CA2220495C - Water resistant electrical insulation compositions - Google Patents

Abstract

An electrically insulating cross-linked polyethylene composition for use in high voltage electrical cables, said cross-linked polyethylene being obtained by cross-linking a composition consisting essentially of a low density, peroxide cross-linkable polyethylene, 1-2% w/w of a terpolymer of ethylene, vinyl acetate and vinyl alcohol and at least 0. 15 % w/w of a sterically hindered amine stabiliser of said composition;
said terpolymer being obtained from about 40-45%
hydrolysis of an ethylene-vinyl acetate copolymer having a 20-30% w/w vinyl acetate content; and wherein said polyethylene and said terpolymer have substantially the same melt index. The cross-linked polyethylene compositions have improved resistance to moisture induced degradation.

Description

CA 0222049~ 1997-11-07 Water Re~istant Electrical Insulation ComPOSitiOnS

Field of the Invention This invention relates to polyethylene compositions for use as high voltage electrical cable insulation and more particularly to such insulation having enhanced resistance to moisture-induced degradation.

Backqround to the Invention In the mid 1960's, underground power distribution cables were insulated with high molecular weight polyethylene (HMWPE). Although PE is one of the most moisture resistant polymers available, these cables started to fail in service after three to four years. By the time these cables were ten years old, failure rates became a financial burden to electrical utilities and the lack of reliable service to consumers became an issue.
Notwithstanding a lack of in-depth knowledge of the failure mechanisms, it was recognised that the insulation system deteriorated with time due to the combined action of moisture, temperature and electrical stress.
Subsequently, insulation compositions of newer technologies, such as cross-linked polyethylene (XLPE) and ethylene-propylene rubber (EPR) were introduced in the market place.
With the recognition that water is of primary concern in high molecular weight polyethylene (HMWPE) cable failures and with an objective to assure that cables operate reliably on an electric power utility system, the Association of Edison Illuminating Companies (AEIC), as a cable specification body, introduced an accelerated test to evaluate the effect of moisture directly on full size cables to determine AC breakdown strength and impulse strength of aged cables. Cable qualification tests were thus made very stringent with the aim to improve the operational reliability of a distribution network CA 0222049~ 1997-11-07 consisting of these cables and to obtain a better performing cable in service.
The mechanism by which the electrical strength of cable insulation is reduced by moisture induced degradation is still not understood. Tree-like patterns, named water trees, which occur during accelerated aging and field aging of the cable insulation are generally believed to be responsible for this degradation process.
These water trees comprise water-filled micro cavities, which originate inside the insulation, usually from a void, imperfection or a cont~m-n~nt and grow in the electric field direction. The trees can also originate at the insulation interface with the semiconductive polymer compound applied as a shield between the conductor and insulation, and on top of the insulation. Water trees, once initiated, slowly develop and lead to a reduction in dielectric strength of the insulation. However, because of the lack of a proper definition of tree retardancy and a substantial correlation to the cable aging process, it is not evident or obvious that a water tree retardant insulation will automatically result in improved service reliability. For example, contrary to what would be expected of EPR, which is a non-water tree retardant insulation, cable insulated with EPR is more or equally reliable to the performance of an improved or tree retardant XLPE. This, however, did not impede a proliferation of fundamental studies on water tree growth.
Based on screening work employing moulded samples or model cables, numerous methods to improve the performance of XLPE insulation against dielectric deterioration by water tree generation and growth have been described in the literature. US Patent No. 4,144,202 issued March 13, 1979, to Ashcraft et al relates to the inhibition of water tree growth by use of certain organosilane compounds. US patent 4,206,260 describes compositions containing an effective amount of an alcohol containing 6-24 carbon atoms as being an efficient water - - - - - - - -CA 0222049~ 1997-11-07 and electrical tree retardant insulation. German patent 2,737,430 discloses that certain alkoxysilanes act as tree retardant additives in polyethylene insulation. European patent 0,166,781, published January 8, 1986 to Sumitomo Electric Industries Limited describes a blend of ethylene and vinyl acetate copolymer as a water tree retardant material. Certain aliphatic carboxylic acid derivatives when incorporated in suitable amounts in XLPE are also reported to suppress water tree growth. Japanese application 63-226,814 published September 21, 1988 and Canadian application 2,039,894 published October 6, 1992 to Sarma et al disclose an insulation composition comprising a low density PE in admixture with an ethylene-vinyl acetate-vinyl alcohol copolymer as a water tree retardant composition.
It is also recognised in the industry that apart from moisture induced degradation of cable insulation, an additional gradual degradation in dry conditions leading to electrical tree initiation occurs at electric fields much lower than the breakdown strength of the insulation.
US patent 4,870,121 issued September 26, 1989 to Bamji et al. discloses the use of ultraviolet stabilisers, preferably in combination with reduced concentration of oxygen in the polymer, to significantly extend the time to initiation of electrical treeing by preventing photo-degradation of the polymer. It is, however, not obvious to one skilled in the electrical art that these additives or others cited in the literature for suppressing electrical trees would be stable and resistant to a moisture induced degradation process. Neither this patent nor any other teaches the possibility of W light emission during a moisture induced degradation process, nor the effect of activation or deactivation or extraction of these stabilisers by moisture in the insulation. The process of conversion of water trees to electrical trees leading to the final breakdown of the insulation and the influence of these additives on this conversion process CA 0222049~ 1997-11-07 , are other factors not taught by the prior art on electrical treeing.
Thus, there remains a need for a moisture resistant polyethylene composition for use in a high voltage electric cable, which, notwithstanding being a water tree retardant, should have a longer time to failure.

SummarY of the Invention It is an object of an aspect of the present invention to provide an improved electrical insulation composition for use in high voltage electrical cables having enhanced resistance to moisture induced degradation.
Accordingly, an aspect of the present invention - provides an electrically insulating cross-linked polyethylene composition for use in high voltage electrical cables, said cross-linked polyethylene being obtained by cross-linking a composition consisting essentially of a low density, peroxide cross-linkable polyethylene, a terpolymer of ethylene, vinyl acetate and vinyl alcohol and at least 0. 15 % w/w of a sterically hindered amine stabiliser; said terpolymer being partially hydrolyzed ethylene vinyl acetate copolymer; and wherein said polyethylene and said terpolymer have substantially the same melt index, said composition being optically transparent after heating to 120~C in hot oil for 15 minutes. Optical transparency is measured by the procedure of AEIC CS5-94.
A further aspect of the present invention provides an electrically insulating cross-linked polyethylene composition for use in high voltage electrical cables, said cross-linked polyethylene being obtained by cross-linking a composition consisting essentially of a low density, peroxide cross-linkable polyethylene, a terpolymer of ethylene, vinyl acetate and vinyl alcohol and at least 0. 15 % w/w of a sterically CA 0222049~ l997-ll-07 hindered amine stabiliser; said terpolymer being a partially hydrolyzed ethylene vinyl acetate copolymer with from about 40-45% hydrolysis and formed from an ethylene-vinyl acetate copolymer that has a 20 - 30 % W/W vinyl 5 acetate content; and wherein said polyethylene and said terpolymer have substantially the same melt index, said composition being optically transparent after heating to 120~C in hot oil for 15 minutes.
Another aspect provides an electrically insulating cross-linked polyethylene composition for use in high voltage electrical cables, said cross-linked polyethylene being obtained by cross-linking a composition consisting essentially of 98 % W/W of a low density, cross-linkable polyethylene, 1-2 % W/W of a terpolymer of 15 ethylene, vinyl acetate and vinyl alcohol and at least 0.
15 % W/W of a sterically hindered amine stabiliser; said terpolymer being obtained from about 40-45 % hydrolysis of an ethylene-vinyl acetate copolymer having 20 - 30 % W/W
vinyl acetate content; and wherein said polyethylene and 20 said terpolymer have substantially the same melt index.
The low density ethylene polymer (LDPE) of value in the practice of the invention has a density of about 900 to 950 kg/m3 (ASTM 1505 test procedure with conditioning as in ASTM D1248) and a melt index (MI) of 25 about 0. 5 to 10 decigrams per minute (ASTM D1238 test procedure). It may be made by a variety of techniques known to those skilled in the art of manufacture of low density polyethylene, for example, under high pressure using a tubular or autoclave reactor with any of the known 30 free radical initiators or coordination catalysts, using slurry, solution or gas phase polymerization techniques with coordination catalysts, including the catalysts known as metallocene catalysts and the transition metal catalyst systems such as Zeigler-Natta catalysts. Some of the low 35 density ethylene polymers obtained in such polymerization processes are frequently referred to as linear low density polyethylene. All such low density ethylene polymers are CA 0222049~ 1997-11-07 cross-linkable low density polyethylene of the compositions of the invention.
The ethylene-vinyl acetate-vinyl alcohol terpolymer of use in the practice of this invention will, hereinafter, be referred to as EVA(OH) terpolymer and is preferably obtained by the hydrolysis of ethylene-vinyl acetate (EVA) copolymer having a vinyl acetate (VA) content of approximately 20-30% w/w and especially about 25% w/w. Hydrolysis to give the terpolymer is carried out on the EVA copolymer in a reactive extrusion as a continuous process. Hydrolysis of the acetate group to the hydroxyl radical is to effect a minimum of 38 hydrolysis, preferably 40-45 % hydrolysis, under hydrolysis conditions known to the art. The amount of the terpolymer used in the composition is preferably limited to a maximum of 2% w/w. The terpolymer preferably has a melt index matching as close as possible that of the ethylene polymer.
Sterically hindered amines of use in the invention are present in the insulation composition at a minimum of about 0. 15 % w/w, preferably about 0. 3 % w/w, and may be either low molecular weight, polymeric or graftable to the polyethylene.
Examples of polymeric sterically hindered amine light stabilizers (HALS) of use in the practice of the invention are compounds of the general formula:
1. HALS of Formula I:
-H H
>

'~ ~r N--~CH2) 6--N
N ~N
Cl ~3 ~CH3 NH-C-CH2-f--CH3 --n where n - 2, 040 to 3, 500 S

CA 0222049~ 1997-11-07 having a melting point of 115~C - 150~C, density of 1.01 g/CM3 (20~C) and a pH (at 100 g/l water) of 8.5.
2. HALS of Formula II, being poly-6-morpholino-s-triazine=2,4-diyl((2,2,6,6-tetramethy-4-piperidyl)imino)hexamethyl((2,2,6,6-tetramethyl-4-piperidyl)imino) having a softening range of 110~C -130~C and volatility during thermogravimetric analysis (heating ratelO~C/minute), such that a 10%
weight loss is observed at 340~C.
3. HALS of Formula III, having a formula (C20Hs2N4) n where n = 1-12, and as further disclosed in U.S. Patent No. 4,104,284 (1978) having a melting point of 93~ -98~C (199~ - 208~F) and an average molecular weight of approximately 2000. As measured by thermogravimetric analysis, it exhibits essentially no weight loss up to 300~C (572~F).
Examples of such amines are as follows:
AO-1 poly-(N-~-hydroxyethyl-2,2,6,6-tetra-methyl-4-hydroxy-piperidyl succinate) AO-2 bis-2,2,6,6-tetramethyl-4-piperidyl sebecate AO-3 2-hydroxy-4-n-octoxy-benzophenone AO-4 poly-{[6-(1,1,3,3-tetramethylbutyl)-imino]-1,3,5-triazine-2,4-diyl][2-(2,2,6,6 tetramethylpiperidyl)-amino]-hexamethylene-[4-(2,2,6,6-tetramethylpiperidyl)imino]} AO-5 N,N'-bis[2,4-di(allylamino)-1,3,5- triazinyl-6]-N,N'-bis-(2,2,6,6-tetramethylpeperidyl-4)-hexamethylene diamine The preferred hindered amine used for a favourable advantageous synergistic effect is AO-4, which is sold under the Trademark "Chimasorb 944", by Ciba-Geigy, Basle, Switzerland.
Chemical cross-linking agents, either alone or in combination with a coagent, are used in appropriate amounts to cross-link the polymer insulation as would be well-known in the art. A preferred cross-linking agent is dicumyl peroxide. Other organic peroxides which can be employed are, for example, ditertiarybutyl peroxide (DTBP) and tertiarybutyl peracetate (TBPA).
The insulation composition of the invention preferably includes a further stabiliser such as the phenolic ester, sold under the Trademark " Irganox 245 "
by ~iba-Geigy, to obtain satisfactory cross-linking thermal stability.
Other additives which may be employed in the composition of the present invention include, for example, plasticisers, coupling agents, colorants and chelating agents.
Thus, this invention provides a synergistic combination of hydrolysed ethylene-vinyl acetate copolymer and a hindered amine light stabiliser as additive in a polyolefin dielectric composition.
In a further aspect the invention provides a high voltage electrical cable comprising an insulating layer of a cross-linked polyethylene composition as hereinabove defined.
The cable according to the invention may be manufactured according to known processes in the art wherein a conducting core to be coated is pulled through a heated extrusion die, generally a cross-head die, and layers of conductor shield, insulation and insulation shield are applied to the core. The coated cable then passes through a high pressure vulcanisation system to cross-link the polymeric layers.
Descri~tion of Preferred Embodiments In order that the invention may be better understood, preferred embodiments will now be described with reference to the following examples.
The insulation compositions of use in the present invention having surprisingly enhanced resistance to moisture induced degradation, of which water tree CA 0222049~ 1997-11-07 growth resistance forms only a part thereof, were tested to meet two standards. One standard was to verify the susceptibility to water treeing while simultaneously complying to all other electrical and non-electrical 5 requirements stipulated by the industry standards such as AEIC Cable Qualification CS5-94. In addition, the expected service reliability standard was also verified by accelerated life time tests on real size cable as stipulated by industry standards.
Extra clean compounds with contaminants of size < 2 mil at less than 1 per 12 c.c., were used in all testing. To allow an efficient inspection of raw material insulation and also final inspection of the cable insulation in its integrity with other compounds such as 15 conductor shield and insulation shield, the insulation compound was transparent after cross-linking. This transparency was examined under silicone oil heated to 120~C. When the cable composition was heated beyond the crystalline melting point of the insulation, it was rendered transparent and facilitated the optical ex~m'n~tion of the quality of the interfaces between insulation and semiconductives.
One physical property which is relevant to service behaviour is the dielectric loss of the cable insulation at operating (25~C-35~C), emergency (90~C) and short circuit (130~C) conditions. AEIC specification calls for a maximum value of 0.5% at any of these measurement temperatures. Insulation compositions with dielectric loss values greater than 0.5% are not 30 considered suitable for practical service applications.
Cable qualifications also deal with an accelerated water tree test looking at the changes in the dielectric strength after 120 days of aging. The cables are aged in water filled pipes under an electric field of 35 150 volts/mil with a load current that heated the conductor to 90~C for 8 hours and with the load current being off for the next 16 hours. These conditions are CA 0222049~ 1997-11-07 maintained for 120 days after which the AC breakdown strength is measured. The specification requirement is 620 volts/mil before aging and 260 volts/mil after aging, signifying the necessity of an improved product for service reliability.
Accelerated life tests are conducted with the cables in water-filled tanks. Fifteen kV rated cables with 0. 175 in (4.4 mm) of insulation are energised to four times the operating voltage in a set of twelve in a series circuit, submerged in a water filled tank with water inside the conductor strands. Voltage is maintained continuously while the cables are load cycled with current to a conductor temperature of 90~C for eight hours each day. The cables are tested to failure and the data analyzed by Weibull statistics. The results are represented as "an, the mean time for 63.5% failure probability and ~n I the statistical spread in the data.
Although the correlation of actual service life with the data derived from this test has not been quantified, it is generally agreed by those skilled in the art that a cable system with an improved test performance i.e. longer time to failure, will also carry over into its field performance reliability and extension of life.
The susceptibility of the insulation composition to water tree growth was tested on moulded samples of the same. This laboratory material test utilised compression moulded dish shaped specimens with built-in protrusion defects of 20 micrometer tip radius. The experimental design and test procedure is identical to that described in aforesaid Canadian patent application No. 2,039,894.
Tests at an insulation temperature of 65~C were also carried out to simulate the operating conditions of a commercial size cable. Dielectric loss measurements on pressed and cross-linked plaques were also carried out in parallel to cable measurements.
Under no circumstances was it considered CA 0222049~ 1997-11-07 permissible to realize enhanced resistance to moisture induced degradation behaviour at the expense of other properties relevant to service behaviour and operational reliability.
Commercially available insulation compound HFDA
4202 supplied by Union Carbide Chemicals and meeting the cleanliness requirements as specified by AEIC was used for comparison. This compound is based on the disclosure in US patent 4,144,202 and believed to contain approximately 10 5 % w/w of a silane of the formula C6HsCH=N(CH2) 3Si (OCH2CH3) 3 Results The results illustrate how a cross-linked synergistic specific EVA(OH) terpolymer and a hindered amine in polyethylene renders the resulting composition more resistant to moisture induced degradation of its electrical properties thereby resulting in a cable of improved reliability. All compositions and results are described in the following examples.

Example 1 In~ulation Com~o~ition:

Components % w/w Polyethylene 92.6 EVA(OH) 5 Irganox 1035 0.15 Irganox 802 0.35 Dicumyl peroxide 1.9 * Irganox 1035 is a hindered phenolic ester and Irganox 802 a thioester, both supplied by Ciba Geigy. EVA(OH) is a terpolymer of melt flow index of 5 g/lOmin made from EVA
copolymer of 28% VA, with a degree of hydrolysis of 98%
and supplied by Tosoh, Japan.

CA 0222049~ l997-ll-07 The number density of the contaminants of different sizes in this compound was measured using a Intec Laser Inspection system and found to comply to the AEIC cable requirements. The typical values were:

Defects/cu.in of Exam~le 1 A B C D
(< 2mil) (2-5mil) (5-lOmil) (~ lOmil) 0.13 0.65 0.06 0.003 A 15 kV rated cable with this compound as insulation and commercially available semiconductive shield compounds (AT plastics Compounds AT 377 and AT389) was extruded onto a 1/0 AWG (19 wires) compressed Aluminium conductor. Concentric neutrals were then laid on these cables. A reference cable with HFDA 4202 as insulation was also simultaneously extruded.
Accelerated Cable Life Tests with Exam~le 1 These cables were tested in water filled tanks using the following experimental parameters.
(1) Test voltage 4 x Vg (35 kV; average stress of 200 volts/mil).
(2) Test frequency 60 Hz.
(3) Load cycling with 90~C conductor temperature in air, 8 hours on and 16 hours off load cycle.

(4) Tank water temperature maintained at 50~C + 1Ø

(5) Ten 30 ft cable samples wound into two loops (2 x 360 with an active length of 25 ft in each water tank.

(6) Cable strands and tanks filled with deionised water.~5 (7) Tests conducted until 6 out of 10 samples fail, with the remaining four samples being used to measure residual breakdown strength after aging.

~tNI' BY: CA 0222049~ IM997 1l 071M ~~ McBLlRNEy- B19 953 953~:# ~ 2 t ~f~ ton t~ mr~l ~ rnm~ i t.i ~ Ui~
Exa~ple I in the form o~ ln~ulation fail~ witl~in lG00 hour~ (fi~ day~) of te~t time. There were no failure~
amon~ cable~ with HFDA 4202 a~ in~ulation. The ~ailure S d~ta is a~ reported below:

Cahle F~ilurç Dat~ with Ex~mPle 1 Sample No. Time to fall Comment ~hours) 1 350.7 failed in air 19" above ~ater 12 1142.2 failed 18" below water 8 ~370.~ ~iled in ~ir 3~ ~bove water 4 1545.5 faile~ in air 13" above water 2 1545.~ failed 12" below water 11 1551,4 failed 41" below wate~
15(replacemen~)60g.9 f~iled ~1" below w~ter The da~a wa~ andlyzed usin~ Weibull ~t~tisti~
to yield the p~ra~eter~ a (~3.5% fail~re proba~ility~ a~
1300.4 hours ~nd ~ ~mea~ure of the spread) as 2.96 For the ~ir~t three failures, cable se~tion.~ ~lo~e to ~reakdown sites re~ealed a laL-~e number of bow-tie trees in the 2-S mil size ranye with no ven~ed tree~. T~le ~ver~ge brea~down ~tren~th of the vir~in cable~ with Example I a~ in.~ul~tion wa~ 1032 volts/mil e~ceeding the AEIC ~pecification. Ho~ever, after 1~0~ hour~ of accelerated test in water fille~ tanks, thi~ dropped to a value of 400 volt~/mil. On the contrary, th~ reference çdble retained 85~ of it~ ori~i~al ~alue of 1030 ~olt~/mil for it~ breakdown ~rength. ~he a~erage failure time of eable with Ex~mple I in~ulation Wd~ e~uivalent to that of ~ con~en~ional ur~modif ied XLPE c~ble . It should be noted that the addition of EVA(OH~ of g5 ~ hydroly~i~ did not re~ult in any improvement in the re~i~tance to ~oi~re induced ~e~ra~ation.

CA 0222049~ l997-ll-07 AEIC Oualification Tests With Example 1 The cable with Example I was also tested as per AEIC qualification test procedures. Several anomalies were observed.
(1) The cable insulation did not render itself transparent in hot oil at 120~C and thus it was not possible to examine cable interfaces. The insulation became even more opaque than at room temperature, which indicates structural and morphological incompatibility of terpolymer with polyethylene.
(2) With an applied voltage of 13.0 kV, the dissipation factors measured at ambient temperature and at 130~C were, respectively, 0. 14% and 2.64%, with the latter far exceeding the specification of 0.5% maximum.
(3) The AC breakdown strength after 120 days of accelerated water tree aging was 540 volts/mil with a very large number of non-vented water trees.
These results confirm that the terpolymer of higher degree of hydrolysis did not enhance the insulation resistance to moisture induced degradation as tested by the industry standards and hence not useful for actual service applications. At the concentrations used in Example 1, it is not structurally compatible with polyethylene.

Moulded Plaoue tests with ExamPle 1 Accelerated water tree growth studies were conducted using moulded and cross-linked plaques of Example I at ambient temperature and at 65~C. The size of water trees was measured after a fixed test time. HFDA
4202 was used as reference sample against which comparisons were made:

CA 0222049~ 1997-11-07 Sample Water tree size (micrometers) at 65~C at ambient temperature 2300 hrs test 1500 hrs test Example 1 330 500 These results clearly show the water tree retardancy inherent in the composition of Example 1, and demonstrate a superiority over the HFDA 4202 at 65~C test temperature.
Thus, the results obtained for the composition of Example 1 teach the following:
1. An improved cable with enhanced resistance to moisture induced degradation is not self-evident from a water tree retardant composition as judged from moulded plaque tests alone.
2. EVA(OH) terpolymer of very high degree of hydrolysis, such as the one used in Example 1 from 95% hydrolysis, is structurally incompatible with polyethylene and renders it opaque at temperatures required for cable defect examination. This ex~m'n~tion step, therefore, is thus not possible and hence a cable with such insulation cannot be certified for improved service reliability. This is reinforced further by the Examples described hereinbelow, wherein EVA(OH) of higher hydrolysis is unsuitable for cable applications at any concentration level;
3. EVA(OH) of higher hydrolysis when used at 5 % w/w level, also increases the dielectric loss of the XLPE
insulation to an unacceptable level.
Further modifications were therefore necessary to make the composition highly useful for practical service applications.

CA 0222049~ 1997-11-07 ImPLGv~..~..t in transParencY and dielectric loss In view of the importance and necessity of a transparent XLPE insulation, the compatibility of EVA(OH) with PE was examined by employing 5 mm plaques of Examples 2-5. The details of the embodiments of these compositions with the test data are presented hereinbelow.

Examples 2-5 Components Example 2 Example 3 Example 4 Example 5 ( % w/w) ( % w/w) ( % w/w) ( % w/w) Polyethylene 95.4 96.4 95.4 92.4 EVA(OH) 2 1 2 5 Irganox 1035 0.15 0.15 --- ---Irganox 802 0.35 0.35 --- ---Irganox 245 --- --- 0.2 0.2 Chimasorb 944 --- --- 0.3 0.3 Dicumyl peroxidel.9 1.9 1.9 1.9 Transparency hot oil test opaque opaque transparent opaque cloudy cloudy cloudy Dielectric loss (%) @ 2 kV/mm @ 130~C 1.45 0.56 0.54 1.40 @ 90~C 0.72 0.21 0.22 0.63 Examples 2 and 3 are similar to Example 1 except for the EVA(OH) concentration in the compositions. In Examples 4 and 5, an EVA(OH) terpolymer with 40%
hydrolysis, also supplied by Tosoh, Japan, was used;
Irganox 245 is a hindered phenolic ester and Chimasorb 944 is AO-4 a hindered amine, both supplied by Ciba Geigy.
Surprisingly, it was found that EVA(OH) of higher hydrolysis was not useful in any concentration range studied, notwithstanding the dielectric loss of the composition containing I % EVA(OH), Example 3, may come close to the specification requirements as shown above.
On the contrary, EVA(OH) of 40% hydrolysis used at 2% w/w concentration as in Example 4 rendered the XLPE optically transparent in hot oil test indicating physical -CA 0222049~ 1997-11-07 compatibility. In addition, to meet the electrical requirement it can only be used at or below 2% w/w concentrations. The latter is evident from the data on dielectric loss as given above.

SYnerqistic effect of EVA(OH) and h;n~red amine on water tree qrowth Three types of tests were carried out to elaborate on the surprising effect of synergism between EVA(OH) terpolymer and hindered amine. Moulded plaque tests for accelerated water tree growth were carried out with the following test conditions:

Test Conditions 1. 6kV, IkHz,0.1M NaCl, 1500 hrs at RT
2. 6kV, lkHz,0.1M NaCl, 2300 hrs at 65~C
3. 6kV, lkHz,0.1M NaCl, CuSO4.5H20 and (NH4)2S2O8, 1000 hrs at ambient temperature The details of the embodiments of the test compositions are given below. EVA(OH) of 95% hydrolysis was used in all:
Exa~ les 6-7 Example l Example 6 Example 7 Com~onents (% w/w) (% w/w) (% w/w) Polyethylene 92.6 92.6 92.l EVA(OH) 5 5 5 Irganox 1035 0.15 --- 0.15 Irganox 802 0.35 --- 0.35 Irganox 245 --- 0.2 ---Chimasorb 944 --- 0.3 0.5 Dicumyl peroxide l.9 l.9 l.9 Example 1 Example 6 Example 7 Water tree size (micrometers) Test condition 1 500 170 NA
Test condition 2 330 130 NA
Test condition 3 380 --- 220 CA 0222049~ 1997-11-07 , The results obtained above show that the addition of Chimasorb 944 resulted in smaller, water tree size for the same type and concentration of EVA(OH), proving the additional moisture degradative stability offered by the hindered amine. Although the mechanism for this synergistic effect is not known, and not being bound by theory, the observed smaller water tree size could be related to lower degree of oxidative degradation due to the combined action of moisture, ions, electric stress and temperature. This enhancement in the resistance to water tree growth through the use of an hindered amine in electrical insulation is most surprising.

Cable test~ usinq the insulation comPosition of the ~resent invention The final test composition was that of Example 4 because the addition of EVA(OH) with 40% hydrolysis at 2%
w/w level to polyethylene had rendered the insulation transparent and limited the dielectric loss within the specification requirements and because of the unexpected synergism between EVA(OH) and hindered amine. The compound was produced to the same cleanliness standards as of Example 1, which had not convincingly demonstrated any improvement in the resistance to moisture induced degradation in full size cable tests. A 15 kV rated cable with this compound as insulation and commercially available semiconductive shield compounds (AT Plastics Compounds AT 377 and AT389) was extruded onto a 1/0 AWG
(19 wires) compressed aluminium conductor. Concentric neutrals were then laid on these cables. These cables were used for AEIC qualification tests. A second 15 kV
cable with the same materials, but on #2 AWG (7 strands) aluminium conductor was extruded for life tests. A
reference cable with HFDA 4202 as insulation was also simultaneously extruded.

CA 0222049~ 1997-11-07 Accelerated Cable Life tests with Example 4 The test conditions were as follows.
(1) Test voltage 4 x Vg (35 kV; average stress of 200 volts/mil).
(2) Test frequency 60 Hz.
(3) Load cycling with 90~C conductor temperature in air and 8 hours on and 16 hours off load cycle.
10 (4) tank water temperature maintained at 50~C +3Ø
(5) twelve 16 ft cable samples wound into 1 loop (1 x 360) with an active length of 13 ft in each water tank.
(6) Cable strands and tanks filled with water of initial resistivity of ~200 Kohm.cm (7) Tests conducted until all samples fail.

In this series of tests the entire cable population insulated with HFDA 4202 failed before 270 20 days, with the initial failure occurring after 173 days. The failure data for this cable was as follows:
Sample No. Time to fail Comment (days) 9 173.5 failed 8" below water 3 178.5 failed 28" below water 192 failed 12" below water 12 196 failed 28" below water 4 200.8 failed 5" below water 7 213.3 failed 18" below water 6 218.8 failed 22" below water 8 221.6 failed 17" below water 226.3 failed at water-line 11 241 failed 26" below water 2 250.8 failed 27" below water 268.6 failed 27" below water The statistical analysis of the failures in 40 HFDA 4202 insulated cable yielded the values for the parameters a (63.596 failure probability) as 227.8 days CA 0222049~ 1997-11-07 and ~ (spread) as 8.01.
The failure data for the cable insulated with the composition of the present invention is given below:

Sample No. Time to Fail Comment (Days) 4 12.3 Failed 7" below water line*
1 291 Failed 10" below water line 9 348.7 Failed 13n below water line 429.6 Failed 10" below water line 458.8 Failed 29" below water line 11 478 Failed 30" below water line 12 483 Failed 17" below water line 6 486.2 Failed 1" below water line 7 497 Termination failure 8,2,3 ---- Removed after 254 days * diagnosed to be a premature failure and statistically outside the significance bounds The statistical analysis of the failure data yielded the values for the parameter ~=504.1 with ~=6.87, thus establishing the superiority of the cable performance with reference to the commercially available insulation system viz. HFDA 4202 insulated cable.

AEIC Oualification Tests with Example 4 The cable with Example 4 as insulation was also tested as per the procedures under this qualification test and the improvements in terms of enhanced resistance to moisture induced degradation were confirmed:

(1) The cable insulation rendered itself transparent enabling the physical and microscopic examination of the semiconductive interface. This reassures the claim made based on the plaque test for the compatibility of EVA(OH) of lower hydrolysis with polyethylene.~0 (2) The dielectric loss factors were measured with an applied voltage of 8.8 kV at ambient temperature, CA 0222049~ l997-ll-07 90~C and 130~C. Conductor temperatures were, respectively, 0.021%, 0.130% and 0.200% and below the 0.5% requirement. This reinforces the claims made hereinbefore through the plaque tests.
(3) There was an observed 64% retention on the AC
breakdown strength after accelerated water tree tests.

Although this disclosure has described and illustrated a preferred embodiment of the invention, it is to be understood that the invention is not restricted to both particular embodiments which are functional or mechanically equivalents of the specific embodiment and features that have been described and illustrated.

Claims (8)

1. An electrically insulating cross-linked polyethylene composition for use in high voltage electrical cables, said cross-linked polyethylene being obtained by cross-linking a composition consisting essentially of a low density, peroxide cross-linkable polyethylene, a terpolymer of ethylene, vinyl acetate and vinyl alcohol and at least 0. 15 % w/w of a sterically hindered amine stabiliser; said terpolymer being partially hydrolyzed ethylene vinyl acetate copolymer; and wherein said polyethylene and said terpolymer have substantially the same melt index, said composition being optically transparent after heating to 120°C in hot oil for 15 minutes.

2. An electrically insulating cross-linked polyethylene composition for use in high voltage electrical cables, said cross-linked polyethylene being obtained by cross-linking a composition consisting essentially of a low density, peroxide cross-linkable polyethylene, a terpolymer of ethylene, vinyl acetate and vinyl alcohol and at least 0. 15 % w/w of a sterically hindered amine stabiliser; said terpolymer being a partially hydrolyzed ethylene vinyl acetate copolymer with from about 40-45% hydrolysis and formed from an ethylene-vinyl acetate copolymer that has a 20-30% w/w vinyl acetate content; and wherein said polyethylene and said terpolymer have substantially the same melt index, said composition being optically transparent after heating to 120°C in hot oil for 15 minutes.

3. An electrically insulating cross-linked polyethylene composition for use in high voltage electrical cables, said cross-linked polyethylene being obtained by cross-linking a composition consisting essentially of 98 % w/w of a low density, cross-linkable polyethylene, 1-2 % w/w of a terpolymer of ethylene, vinyl acetate and vinyl alcohol and at least 0. 15 % w/w of a sterically hindered amine stabiliser; said terpolymer being obtained from about 40-45 % hydrolysis of an ethylene-vinyl acetate copolymer having 20-30 % w/w vinyl acetate content; and wherein said polyethylene and said terpolymer have substantially the same melt index, said composition being optically transparent after heating to 120°C in hot oil for 15 minutes.

4. A composition as claimed in any one of Claims 1-3 wherein said ethylene-vinyl acetate copolymer has a vinyl acetate content of about 25 % w/w.

5. A composition as claimed in any one of Claims 1-4 wherein said sterically hindered amine is in an amount selected from 0.15 - 0.3 % w/w.

6. A composition as claimed in any one of Claims 1-5 wherein said sterically hindered amine is poly-{[6-(1,1,3,3-tetramethylbutyl)-imino]-1,3,5-triazine-2,4-diyl][2-(2,2,6,6tetramethylpiperidyl)-amino]-hexamethylene-[4-(2,2,6,6-tetramethylpiperidyl)imino]}.

7. A composition as claimed in any one of Claims 1-6 wherein said sterically hindered amine stabilizer is a polymeric sterically hindered amine light stabilizer, said stabilizer and terpolymer being synergistically effective to enhance the resistance of the composition to moisture induced degradation.

8. A high voltage electrical cable comprising an electrical conductor and an electrically insulating cross-linked polyethylene composition as claimed in any one of Claims 1-7.