteristic voltage is found. .The purpose of a tree growth inhibitor is not to prevent tree initiation, but only to suppress the growth of a tree after this has been initiated.

Although the mixture according to the invention mainly consists of LD polyethylene and an alcohol, it should be understood that other conventional additives can be, and are normally present in the mixture. These additives include antioxidants, e.g. polymerized trimethyldichloroquinone, lubricants, e.g. calcium stearate, pigments, e.g. titanium dioxide, fillers, e.g. glass particles, and reinforcements, e.g. fiber materials such as asbestos and glass fibres, etc.

Although an insulation of non-crosslinked LD-polyethylene or cross-linked LD-polyethylene containing a wood growth inhibitor is particularly suitable for a power cable intended for a voltage of at least 15 kV, e.g. 15-220 kV, it is also suitable for lower or higher voltages. In an electric cable, in accordance with the state of the art, a semi-conductive layer will be placed between an electric conductor and an insulation layer. Such a semi-conductive layer usually includes an insulating material which also contains carbon black.

The following examples illustrate the invention:

Insulation mixture for control A and ex. 1- 4

(A) LD polyethylene: Homopolymer

Melt index (ASTM D-1238) 1.8 g per 10 minutes

Density 0.918 g per cm 3 (measured according to ASTM

D-1505-68 (Reapprove 1975))

(B) Antioxidant: 4,4<1->thiobis-(6-tert-butyl-m-cresol), 1500 ppm (C) Tree growth inhibitor: n-dodecyl alcohol, except for cont- troll A

Examples 1 and 2

In Example 1, n-dodecyl alcohol was added to LD polyethylene pellets by drum mixing followed by extrusion mixing. In example 2, the n-dodecyl alcohol was injected using a gear pump into molten LD polyethylene in the mixing zone of the extruder. In each of Examples 1 and 2, the final concentration of the n-dodecyl alcohol was 3% by weight in the polyethylene, measured by IR spectrometry. Nine SPINGS were prepared for each example and used by test method A.

The nine SPINGS in examples 1 and 2 were taken out of the test without any having failed after 1960 hours and 1730 hours respectively. The electrical resistance would thus be over 1960 and 1730 hours respectively.

Further testing of the same SPINGS was carried out. The nine SPINGS for examples 1 and 2 were later withdrawn from the test without any failure after a total of 4000 hours. In each case, the electrical resistance would thus be over 4000 hours.

Comparison of examples 1 and 2 with control A shows that the increase in the electrical resistance would be well over a thousandfold.

Examples 3 and 4

Also in these examples, the additive n-dodecyl alcohol was in a concentration of approx. 3% by weight in the LD polyethylene as measured by IR spectrometry. In Example 3, the polyethylene was mixed with the alcohol in a Banbury mixer, while the mixing in Example 4 was carried out in a Brabender mixer. Nine SPINGS were prepared from the mixture in each example and tested by Test Method A. All SPINGS according to Example 3 were removed from the test after 850 hours, while all SPINGS according to Example 4 were removed from the test after 720 hours. No SPINGS had failed either in example 3 or 4. The electrical resistance was thus greater than 850 and 720 hours. The increase in electrical resistivity compared to control A was over a thousandfold.

In the remaining examples (Examples 5-10) and in controls B-D, test method A was still used, but with the minor changes that each electrode was 1.0 mm in diameter (instead of 0.6 mm), and that the other electrode was grounded at one end to a hemisphere of radius 0.5 mm (instead of 0.3 mm). These changes were made only to simplify the fabrication of the shape of the electrode end by machining, as a thickness of 1.0 mm gives less flexibility than 0.6 mm. Parallel tests carried out with the thicker and thinner electrodes in SPINGS made from the same insulation mixture determined that the same test results were obtained in each case. The test procedure that makes use of thicker electrodes can thus still be referred to as test method A.

Examples 5-8

In these examples, the insulation mixture was similar to that used in examples 1-4. It was still a polyethylene homopolymer with a density of 0.918, but had a melt index of 2.5 and contained approx. 750 ppm of the same antioxidant,

in

and various tree growth inhibitors were used. In each case, the amount of wood growth inhibitor in the polyethylene was 3% by weight.

In each of Examples 5-8, the wood growth inhibitor was added to polyethylene pellets by drum mixing followed by extrusion mixing. Four SPINGS were prepared for each example and tested by test method A. (When four SPINGS were used in the test, the electrical resistive force would be greater than the time to failure of the second SPING, but less than the time to failure of the third SPING.)

In Example 5, the tree growth inhibitor was n-dodecyl alcohol, a primary alcohol. All SPINGS were removed from the test after 600 hours. No SPINGS had failed. The electrical resistance was thus greater than 600 hours.

In Example 6, the tree growth inhibitor was cyclohexanol,

a primary alcohol. All SPINGS were removed from the test after 768 hours. No SPINGS had failed. The electrical resistance was thus greater than 768 hours.

In Example 7, the wood growth inhibitor was 2-decanol, a secondary alcohol. All SPINGS were removed from the test after 552 hours. No SPINGS had failed. The electrical resistance was thus greater than 552 hours.

In Example 8, the wood growth inhibitor was 4-decanol, a secondary alcohol. All SPINGS were removed from the test after 600 hours. No SPINGS had failed. The electrical resistance was thus greater than 600 hours.

In all of Examples 5-8, the increase in electrical resistivity compared to Control A was over a thousandfold.

Controls B, C and D

In controls B-D HD polyethylene was tested.

Control B used polyethylene having a density of 0.960 g/cm 3 and containing 100 ppm "Irganox 10-10" antioxidant. The polyethylene pellets and 3% n-dodecyl alcohol were mixed by drum mixing followed by extrusion mixing.

In control C, 88.11% by weight of pellets of the same polyethylene with a density of 0.960 g/cm 3 and 11.89% by weight of pellets of polyethylene with a density of 0.918 g/cm 3 (containing 700 ppm "Sanotox R" antioxidant) were drum mixed and further mixed by extrusion in a twin screw extruder and pelletized. Control D was the same, except that 52.38% by weight polyethylene with a density of 0.960 and 47.62% by weight of polyethylene with a density of 0.918 were used. In each case, the pellets in the blends were then drum blended with 3% n-dodecyl alcohol, followed by extrusion blending.

SPINGS of each mixture were then prepared for testing purposes. A small chip of polymer was removed from a SPING

of each mixture to measure its density (ASTM D 1505-68).

: 3

The density was: Control B - 0.957 g/cm

Control C - 0.949 g/cm<3>

Control D - 0.936 g/cm<3>

Five SPINGS of each mix were tested by test method

aJ In each of the controls B, C and D, all SPINGS failed before a day had passed. Accordingly, the electrical resistance in each case was less than 24 hours.

Claims (14)

1. Mixture based on non-crosslinked or crosslinked LD-polyethylene for use as electrical insulation material, in particular for high-voltage cables, which polyethylene contains at least 85% by weight of polymerized ethylene units and at least 95% by weight of polymerized olefin units, characterized in that it consists of LD- polyethylene and from U.5 to 10 by weight, calculated on the polyethylene, of at least one straight-chain or non-straight-chain monohydric aliphatic alcohol with 6-24 carbon atoms. 2. Mixture according to claim 1, characterized in that the polyethylene contains 100% by weight of polymerized ethylene units. 3. Mixture according to claim 1 or 2, characterized in that the alcohol contains 8-12 carbon atoms. 4. Mixture according to claim 3, characterized in that the alcohol is n-decyl alcohol. 5. Mixture according to claim 3, characterized in that the alcohol is n-dodecyl alcohol. 6. Mixture according to claims 1-5, characterized in that the alcohol content is from 1 to 5% by weight. 7. Mixture according to claims 1-6, characterized in that the density of the LD polyethylene is 0.918 g/cm<3>. 8. Electrical conductor insulated with a mixture based on non-crosslinked or crosslinked LD polyethylene, which polyethylene contains at least 85% by weight of polymerized ethylene units and at least 95% by weight of polymerized olefin units, in particular a high-voltage cable, characterized in that the mixture consists of LD polyethylene and from 0.5 to 10% by weight, calculated on the polyethylene, of at least one straight-chain or non-straight-chain monovalent aliphatic alcohol with 6-24 carbon atoms. 9. Electrical conductor according to claim 8; characterized in that the polyethylene contains 100% by weight of polymerized ethylene units. 10. Electric conductor according to claim 8 or 9, characterized in that the alcohol contains 8 - 12 carbon atoms. 11. Electrical conductor according to claim: 10, characterized in that the alcohol is n-decyl alcohol. 12. Electrical conductor according to claim 10, characterized in that the alcohol is n-dodecyl alcohol. 13.jElectrical conductor according to claims 8-12, characterized in that the content of alcohol in the mixture is 1 - 5% by weight. 14. Electrical conductor according to claims 8-13, characterized in that the density of the LD polyethylene is 0.918 g/cm<J>.