WO1996020974A1 - Process for stabilizing ethylene polymers during high temperature melt processing operations - Google Patents

Abstract

Ethylene polymers, such as linear low density polyethylene, with or without pendent polar functionality, are stabilized against changes in viscosity due to crosslinking during high temperature (e.g. in excess of 175 °C) melt processing operations by adding to the polymer, either prior to or during the operation, a viscosity-stabilizing amount of a transition metal, e.g. copper metal or a copper salt such as copper stearate, optionally in combination with one or both of a controlled metal deactivator, e.g. an oxamide or a hydrazine, and an antioxidant, e.g. a hindered phenol or an aryl amine. Certain transition metals other than copper are preferred melt viscosity stabilizers when aluminum compounds, such as hydrotalcite acid neutralizer, is present. Ethylene polymers are also stabilized by adding unhindered phenols, particularly when yellowing is a problem. Recycling of polyolefins can be improved by adding a viscosity-stabilizing amount of a transition metal, optionally in combination with a controlled metal deactivator and/or antioxidant, to the recycled polyolefin stream either before or during melt processing.

Description

PROCESS FOR STABILIZING ETHYLENE POLYMERS DURING

HIGH TEMPERATURE MELT PROCESSING OPERATIONS

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is a continuation-in-part of pending application USSN 08/329,734 filed October 26, 1994, which is a continuation-in-part of USSN 08/033,666 filed March 16, 1993, now abandoned, which is a continuation-in-part of USSN 07/850,731 , filed March 13, 1992, now abandoned, which is a divisional of application USSN 07/641,028, filed January 14, 1991, now USP 5,096,955.

BACKGROUND OF THE INVENTION

This invention relates to ethylene polymers. In one aspect, the invention relates to ethylene homopolymers while in another aspect, the invention relates to ethylene interpolymers. In yet another aspect, this invention relates to stabilizing ethylene polymers against changes in viscosity due to crosslinking under high temperature (i.e. in excess of 175 C) melt processing conditions by adding to the polymer a viscosity-stabilizing amount of a transition metal, optionally in combination with one or both of a control led-metal deactivator and an antioxidant.

Ethylene homopolymers and interpolymers are known classes of thermoplastic polymers, each having many members. They are prepared by homopolymerizing ethylene or interpolymerizing (e.g. copolymerizing) ethylene with one or more vinyl- or diene-based comonomers, e.g. α-olefins of 3 to about 20 carbon atoms, vinyl esters, vinyl acids, styrene-based monomers, monomers containing two or more sites of ethylenic unsaturation, etc., using known copolymerization reactions and conditions.

The viscosity of these homopolymers and interpolymers, including those with pendent polar functionality, i.e. those containing a pendent radical that exhibits an electrical charge, tends to change during high temperature melt process operations such as extrusion molding and the like. Such thermally-induced changes in viscosity have been attributed to the changes in molecular weight and/or linearity of the homopolymers or interpolymers caused by crosslinking.

A wide variety of "stabilizers" have been developed to reduce the changes (e.g. crosslinking) that can occur during melt processing or under conditions of use. Many of the stabilizers are organic compounds which are classified in the plastics industry as antioxidants. Many antioxidants tend to function as free radical scavengers, and they interact with free radicals that are formed during polymerization or in the presence of air or other oxidizing medium. Antioxidants are a known class of stabilizers which includes, for example, hindered phenols, triaryl phosphites, arylamines, hydroxylamines, and the like.

The well-known and well-publicized deleterious effects of copper and many other transition metals on the oxidative stability of hydrocarbons and hydrocarbon polymers is based on the catalytic effects of these metals in decomposing hydroperoxides. Thus, for example, Stivala, Kimura and Gabbay in 'Thermal Degradation and Oxidative Processes", Degradation and Stabilisation of

Polyolefins. pp. 63 et ff., Applied Science Publishers, London (1983), describe the catalytic decomposition of hydroperoxides by metallic catalysts and review literature on the catalytic effects of metals on polyolefin oxidation. Carbonyl absorbance in polypropylene films is reported to increase at a faster rate in the presence of metals than in their absence (Figure 12). This is due to the fact that hydroperoxide decomposition leads to carbonyl formation and therefore, catalyzed hydroperoxide decomposition leads to faster rates of carbonyl formation.

Although the decomposition of hydroperoxide leads to carbonyl formation, the formation of carbonyl groups is not directly related to viscosity changes, and thus the rate of carbonyl formation is not a true measure of viscosity change.

Viscosity changes in polyolefins are directly related to the crosslinking that results from combination of alkyl radicals or the addition of alkyl radicals to double bonds.

However, polyolefins may scission due to β-cleavage of alkyl or alkoxy radicals. The ratio of alkyl radical chemistry to alkoxy/peroxy radical chemistry depends on oxygen concentration. Thus, conditions that favor high rates of carbonyl formation do not necessarily favor high rates of crosslinking. In short, although

hydroperoxide decomposition leads to carbonyl formation, it does not necessarily lead to increased crosslinking (it may lead to increased scission).

Antioxidants have been added as stabilizers to polyethylenes and ethylene interpolymers with mixed success. Typically, such antioxidants have protected articles made from these materials against oxidative degradation at ambient conditions but have not been particularly effective at protecting the polymer against thermally-induced changes in viscosity during melt processing.

The text "Plastic Additives Handbook". 2d Edition, edited by R. Gachter and Muller and distributed in the United States by Macmillan Publishing Co., New York, NY (1985), describes antioxidants in Chapter 1 and the mechanisms by which such compounds are thought to work. It describes polymers that change

properties under melt process conditions due to chain scission (e.g.

polypropylene) and polymers that change properties due to crosslinking (e.g. low density polyethylene (LDPE)) and suggests that processing stabilizer systems commonly used in polypropylene (i.e. phosphites and a long-term heat stabilizer in overall concentrations up to 0.1 wt %) could be used as process stabilizers for linear low density polyethylene (LLDPE). In Chapter 2 the text confirms the literature in describing the thermo-oxidation of polyolefins as proceeding by a free radical chain mechanism in which hydroperoxides are key intermediates.

Hydroperoxides undergo thermally-induced (120 C and higher) homolytic decomposition to free radicals, which in turn initiate new oxidation chains which attack the polymer and cause degradation. This homolytic decomposition reaction is said to be catalyzed in a redox reaction by the presence of catalytic amounts of certain metal ions, particularly transition metal ions, such as iron, cobalt, manganese, copper and vanadium. The author then states that the presence of such metal ions in the auto-oxidation of a hydrocarbon increases the

decomposition rate of hydroperoxides and the oxidation rate to such an extent that even in the presence of antioxidants, the induction period of oxygen uptake is drastically shortened or completely eliminated. Even at rather high concentrations, hindered phenols or aryl amines reportedly do not retard the oxidation rate sufficiently. A more efficient inhibition is allegedly achieved by using metal deactivators (e.g. copper inhibitors).

A variety of metal deactivators are described in Chapter 1 in the Gachter et al. handbook and a method of testing is set forth on page 82. In the test, the polyolefin resin and stabilizer are homogenized (i.e. thoroughly blended) in a suitable lab scale kneader (e.g. Brabender plastograph), or by first milling the resin and then adding 1 wt % fine copper power or 0.1 wt % copper stearate, making a compression molded plaque, and then oven aging the plaque to determine polymer changes over time. Test results are commercially important because of the wide use of polyolefin insulation over copper conductors. In such applications the author states that it is mandatory to combine a metal deactivator with an antioxidant if the metal deactivator does not contain moieties with radical scavenging function. Information is presented in Table 1 on page 84 showing combinations of metal deactivators and phenolic type antioxidants used to protect polyethylene in contact with copper. The need for metal deactivators is

emphasized by the teaching in Additives for Plastics by J. Stepak and H. Daust,

Spring-Verlag New York Inc. (1983) at pages 182-183 that heavy metal ions (Co, Cu, Mn, Fe, Pb) which catalyze the hydroperoxide decomposition are present in polymers from contact with metallic parts of reactors and processing machines.

Attempts have been made to counteract the metal/metal ion catalyzed peroxide decomposition reaction in polyolefins by including a material in the polymer which reacts preferentially with the peroxide or its decomposition products. Such materials are known, among other names, as "sacrificial reducing agents". For example, Black in USP 4,122,033 allegedly stabilized organic materials against auto-oxidation by including at least 100 parts per million (ppm) of a transition metal containing compound and certain (1) aliphatic amines, (2) alkyl selenides, or (3) alkyl phosphines or phosphites.

Similarly, Chiquet in USP 4,931 ,488 included starch in a thermoplastic polymer (e.g. polyethylene) to make a thermoplastic composition which allegedly degrades under the action of heat, ultraviolet light, sunlight and/or composting conditions. Chiquet used iron and another transition metal compound (e.g. copper stearate) to catalyze the degradation of the starch.

In view of these representative teachings about the catalytic effect which transition metals and metal ions have on the thermo-oxidation of polyolefins (e.g. polyethylenes) and how this catalytic effect is controlled by adding materials which react preferentially with peroxides to "protect" the polymer, the discovery that small amounts of a transition metal or its salt can thermally stabilize certain ethylene polymers under melt processing conditions without the benefit of metal deactivators was a surprise indeed. This discovery is fully described in USP 5,096,955 which is herein incorporated in its entirety by reference.

While the discovery described in USP 5,096,955 has advanced the art of stabilizing certain ethylene polymers, areas of practice remain to be improved. For example, other classes of ethylene polymers also require stabilization against viscosity while under melt processing conditions. In addition, transition metals tend to reduce the long-term stability of the ethylene polymer due to their catalytic role in hydroperoxide decomposition. This effect is not prevalent during high temperature extrusion, but it is important at lower temperatures such as those found under oven-aging or long-term storage conditions. This effect can be tempered by the addition of phenolic antioxidants, but these tend to discolor the ethylene polymer over time.

Improvements are also needed in the melt processability of aluminum-containing polymeric compositions, because the inventors found that aluminum compounds and certain preferred melt viscosity stabilizing transition metals taught in USP 5,096,955 are antagonistic to each other.

Improvements are also needed in recycling polyolefins. Recycled

polyolefins have had problems, because they undergo more severe processing histories than virgin polyolefins leading to degradation, discoloration, etc., and because they are generally not free of contamination by labels, dirt, glue, and other polymers leading to melt processing problems such as gelling.

Gelling during melt processing must be prevented in both virgin and recycled polyolefins, because it leads to defects in the polymer end product. It is due to excessive crosslinking of some portion(s) of the polymer composition during melt processing and is associated with thermo-oxidation of the polymer or the presence of contaminants during melt processing as reported, for example, in Johnston, "Degradation and Stabilization of LLDPE During Melt Processing," pp. 60-78 of "Broadening the Horizons of Linear Low Technology" presented at the Regional Technical Conference of the Society of Plastics Engineers, Inc., Akron, Ohio, on October 1-2, 1985, which is hereby incorporated herein by reference. These and other problems are solved by the present invention as described below.

SUMMARY OF THE INVENTION I

In one aspect of this invention, ethylene polymers are stabilized against changes in viscosity due to crosslinking under high temperature melt processing conditions by adding to the polymer a viscosity-stabilizing amount of a transition metal other than nickel, optionally in combination with one or both of a controlled-metal deactivator and an antioxidant.

In one embodiment of this invention, ethylene interpolymers having pendent polar functionality or ethylene/diene interpolymers are stabilized against changes in viscosity due to crosslinking under high temperature melt processing conditions by adding to the polymers a viscosity-stabilizing amount of a transition metal other than nickel, optionally in combination with one or both of a controlled metal deactivator and an antioxidant. In another embodiment of this invention, ethylene homopolymers or ethylene interpolymers made by the interpolymerization of ethylene and at least one α-olefin and/or at least one diene are stabilized against changes in viscosity due to crosslinking under high temperature melt processing conditions by adding to the polymers a viscosity-stabilizing amount of a mixture or complex of a transition metal other than nickel and a controlled-metal deactivator, optionally in combination with an antioxidant. Another aspect of this invention is a process for recycling polyolefin materials by adding to a recycled polyolefin composition a viscosity-stabilizing amount of a transition metal other than nickel, optionally in combination with one or both of a controlled-metal deactivator and an antioxidant and melt processing the recycled polyolefin composition. This aspect includes melt viscosity stabilized compositions obtained by adding a viscosity-stabilizing amount of transition metal other than nickel, with or without a controlled metal deactivator and an antioxidant, to a polymer composition comprising recycled polyolefin, stabilizer concentrates for making the melt viscosity stabilized compositions, and recycled products made by melt processing the melt viscosity stabilized compositions.

Another aspect of this invention is a process for stabilizing aluminum compound-containing polyolefins against changes in viscosity due to crosslinking under melt processing conditions by adding to the polyolefin a viscosity-stabilizing amount of chromium, manganese, ruthenium, cobalt, rhodium, iridium, rhenium, and/or osmium, optionally in combination with one or both of a controlled-metal deactivator and an antioxidant and melt processing the polyolefin. This aspect includes melt viscosity stabilized compositions for use in this process, which optionally can include adding the aluminum compound together with the viscosity-stabilizing amount of transition metal selected from the foregoing, with or without a controlled metal deactivator and an antioxidant, to a polyolefin, stabilizer concentrates for making the aluminum compound-containing melt viscosity stabilized compositions, and polymer materials made by melt processing the melt viscosity stabilized compositions of this invention.

Another aspect of this invention is the gel-free, or substantially gel-free, polyolefin polymer, preferably comprising an ethylene polymer, obtainable by melt processing a polyolefin polymer containing a viscosity-stabilizing amount of a transition metal other than nickel and, optionally, a controlled metal deactivator, antioxidant and/or an aluminum compound.

Important to the understanding of this invention is that the deleterious, undesirable, pro-oxidant effects of metals are due to their catalytic decomposition of hydroperoxide. As shown by this invention, when this reaction is eliminated or substantially retarded, the metal(s) can then have a positive effect on polymer stability under conditions in which alkyl radical chemistry (e.g. crosslinking) is important. While not bound by theory, the metals used in this invention are believed to act as alkyl radical traps which prevent or retard alkyl radical chemistry. Moreover, in the case of relatively stable alkyl radicals, such as those found in polyethylene, the probability of a metal trapping the radical before it reacts with another alkyl radical or an unsaturated group is improved (relative to relatively unstable alkyl radicals such as those found in polypropylene, which readily undergo beta scission reactions), even with surprisingly minor amounts of metal. As used here, "stable" means propensity to undergo β-scission as opposed to stable in the traditional thermodynamic sense. Of course, these positive benefits are not available if the metals are reacting with hydroperoxides.

The resulting ethylene polymer compositions are new, melt-stabilized, crosslink-resistant, substantially gel-free polymer compositions having many uses and which are capable of being recycled through several iterations with little or no degradation, discoloration, etc.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows torque curves for maleic acid anhydride grafted high density polyethylene alone, with Irganox™ 1010 and with Irganox™ 1010 combined with copper stearate during melt processing at 250 C. Figure 2 shows torque curves at 250 C for maleic acid anhydride grafted high density polyethylene blends with ASPUN fiber grade linear low density polyethylene alone, with Irganox™ 1010, with copper acetate (abbreviated CuAc) and with a combination of copper acetate and Irganox™ 1010. Figure 3 shows torque curves at 250 C for ethylene acrylic acid copolymer resins alone, and with cobalt stearate, with the combination of copper stearate and distearylamine, with cobalt phthalocyanine and with a combination of cobalt stearate and Irganox™ 1010.

Figure 4 shows torque curves at 250 C for a 20/80 blend of ethylene acrylic acid copolymer resin and linear low density polyethylene alone and with Irganox™

1010 and with a combination of copper acetate and Irganox™ 1010.

Figure 5 shows torque curves at 250 C for linear low density polyethylene alone and with 1% Ox-PE wax, with copper stearate, with a combination of copper stearate and 1 % Ox-PE wax, and with a combination of copper stearate

(abbreviated CuSt2), 1% Ox-PE wax, and Irganox™ 1010.

Figure 6 shows torque curves at 250 C for linear low density polyethylene alone and with Irganox™ 1010, with copper stearate, and with a combination of copper stearate and Irganox™ 1010.

Figure 7 shows torque curves at 250 C for low molecular weight ethylene acetic acid wax alone and with 1 % Allied EAA 5120 and with a combination of 1%

Allied EAA 5120 with copper stearate.

Figure 8 shows torque curves at 250 C for a sodium ethylene acetic acid ionomer alone and with Irganox™ 1010, with a combination of Irganox™ 1010 and copper acetate, with cobalt stearate, and with cobalt phthalocyanine. Figure 9 shows torque curves at 250 C for ethylene carbon monoxide copolymer containing 500 ppm Irganox™ 1010 alone and with 20 ppm copper stearate, with 20 ppm copper stearate and an additional 500 ppm Irganox™ 1010, and with 50 ppm copper stearate. Figure 10 shows torque curves at 250 C for ethylene carbon monoxide copolymer containing 500 ppm Irganox™ 1010 alone and with copper/Irganox MD1024 complex, with Irganox MD1024, with copper sulfate pentahydrate, with copper acetate, with copper phthalocyanine, with 10 ppm cobalt phthalocyanine, and with 1000 ppm cobalt phthalocyanine. Figure 11 shows torque curves at 250 C for 12% vinyl acetate copolymer,

Elvax 3135X, alone and with Irganox™ 1010, with copper stearate, and with a combination of copper stearate and Irganox™ 1010.

Figure 12 shows torque curves at 250 C for Elvax 3135X alone and with two combinations of copper stearate and Irganox™ 1010, one trial at 20 ppm copper stearate and another trial at 50 ppm copper stearate.

Figure 13A shows torque curves at 250 C for additional combinations of Elvax 3135X with copper stearate and Irganox™ 1010 compared with Elvax 3135X alone.

Figure 13B shows the torque curve at 250 C for Elvax 3135X with a complex of copper sulfate pentahydrate and Irganox MD1024.

Figure 13C shows a torque curve at 250 C for Elvax 3135X with VERSENE AG™.

Figure 14 shows torque curves at 250 C for Elvax™ 3120 alone and with Irganox™ 1010, with 10 ppm copper stearate, with a combination of 10 ppm copper stearate and 1000 ppm Irganox™ 1010, and with a combination of 20 ppm copper stearate and 2000 ppm Irganox™ 1010.

Figure 15 shows torque curves at 250 C for the same combinations of stabilizer/ antioxidant as in Figure 14, except Elvax™ 3120 is replaced by Elvax™ 3175.

Figure 16 shows torque curves at 250 C for ethylene/ethylene acetate (18%) copolymer alone and with Irganox™ 1010, with a combination of copper acetate and Irganox™ 1010, with copper acetate and with a combination of copper acetate and Irganox™ 1010. Figures 17A and 17B show torque curves at 250 C for various combinations of antioxidant and transition metal stabilizer with linear low density polyethylene produced by constrained geometry catalysis (CGC LLDPE).

Figure ,17C shows torque curves at 250 C for Dowlex 2045 polyethylene resin combined with various amounts of molybdenum Ten-Cem™ (molybdenum neodecanoate from Mooney Chemicals).

Figure 18 shows torque curves at 250 C for Dowlex 2045 polyethylene resin alone and with para-t-butyl phenol (PTBP) and a combination of copper stearate and PTBP.

Figures 19A through 191 show torque curves at 250 C of Dowlex 2045 polyethylene resin containing melt viscosity stabilizers and antioxidants according to the present invention with and without the presence of various aluminum compounds.

Figures 20A through 20E show torque curves at 250 C for recycled polyethylene alone and in combination with antioxidant and melt viscosity stabilizers according to the present invention. These figures are described in more detail in the examples below.

DETAILED DESCRIPTION OF THE INVENTION

The ethylene polymers that can be melt stabilized by the practice of this invention are well known, and include both ethylene homopolymers and

interpolymers of ethylene and one or more other vinyl or diene-based monomers.

The ethylene homopolymers include both high and low density polyethylene (i.e. HDPE and LDPE). The ethylene interpolymers (i.e. copolymers) are those that contain ethylene and minor amounts of one or more vinyl- or diene-based monomers, polyenes, etc. (e.g., α-olefins, ethylene acrylic acid (EAA), ethylene acetate (EVA), 1 ,4-hexadiene, 1 ,4,7-octatriene, ethylidene norbornene, the naphthenics (e.g., cyclopentene, such as cyclopentadiene, cyclohexene, and cyclooctene, etc.) in interpolymerized form. Examples of ethylene interpolymers that can be used in the practice of this invention are: copolymers of ethylene and C₃-C₂₀ α-olefins (e.g. propylene, butene-1, hexene-1 , octene-1 , 4-methyl-1-pentene, etc.); copolymers of ethylene and vinyl-based monomers having polar functionality (e.g. EAA, EVA, ethylene methyl acrylate (EMA), ethylene carbon monoxide (ECO), etc.); various polyethylene polymers and copolymers grafted with an unsaturated organic compound containing at least one double bond and at least one functional acid or anhydride group (e.g. high density polyethylene grafted with maleic anhydride (HDPE-g-MAH)); various oxidized polyethylene polymers and copolymers (e.g. oxidized polyethylene wax); copolymers of ethylene interpolymerized with two or more C₃-C₂₀ α-olefins; and copolymers of ethylene interpolymerized with one or more dienes (e.g. ethylene/propylene/diene monomer (EPDM), ethylene/octene/diene monomer (EODM), ethylene-propylene rubber (EPR), and the like). For purposes of this invention, "copolymer" includes polymers made from two or more monomers, and "diene" includes aliphatic and cycloaliphatic monomers containing two or more sites of ethylenic unsaturation. The ethylene polymers are prepared by known polymerization processes, including: high pressure, free radical initiated polymerizations to make LDPE and HDPE resins; the "Phillips" process which uses a chromium catalyst to make HDPE resins; gas phase polymerizations and solution phase polymerizations in which ethylene is homopolymerized, or copolymerized with an alkene of 3 to 20 carbon atoms (e.g. butene-1 , hexene-1, octene-1. etc.) over a suitable transition metal catalyst to make linear low density polyethylene (LLDPE); the slurry process for making HDPE; and other known techniques. The polymerization techniques are broadly classified as solution, gas phase or slurry polymerization reactions. The ethylene polymers and their methods of preparation are described in many sources. See, for example, the Kirk-Othmer Encyclopedia of Chemical

Technology, Volume 16, pages 385-452; Modern Plastics Encyclopedia/89. pages 63-72, from McGraw-Hill, Inc.; Chemical Technology Review No. 70 entitled "Polyolefins Production Processes Latest Developments" by Marshall Sittig, Noyes Data Corporation (1976); the compilation of technical papers presented at the

Golden Jubilee Conference held June 8-10, 1983, by the Plastics and Rubber Institute under the title "polyethylenes 1933-83", and the like. The structure of the ethylene polymers can be branched or linear in molecular configuration, and the physical properties (e.g. melt index, density) can be varied by the reaction conditions, catalysts and olefin monomers used in the polymerization.

The ethylene polymers that can be stabilized by the process of this invention specifically include the elastic, substantially linear ethylene polymers comprising at least one α-olefin comonomer and made through the action of a constrained geometry catalyst. These polymers and their method of preparation are fully described in USP 5,272,236 and 5,278,272, each of which is incorporated herein in its entirety by reference.

For purposes of this invention, "ethylene interpolymers having pendent polar functionality" and like terms include homopolymers of ethylene that have pendent polar functionality, e.g. HDPE-g-MAH and the like. With respect to grafted ethylene polymers and copolymers, the unsaturated organic compound typically contains a double bond conjugated with the double bond of an acyl group, e.g. the acids and anhydrides of maleic, fumaric, acrylic, methacrylic, itaconic, crotonic, α-methyl crotonic, cinnamic and the like. The ethylene polymers of this invention do not include ethylene vinyl alcohol (EVOH). All the ethylene

interpolymers having pendent polar functionality that can be used in the practice of this invention are prepared by known copolymerization, grafting and/or oxidation processes. In addition to these ethylene polymers, recycled polyolefin materials may be used to make melt viscosity stabilized polyolefin compositions according to this invention. The inventors have surprisingly found that such melt viscosity stabilized compositions can be processed under conventional melt processing conditions without excessive and deleterious changes in polymer molecular weight or viscosity. The absence of such changes is unexpected, because recycled polyolefins undergo more severe processing histories than virgin polyolefins, particularly when they are post-consumer. Not only must post-consumer recycled polyolefins undergo melt processing the first time they are fabricated, they are exposed to deterioration caused by long term storage, exposure to a variety of organic and inorganic products during use, and exposure to harsh chemicals used for removing such products and other contamination such as labels, dirt, glue, etc. Sorting and cleaning is not 100% efficient, so that traces of such contaminants can persist and the recycle stream may contain other polymer types as well,

complicating the requirements for reprocessing. As mentioned above,

contamination is known to cause gelling during melt processing.

The inventors have found that these and other problems are solved by adding at least one of the aforementioned transition metals to a polymer

composition comprising recycled polyolefins, with or without a controlled metal deactivator and/or an antioxidant. In a preferred embodiment an antioxidant is present (e.g., a hindered or unhindered phenol), in another preferred embodiment a controlled metal deactivator is present, with or without a hindered or unhindered phenol functionality, and in yet another preferred embodiment the transition metal is molybdenum. The recycled polyolefin and/or stabilized composition therefrom may be blended with one or more virgin polymers such as polyethylene, polypropylene, EVA copolymer, etc., in any amount. The amount may be from about 1 wt% or less to about 95 wt% or more, but is more typically in the range from about 5 wt% to about 30 wt%.

As used herein, the term "virgin polymer" refers to polymer which has not yet been fabricated for the first time (i.e., it has not yet been recycled).

The metals used in the present invention are transition metals of Groups 5 through 11 , rows 4 through 6 of the CAS Version of the Periodic Chart, Handbook of Chemistry and Physics. 69th Edition, edited by Robert C. Weast, CRC Press (1989-1990). For purposes of this invention, "transition metal," "transition metals," and like terms include the metals in both their metallic state, e.g. as finely-divided particulate solids, and in their ionic states, e.g. as metal salts. In either state, the metals are used in a form such that they are dispersible in the ethylene polymers.

The metals are preferably used as dispersible metal carboxylate salts, and most preferably as metal carboxylate salts of fatty acids. Vanadium, chromium, manganese, cobalt, copper, molybdenum, ruthenium, palladium, platinum, rhodium and iridium, and the dispersible carboxylate salts of such metals are preferred, with metallic copper, cobalt and molybdenum, and their carboxylate salts, most preferred. Copper and molybdenum and their respective salts are particularily preferred. In addition to high temperature melt processing stability, molybdenum salts also provide desirable long-term heat stability to the

polyethylenes in which they are dispersed. With respect to ethylene polymers that do not contain polar functionality, e.g. HDPE, LLDPE, LDPE, ultra low density polyethylene (ULDPE), etc., iron and nickel (and their carboxylate salts) are operable in the present invention, but they are less effective in providing viscosity stabilization than comparable amounts of manganese, molybdenum, cobalt and copper, elements which are near or adjacent to iron and/or nickel in the Periodic Chart. Similarly, silver (and its carboxylate salts) do not perform as well as comparable amounts of copper, a metal which is adjacent to silver in the Periodic Chart. Suitable metals include, for example, the finely divided metals and the formate, acetate, acetylacetonate, octanoate, neodecanoate, and stearate salts of chromium, manganese, cobalt, copper, molybdenum and the various mixtures of one or more of these materials.

With respect to certain ethylene interpolymers with pendent polar functionality, some of the metals perform their stabilizing function better than others. For example, copper salts are only marginally effective with EAA copolymers, but cobalt stearate is very effective with this particular interpolymer.

Similarly, copper carboxylates are only marginally effective with ECO

interpolymers, but cobalt carboxylates and complexes of copper metal and an antioxidant are very effective.

According to this invention, the melt processing temperature of the polyethylene is maintained at a temperature in excess of 175 C, preferably in excess of 200 C, and more preferably in excess of 225 C. At these temperatures, thermal decomposition of hydroperoxide is so fast that the presence of a catalyst has little, if any, influence on the overall decomposition rate of hydroperoxide. However, under these conditions, the maximum hydroperoxide concentration in the melt is very low because other mechanisms (e.g. bimolecular decomposition, catalytic effects of oxidation products, etc.) cause the hydroperoxide to

decompose as fast or faster than it is formed. As such, even low concentrations of metal will still be enough relative to hydroperoxide concentration to enable the metal to act as an alkyl radical trap.

Any material that retards the reaction of a transition metal with

hydroperoxide, but that does not practically reduce the high temperature melt viscosity stabilizing action of the transition metal in an ethylene polymer can be used as the controlled-metal deactivator. Preferably, these compounds contain at least two carbonyl groups and at least two nitrogen atoms, e.g. derivatives of oxamides and hydrazines (e.g., oxamide, oxamido, hydrazine and hydrazide compounds). More preferably, these include compounds having one or more

or groups.

Even more preferably these include compounds of the formula

and

where R and R' are independently a hydrocarbyl, inertly-substituted hydrocarbyl radical preferably of 1 to about 20 carbon atoms, but not -N=CH-C₆H₅, the hydrocarbyl optionally having one or more additional and/or groups.

By "inertly substituted" is meant that the substituents are essentially nonreactive with the ethylene polymer under high temperature melt processing conditions other than in its capacity as an antioxidant. For those compositions in which optimal color, e.g. minimal discoloration, is important, R and R' do not contain aryl amine or hindered phenol functionality, but may contain one or more unhindered phenol functionalities derived from the unhindered phenol antioxidants described below. In those compositions in which maximum high temperature melt viscosity stabilization is more important than optimal color, one or both of R and R' preferably contains aryl amine or hindered or unhindered phenol functionality (i.e., functionality derived from the antioxidants described below), especially 2,6-di-t-butylphenol functionality. Examples include 2,2'-oxamido bis[ethyl-3-(3,5-di-t- butyl-4-hydroxyphenyl)propionate] and N,N'-bis[3-(3,5-di-t-butyl-4-hydroxyphenyl)propionyl] hydrazide; 1 ,2-diacetylhydrazine; 1-acetyl-2-phenylacetylhydrazine; oxalic bis(benzylidenehydrazide); 1 ,2-dibenzoylhydrazine; N,N-bis-(2-hydroxyethyl)oxamide; 3-aminophthalhydrazide; N,N'- dihexadecyloxamide; 1 ,2-dicarbethoxyhydrazine; diethyl-1,2-dimethyl-1,2- hydrazine-dicarboxylate; and 1,2-diethylhydrazine dihydrochloride.

Another preferred class of controlled metal deactivators are those of the formula R₁R₂NR₃NR₄R₅ where R₃ is (-CH₂-)n in which n is preferably 0-2 and in which R₁, R₂, R₄ and R₅ are -CH₂C(=O)OH. An example of this preferred class of controlled metal deactivator is ethylenediaminetetraacetic acid (EDTA).

Any effective combination of transition metal and controlled metal deactivator can be used to stabilize a polyethylene resin during melt processing. Preferably the controlled metal deactivator contains nitrogen atoms, and the metal and deactivator are present in a mole ratio of at least about 2:1, more preferably of at least about 4:1, based on the ratio of nitrogen atoms in the deactivator to metal. Nonlimiting examples of controlled metal deactivator/metal systems include copper stearate, cobalt stearate, ruthenium acetylacetonate, or copper sulfate solution blended with any of the controlled metal deactivators described above, and the copper diammonium complexes of EDTA in aqueous solution (manufactured and sold by The Dow Chemical Company under the trademark VERSENE).

In certain embodiments of this invention, the transition metal is used in combination with an antioxidant, particularly a hindered phenol or aryl amine antioxidant. These antioxidants are well known in the art, and the preferred hindered phenol and hindered amine antioxidants are more fully described in USP 3,979,180 which is herein incorporated in its entirety by reference. Representative antioxidants include octadecyl 3-(3,5-di-t-butyl-4-hydroxy-phenyl)-propionate; tetrakis-(2,4-di-t-butyl-phenyl)4,4'-biphenylphosphonite; di(stearyl)pentaerythritol diphosphite (+l% triisopropanolamine). Irganox™ 1010 (tetrakis[methylene(3,5-di- t-butyl-4-hydroxyhydrocinnamate)] methane) manufactured by Ciba-Geigy is a preferred antioxidant.

In addition to these antioxidants, the inventors have found that when non- hindered phenols are added to polyolefin compositions together with the transition metals in accordance with this invention, the polyolefin compositions have melt stabilization characteristics at least as good as that obtained with hindered phenols and transition metals while reducing or preventing the discoloration caused by the highly colored oxidation products formed by oxidation of the hindered phenol antioxidants. While not being bound to a particular theory, the ability of the nonhindered phenols to achieve stabilization against oxidation while resisting discoloration may be due to the oxidation resulting in coupling at the phenoxy functionality rather than formation of quinone compounds having intense color. Therefore, nearly any nonhindered phenol may be used as the antioxidant or antioxidant functionality of this invention. These include phenol, mono-substituted phenols, and polysubstituted phenols not having more than one bulky group in the ortho position. Bulky groups include bulky hydrocarbyl groups such as the t-butyl group and hydrocarbyl having more than four carbon atoms. A specific example of an unhindered phenol is para-t-butyl phenol. The transition metals are added to the ethylene polymers in small but viscosity-stabilizing amounts prior to or during melt processing operations.

Typically, they are added in amounts of at least about 0.5 parts per million (ppm), which is about 0.00005 wt % based on the weight of the metal relative to the weight of the ethylene polymer, and amounts up to about 0.1 wt % or more can be used. By "viscosity-stabilizing" is meant an amount sufficient to reduce the absolute value of the slope of a torque curve (see Figure 1). The goal is to approach a flat line with a slope of zero, but any reduction in the absolute value of the slope relative to the ethylene polymer alone is an unexpected improvement and the amount of metal added is a viscosity-stabilizing amount.

In those embodiments in which a controlled metal deactivator is used, it is generally used in an amount of about 5 to about 20,000 ppm, preferably about 50 to about 5000 ppm, which is about 0.0005 to about 2, preferably about 0.005 to about 0.5 wt % based on the weight of the deactivator relative to the weight of the ethylene polymer. The most preferred amount of deactivator will vary with the nature of the deactivator, the ethylene polymer, the transition metal and its physical state, the melt processing conditions, and the like.

In those embodiments in which an antioxidant is used, it is generally used in an amount of about 100 to about 5000 ppm, preferably about 200 to about 1000 ppm, which is about 0.001 to about 0.5 wt %% preferably about 0.002 to about 0.1 wt %, based on the weight of the antioxidant relative to the weight of the ethylene polymer. The most preferred amount of antioxidant will vary with the nature of the antioxidant, the ethylene polymer, the metal and its physical state, the nature of the controlled metal deactivator, the melt processing conditions, and similar variables. When, for example, an unhindered phenol is used as the antioxidant the amount thereof is preferably in the range from about 0.01 to about 3.0 weight percent, more preferably in the range from about 0.05 to about 0.3 weight percent range.

The transition metal can be added to the ethylene polymer in any convenient manner which results in the metal being substantially dispersed throughout the resin. The sequence of adding the antioxidant and the metal deactivator/metal compound system will vary with the polymer and processing conditions. The controlled metal deactivator/metal compound system can be made in situ or blended with one another prior to addition to the polymer melt.

Significant amounts of aluminum is often present in polymeric compositions as metal and as aluminum-containing compounds. Sources of aluminum include aluminum-containing acid neutralizers such as hydrotalcite (e.g., DHT-4A from Kyowa Co.); aluminum-containing fire retardants (e.g., aluminum trihydrate), aluminum-containing catalyst residues and aluminum-contamination from processing equipment, etc. The range of what may be present as aluminum in this aspect of this invention can include inorganic aluminum salts, organic aluminum salts, and finely dispersed aluminum as metal. It may be present in any amount, but is typically present in the range from about 10 ppm to about 10 wt.% based on the weight of aluminum. The inventors have found that there is an antagonistic effect between aluminum and the viscosity stabilizing effect of the otherwise preferred transition metals in polymeric compositions, reducing the effectiveness of those transition metal stabilizers.

The inventors have further found, however, that this problem is overcome by using transition metal stabilizers having transition metals selected from the group consisting of chromium, manganese, ruthenium, cobalt, rhodium, iridium, rhenium, and osmium, preferably manganese, cobalt and iridium, and most preferably cobalt. The transition metal stabilizer is preferably a transition metal compound, such as one comprising an inorganic anion, a carboxylate (e.g., a stearate), or other organic molecule such as acetylacetonate. These transition metals and transition metal compounds are preferably present in the range from about 0.1 ppm to about 1 wt%, preferably from about 0.5 ppm to about 0.1 wt %, based on the weight of the transition metal. Preferred polymeric compositions in this aspect of the present invention are those comprising polyolefin, more preferably an ethylene polymer, and a preferred aluminum compound is hydrotalcite.

A particularly surprising result is that not only does the use of cobalt as the transition metal stabilizer avoid antagonism with aluminum-containing compounds such as hydrotalcite, its effect as a melt viscosity stabilizer is enhanced by the presence of aluminum. This synergism is unexpected from the prior art.

Hydrotalcite is known as a metal deactivator so that from the prior art one would have expected less, not more, transition metal stabilization.

One method of adding these materials to the polymer is by blending them in a polymer masterbatch under conditions where the components may be intimately mixed, e.g. in the mixing section of an extruder. Other methods include dissolving or dispersing the components in a solvent and spraying the resulting mixture onto pellets or powders of the polymer prior to melt processing, or dry-blending the components with pellets or powders of the polymer prior to melt processing. Other methods will be apparent to those skilled in the art.

In a preferred embodiment, a stabilizer concentrate is prepared which comprises a polymer, sometimes referred to herein as a "stabilizer base polymer", and the metal, with or without a controlled metal deactivator and/or antioxidant, the latter dispersed throughout the former (i.e. the stabilizer base polymer). This stabilizer concentrate (generally molten) is then added to the molten ethylene or other polyolefin polymer as it passes through an extruder or other processing equipment. For compatibility reasons, the stabilizer base polymer is preferably the same type of polymer as the molten polymer passing through the processing equipment, and may be the same as the molten polymer. The stabilizer

concentrate is a convenient form of storage and handling for the metal, with or without controlled metal deactivator and/or antioxidant, and it can be easily metered into the extruder in reasonably precise quantities.

The following examples are illustrative of certain specific embodiments of this invention. Unless indicated to the contrary, all parts and percentages are by weight.

SPECIFIC EMBODIMENTS Description of Materials:

The various ethylene interpolymers and additives used in these examples are described below. Except where indicated, all ethylene interpolymer and additive characterization data were supplied by the manufacturers or distributors.

Ethylene Interpolymers:

Maleic Anhydride-graft-High Density Polyethylene (MAH-g-HDPE)

1.2% MAH content MAH-g-HDPE produced by reaction extrusion on a twin screw extruder.

Ethylene Acrylic Acid Copolymer (EAA)

[-CH₂CH(CO₂H-]_n

Ethylene acrylic acid copolymer with a melt index (g/10 min, Ml) of 1.5 and

9.7% acrylic acid content manufactured by The Dow Chemical Company (Dow). It nominally contained 200 ppm Irganox 1010 hindered phenolic antioxidant. EAA™ 5120

Ethylene acrylic acid copolymer having an acid number of 120

manufactured by Allied Chemical Corp (Allied).

Oxidized Polyethylene Wax

Low molecular weight (approximately 3000) with an acid number of 41, and manufactured by Allied.

Ionomer

[-CH₂CH(CO₂Na)-]_n

Sodium ionomer (1.3 Ml) prepared from an EAA copolymer. This ionomer was obtained after 33% neutralization with NaOH of an EAA copolymer (20 Ml,

9.7% acrylic acid) containing no stabilizer. The ionomer was a Dow product.

Ethylene Carbon Monoxide Copolymer (ECO)

[-CH₂CHCO)-]_n

An ECO copolymer having 10.2% CO content, a Ml of 15, and containing a nominal 500 ppm of Irganox 1010 antioxidant.

Ethylene Vinyl Acetate Copolymer ( EVA)

[-CH₂CH(O₂CCH₃)-]_n

Elvax™ 3135x was a nominal 12% vinyl acetate copolymer having a 0.4 Ml.

Elvax™ 3175 was a nominal 28% vinyl acetate (VA), 3.0 Ml resin. Elvax™ 3120 was a nominal 7.5% VA, 1.2 Ml resin. All Elvax™ copolymers were products of the E.I. DuPont de Nemours, Co.

Ethylene Ethyl Acrylate Copolymer (EEA)

[-CH₂CH(CO₂C₂H₅)-]_n

20 Ml, 0.930 g/cc density ethylene ethyl acrylate copolymer with an 18% ethyl acrylate (EA) content. It was void of inhibitors, it had a 240 F softening point and it was obtained from Aldrich Chemical Co., Inc. (Aldrich).

LLDPE Copolymer Produced by Constrained Geometry

Catalysis (CGC LLDPE)

The CGC LLDPE used in these examples are ethylene-1-octene

copolymers analyzed as having a 0.96 Ml (l₂), 7.28 melt flow ratio (l₁₀/l₂), and 0.9005 density (g/cm³).

Linear Low Density Polyethylene (LLDPE)

An additive-free version of Dowlex 2045 (a trademark and product of Dow) with a nominal density of 0.92 g/cc and a measured Ml of 0.92. This ethylene/1-octene interpolymer had approximately 0.31 vinyl groups per 1000 carbons, and it was prone to significant amounts of crosslinking during melt processing.

Dowlex 2047A (a trademark and product of Dow) has a nominal density of 0.92 and Ml of 2.3 and contains 500 ppm DHT-4A hydrotalcite acid neutralizer, 500 ppm Irganox™ 1076 and 800 ppm PEPQ. Metal Salts:

Controlled Metal Deactivators/Antioxidants.

Irganox™ 1010 is tetrakis[methylene(3,5-di-t-butyl-4-hydroxyhydrocinnamate)] methane, an antioxidant from Ciba-Geigy.

Irganox™ 1076 is n - octadecyl-beta-(3,5 - dibutyl - 4 - hydroxy phenyl)-propionate, an antioxidant from Ciba Geigy.

Cu-lrganox™ MD 1024 complex (Irganox MD1024 is N,N" -bis[3-(3,5-di-t-butyl-4-hydroxy-phenyl)propionyl]hydrazide, from Ciba-Geigy). The complex was prepared by placing 0.0120 g copper sulfate pentahydrate (CSP) and 1.2010 g Irganox™ MD1024 in a 150 ml beaker containing Isopar E hydrocarbon solvent (obtained from Exxon). The mixture was heated and stirred until the temperature reached 85 C. Water (15 ml) was then added with vigorous stirring. The copper compound changed color from blue to green, and then was observed to transfer from the aqueous phase to the hydrocarbon phase. The mixture was placed in a glass dish and the solvent/water were evaporated.

Copper sulfate pentahydrate (CSP) - Irganox™ MD1024 complex prepared by placing 6.7737 g CSP in 300 ml of deionized water, and heating the mixture until a clear, light blue solution was obtained. Irganox™ MD 1024 (30.00 g) was heated with stirring to 50 C in 1500 ml acetone. The aqueous CSP was stirred into the acetone/MD 1024 solution, and immediately an olive-green precipitate was formed. The mixture was cooled to 45 C, and then it was poured into a glass dish from which the solvent/water was evaporated at room temperature. The resulting product had a melting point of 197 C, and it was a light green, fine powder. VERSENE AG (a trademark of Dow Chemical Co., referring to a copper diammonium complex of EDTA in aqueous solution, containing 7.5% copper).

PTBP is para-t-butyl phenol, a nonhindered phenol available from

Schenectady Chemicals.

PEPQ refers to Sandostab PEPQ™, an antioxidant which was available from Sandoz and is now available from Clariant Corp., tetrakis-(2,4-di-t-butyl-phenyl-Y,Y'-biphenyl phosphonite).

Equipment and Procedures:

A Rheocord System 40™ torque rheometer from HaakeBuchler Instruments was used to test the melt processing stability of the various formulations. The mixer was a Rheomix 600™ model equipped with roller style blades. The mixer was modified, with -glass-filled TEFLON™ or stainless steel bushings replacing the original bronze brushings. The torque rheometer was operated isothermally at temperatures between 150-275 C depending on the particular polymer. The ram was kept up, allowing exposure to air. Mixing speed was 10 rpm. A 60 second adiabatic initial mixing period at 60 rpm was used at the start of each experiment. The additive was incorporated accurately by weighing it into a capsule molded from the base interpolymer, then adding the additive package and polymer into the mixer via a loading chute and ram and then, unless stated otherwise below, processing for 1000, 1500 or 3000 seconds. Sample size was typically 40 g but in interpolymers with higher melt densities, larger amounts were used in order to fill approximately the same mixer volume each time. Mixing conditions were not selected based on typical fabrication processing conditions for the given interpolymer, but were based on conditions believed most likely to reflect a reasonable rate of thermo-oxidative degradation and opportunity to observe the possible stabilizing action of metal compounds.

Results:

MAH-g-HDPE and Its Blends with LLDPE

Figure 1 shows that MAH-g-HDPE by itself crosslinked so severely that by approximately 2500 seconds it began to grind to a powder and climb out of the mixer bowl, resulting in a rapid torque decrease due to the reduced polymer volume in the mixer and the cessation of viscoelastic flow as a mechanism of deformation. 1000 ppm Irganox 1010 reduced the rate of torque increase

(crosslinking). A combination of 10 ppm copper stearate and 1000 ppm Irganox 1010 provided an even greater reduction in crosslinking rate.

One application of MAH-g-HDPE is in a blend with ASPUN fiber grade LLDPE (a trademark and ethylene/1-octene copolymer of The Dow Chemical Company with an Ml of 18 and a density of 0.930 g/cc) to produce fibers. Typically, 10% of the MAH-G-HDPE is used with 90% LLDPE. Figure 2 shows the stabilization of this blend with copper acetate. 10 ppm copper acetate had a torque reducing effect when used alone, and a combination of copper

acetate/lrganox 1010 produced a lower torque curve than Irganox 1010 alone. Table I shows the results of melt index analyses on samples removed at the end of each 1500 second experiment. These data clearly show the viscosity stabilizing effect of copper acetate formulations.

The sample containing both Irganox 1010 and copper acetate showed the least reduction in Ml during 1500 seconds processing. (An estimate of the initial Ml for the blend is 11.9 g/10 min., calculated using the logarithmic addition rule:

0.9[log(18.0)]+O.1[log(O.29)]=log(blend)).

Ethylene Acrylic Acid Copolymers

Irganox 1010 alone and high concentrations of copper stearate alone had only marginal effect on torque during 250 C processing of EM in the torque rheometer. A combination of 10 ppm copper stearate and 1000 ppm Irganox 1010 also had little effect. So did aqueous copper sulfate solution.

On the assumption that any significant effect of copper may have been tempered through coordination by the acid groups of EAA, two "protected" forms of copper were tested. The first was a complex of copper and Irganox MD1024, a hydrazine functional metal deactivator. The second was copper phthalocyanine. Neither compound had a significant effect at the 10 ppm level. A mixture of copper stearate and distearylamine (Figure 3) had only a slight torque reducing effect. However, Figure 3 also shows that EAA torque was significantly reduced by cobalt compounds. This was unexpected, since copper compounds are generally more effective than cobalt in ethylene polymers. 1000 ppm of cobalt

phthalocyanine substantially eliminated torque changes for approximately 500 seconds. One common use of EAA is in multilayer coextruded structures with polyethylene. Since these may be ground and recycled in fabrication processes, the effect of copper acetate and Irganox 1010 on the stability of a 20/80 blend of EAA 1410 and LLDPE was studied. Figure 4 shows that torque was substantially reduced in the presence of copper acetate/lrganox 1010. Oxidized PE Wax/EAA Wax

Figure 5 shows that copper stearate produced a lower torque in LLDPE when used with 1 % Ox-PE wax than when used by itself. A slightly lower torque was obtained in the combination of 1 % Ox-PE wax with copper stearate/lrganox 1010 than with copper stearate/lrganox 1010 alone (Figure 6), but the torque trended upwards after 1000 seconds. Copper stearate was also active when used with Allied EAA 5120 wax (Figure 7).

These results, together with the EAA/LLDPE blend results shown in Figure 4, suggest that polyethylene oxidation products (acids, ketones, aldehydes) and

EAA acid groups do not inhibit the activity of copper stearate unless they are present at very high concentrations (e.g., pure EAA). Alternatively, these groups may not deactivate copper at all. EAA and polyethylene may simply degrade via different mechanisms such that the degradation of EAA is inhibited by Co but not by Cu, and the degradation by polyethylene is inhibited by both.

Sodium EAA Ionomer

Figure 8 shows that a sodium EAA ionomer had slightly improved stability when 1000 ppm Irganox 1010 was added. However, the additional incorporation of 10 ppm copper acetate had no further stabilizing effect. Cobalt stearate or cobalt phthalocyanine produced higher rates of torque increase in sodium EAA ionomer, in contrast to their effects in EAA.

Ethylene Carbon Monoxide Copolymer

Figures 9 and 10 show that the torque stability of an ethylene carbon monoxide copolymer was not substantially affected by up to 50 ppm copper stearate, copper sulfate, copper acetate or copper phthalocyanine. However, 10 ppm cobalt phthalocyanine caused a significant reduction in the rate of torque increase. The "protected form" of copper (Cu/lrganox MD1024) complex) also significantly reduced torques and had a larger effect than Irganox MD1024 alone. When 1000 ppm cobalt phthalocyanine was used, no torque increase was observed. Copolymers of Ethylene and Acetate/Acrylate

Functional-Monomers

Figure 11 shows the effect of copper stearate and copper stearate/lrganox 1010 blends on the torque stability of Elvax 3135X, a 12% vinyl acetate

copolymer. This polymer was so prone to crosslinking that not only was a substantial torque increase obtained, but it also powdered due to shear grinding of the polymer network (similarly to MAH-g-HDPE in Figure 1 ). While 10 ppm copper stearate did not completely inhibit torque increases, it provided a substantial induction time prior to the onset of torque increases. This induction time was greater than 1000 seconds in the case of the copper stearate/lrganox 1010 blend.

Doubling the copper stearate concentration did not have a significant effect (Figure 12) on torque, and even 50 ppm of copper stearate in the blend had only a slight effect on the induction time. Increasing the Irganox 1010 concentration to 2000 ppm had a much larger effect (Figure 13A). These results suggest that Irganox 1010 has the major effect on stability in EVA, but clearly, copper stearate has a cooperative effect.

A comparison of the torque curve in Figure 13B with those in Figure 12 and 13A shows that even better melt processing stabilization is Obtained when a complex of copper sulfate pentahydrate and Irganox MD 1024 is used. Figure 13C demonstrates the efficacy of using VERSENE AG™ as the stabilizer in this

EVA polymer.

Elvax™ 3120 was a lower acetate content resin (7.5% VA). Copper stearate caused more torque reduction in this resin (Figure 14) that in the 12% VA resin. In a 28% VA resin (Elvax™ 3175), the effect of copper stearate may have been less (Figure 15). However, this possible correlation of copper effectiveness and VA content must be considered speculative since the resins varied widely in molecular weight and possibly other characteristics such as VA distribution or additive content. Nonetheless, these data clearly show that copper

stearate/lrganox 1010 together provide improved torque stability over either component alone in a range of EVA copolymers. An additive-free ethylene/ethyl acrylate (18%) copolymer (Figure 16) had significantly better torque stability when copper acetate and Irganox 1010 were used together than when either component was used alone.

Together, these results suggest that the acetate functionality does not substantially inhibit the melt stabilizing effect of copper carboxylates, and that ethylene/acrylate copolymers may be expected to be more stable when processed in the presence of metal stabilizers.

CGC LLDPE

Figure 17A shows that a copper/lrganox MD1024 complex produces a favorable torque response relative to the base resin (the torque curve of which is reported in Figure 17B), and relative to the base resin in combination with Irganox

MD1024 alone. A small torque increase was obtained during the first 500 seconds, but subsequently the torque was stable. Additional copper

concentrations, or the addition of a hindered phenolic phosphonite had little effect on the torque. Figure 17B also reports that arylamine (Ageright White™ from R.T. Vanderbilt, sym. dibetanaphthyl-p-phenylenediamine) are effective in reducing the torque curves of these polymers.

The CSP/lrganox MD1024 complex I was prepared by heating a mixture of 0.0120 g CSP and 1.210 g Irganox MD 1024 in Isopar E with stirring to about 85 C. Since the CSP did not enter the solution, 15 ml of water was added with vigorous stirring. The aqueous phase was at first blue, and then turned green. Eventually the color transferred to the hydrocarbon phase. The solution was transferred to a glass dish from which the solvent/water was evaporated at room temperature.

The CSP/lrganox MD1024 complex II was prepared by placing 6.6738 g Irganox MD1024 into a 1000 ml beaker with 500 ml of Isopar E hydrocarbon solvent. CSP (0.31479 g) was added in solid form. The mixture was heated with stirring to about 85 C, and then 100 ml of water was added with vigorous stirring. The solution changed color from aqua blue to olive green, and then dispersed into the hydrocarbon phase. The solution was then transferred to a glass dish, and the solvent/water evaporated at room temperature.

Hydroperoxide Concentration in Ethylene Polymers

In the absence of metals and at low temperatures in polyolefin oxidation (i.e. below about 175 C), hydroperoxide concentration increases significantly with time, both in static and dynamic oxidation environments, i.e. static oxidation of polymer films and melt processing, respectively. The hydroperoxide concentration eventually reaches a maximum, and it then begins to decrease due to bimolecular hydroperoxide decomposition mechanisms as described by W.L. Hawkins in "The Thermal Oxidation of Polyolefins-Mechanisms of Degradation and Stabilization", Degradation and Stablization of Polymers. John Wiley & Sons, New York, (1975), and/or the catalytic effects of oxidation products such as carboxylic acids on hydroperoxide decomposition. Chakraborty and Scott report in the European Polymer Journal. Vol. 13, pgs. 731-737, Pergamon Press (1977) in an article entitled, "The Effects of Thermal Processing on the Thermal Oxidative and Photooxidative Stability of Low Density Polyethylene", that by increasing the

temperature above 150 C, the maximum hydroperoxide concentration decreased.

Similar trends are reported elsewhere (e.g. Iring et al, Polvmer Bulletin, 16, 159- 165 (1986)). Accordingly, at low temperatures (i.e. less than about 175 C) and high hydroperoxide concentrations, bimolecular reactions predominantly occur as opposed to at high temperatures (i.e. more than about 175 C) or low

hydroperoxide concentrations where homolytic cleavage of hydroperoxides predominantly occurs. In addition to this change in the predominant

decomposition mechanism believed to occur between about 150 and 175 C, there is also a strong increase in overall hydroperoxide decomposition rate as the temperature is increased above about 175 C. This means that as the temperature increases, hydroperoxides increase reaching a high concentration at a relatively low temperature, e.g. 150 C, but rapidly homolytically cleave (possibly assisted by acidic oxidation products) at higher temperatures, i.e. above 175 C). This means that the rate of hydroperoxide decomposition becomes greater than the rate of hydroperoxide formation and that under these conditions, the maximum

hydroperoxide concentration achieved in the polymer melt is likely to be very low. This behavior is further demonstrated by the data reported in Tables 11 A, IIB, and IIC.

The data of Tables HA, IIB and IIC show that the hydroperoxide concentration continues to increase with time at 150 C, but at 175 C a maximum absorption occurs after 3000 seconds of processing. At 200 C, the absorption is simply too weak to measure. The significance of this temperature dependence of hydroperoxide decomposition is that if the processing temperature is sufficiently high and the oxygen concentration sufficiently low, the maximum hydroperoxide concentration will be low, if not extremely low. Under these conditions, even low concentrations of transition metal will be sufficient to act as an alkyl radical trap. Furthermore, because the hydroperoxide decomposition rate by thermal decomposition is already high, a large increase in the overall oxidation rate (e.g. carbonyl formation rate) will not occur. This contrasts with a sharp increase in oxidation rate observed in the presence of metal at lower temperatures (such as that reported by Gugumus in "Advances in UV Stabilization of Polyethylene", Organic Coatings and Applied Polymer Science Proceedings. Vol. 46, (ACS 1981) at 150 C) in which hydroperoxides are relatively stable in the absence of metals, and thus are strongly effected by catalysis. This is further substantiated by the data reported in Tables IIIA and B.

The data of Tables IIIA and B show that high (e.g. 1000 ppm)

concentrations of high redox potential transition metals like cobalt cause increased rates of carbonyl formation even at 250 C, but low (e.g. 100 or less ppm) concentrations of these metals, or either low or high concentrations of copper, do not sharply increase carbonyl formation rates at this high temperature. Moreover, even at a high concentration of these metals, a sharp difference in temperature sensitivity is observed. In Table IIIA, for example, 1000 ppm of cobalt and copper stearate cause 2.6 and 1.4 carbonyls/1000 C, respectively, after 3000 seconds at 250 C as compared to 1.15 for the control resin or in other words, an increase of 116 percent and 16 percent, respectively, in the rate of carbonyl formation. At lower concentrations (e.g. 1 - 53 ppm) but at the same temperature (250 C), the increase in carbonyls is much less (see Table IIIB).

By comparison, a similar LLDPE processed at 150 C with 1000 and 975 ppm cobalt and copper stearate, respectively, produce carbonyl levels of 0.28 and 0.09 carbonyl/1000 C respectively, during 1500 seconds of processing, in contrast to 0.02 carbonyls/1000 C for the base resin during 3000 seconds of processing, or increases of 1400 percent and 350 percent, respectively, despite the metal-catalyzed samples being processed only half as long.

Lono-Term Heat Stability

One aspect of using transition metals as melt processing stabilizers in polyethylene is their tendency to destabilize the polymer during long-term use at relatively low temperatures (e.g. less than 175 C). By catalyzing hydroperoxide decomposition, they accelerate the rate of oxidation and can potentially reduce the shelf life of the polymer by a significant amount. This can be particularly troublesome in applications in which the polyethylene is in contact with a metal, such as the copper conductor in wire and cable applications. To offset this tendency, the transition metal is often formulated with a metal deactivator to maintain the properties of the polyethylene during long term use.

Surprisingly, relatively small amounts of the transition metal stabilizers of this invention do not catastrophically reduce the long-term heat stability of ethylene polymers. For example, 500 ppm copper stearate reduced the long-term heat stability of DOWLEX LLDPE by only 20 percent in a 90 C oven-aging test.

Likewise, 500 ppm cobalt stearate reduced the heat stability of the same polymer by only 40 percent. Reduction of heat stability was measured by the reduction in time to embrittlement. While these reductions are significant, they are not catastrophic, and clearly suggest that even lesser amounts (e.g. 1-50 ppm) which are effective for melt stabilizing ethylene polymers at temperatures in excess of about 175 C should not catastrophically reduce the heat stability of the resultant polymer.

Moreover, trace amounts of copper stearate in combination with

antioxidants impart outstanding long-term stability to ethylene polymers. For example, when 10 ppm copper stearate was used in combination with 1000 ppm Irganox 1010 antioxidant, the LLDPE samples lasted the length of the test (614 days) without embrittlement, as compared to 44-81 days for existing LLDPE based stabilization systems (200 ppm Irganox 1010). This clearly suggests that the melt stabilizer systems of this invention, particularly those of copper stearate and hindered phenol, impart good resistence to physical deterioration. However, some discoloration of the polymer can occur.

Given the good long term stability of trace amounts of transition metal with antioxidants, even better long term stability can be expected for transition metal/controlled metal deactivator systems such as copper/lrganox MD1024, especially when combined with antioxidants. Of particular note, ethylene polymers stabilized with a molybdenum compound, e.g. molybdenum neodecanoate such as Mo TEN-CEM from Mooney Chemical, even at 500 ppm and even in the absence of antioxidant, provides stability against embrittlement over the duration of a 614 day 90 C oven-aging test. In addition, the inventors have dound thta molybdenum by itself in an additive free base resin (Dowlex™ 2045 Polyethylene base resin provides outstanding melt viscosity stability as shown in Figure 17C. Controlled Metal Deactivators

As described above, controlled metal deactivators are those compounds that coordinate with the transition metal such that it (the metal) is inactive towards hydroperoxides at low to moderate temperatures (e.g. less than about 175 C) yet it remains active as a viscosity stabilizer at high temperatures (e.g. in excess of 175

C). These systems, i.e. controlled metal deactivator plus transition metal, impart processing stability without having a strong adverse effect on the long term heat stability of the polymer (as described above).

The efficacy of any particular compound to act as a controlled metal deactivator in a given polymer requires the juxtaposing of two measurements. The first measurement is the extent to which the deactivator passivates the transition metal toward the catalysis of hydroperoxide decomposition, and the other measurement is the extent to which the transition metal (while complexed with the deactivator) is available to viscosity stabilize the polymer at high temperature. The juxtaposing of these measurements is demonstrated in Table IV.

Additive-free LLDPE-hydroperoxide was prepared from a 1 M.I. LLDPE resin in a Banbury mixer for 75 minutes at 145-151 C. This product (i.e. Dowlex™ 2045 Polyethylene) had 0.19 carbonyls/1000 C and 0.39 hydroperoxide/1000 C (according to the 3550 cm^-1 peak method described previously). This was the sample material used to generate the data reported in Table IV. By measuring the change in hydroperoxide concentration during processing under nitrogen (e.g. conditions where further oxidation and hydroperoxide formation cannot occur), the catalytic effect of transition metal compounds on polyethylene hydroperoxide decomposition was determined. The transition metal/metal deactivator compounds used in Table IV were prepared by dissolving the respective transition metal and deactivator in an appropriate solvent, and then either precipitating the compound or evaporating the solvent. In those samples in which the transition metal/metal deactivator systems were added as a premix, the systems were prepared as a solution followed by drying or precipitation of the transition metal/controlled metal deactivator complex. These complexes were either added to the polymer neat or in the form of a polymer concentrate. In the dry blends of Table IV, the respective components of the systems were weighed into the same capsule and then added to the melt to form the system in situ.

Systems containing a controlled metal deactivator in combination with a transition metal should perform no worse than the transition metal at comparable transition metal concentrations relative to hydroperoxide decomposition.

Preferably it should perform the same as the base resin, and more preferably it should perform better than the base resin, (i.e. have a lower hydroperoxide decomposition rate). At the same time, it should impart melt viscosity stability greater than the stability of the base resin itself (i.e. it should flatten the torque curve) and preferably, it should perform as good as a transition metal, and more preferably better than a transition metal, at comparable transition metal concentrations.

While the carboxylate and similar metal salts are useful aids in the dispersion of the metal throughout the polymer melt, these organic radicals are effective at controlling the metal relative to its reactivity towards hydroperoxides.

As such, metal carboxylates and the like are not controlled metal deactivators.

Table IV identifies various controlled metal deactivators within the scope of this invention. The base resin (Control A which is an average of five separate runs and did not contain a transition metal) reported a 20 percent decrease in the hydroperoxide concentration over the period of 400 to 3000 seconds at 150 C. When copper stearate (Control B) was added after 400 seconds at 150 C, the hydroperoxide concentration decreased 63 percent between the period of 400 seconds to 3000 seconds. The controlled metal deactivators of this invention for this particular polymer under these particular conditions should cause a reduction in hydroperoxide concentration no greater than 63%, and preferably 20% or less.

Sample A reports that Irganox™ 1010 is a poor controlled metal deactivator (it does not pacify the copper stearate with respect to catalyzing hydroperoxide decomposition) and consistent with this failure to pacify, the copper is available to stabilize the viscosity at high temperatures (as reported by the 340 m-g peak torque value). In contrast, EDTA demonstrates some effect as a pacifier for copper stearate, but it complexes with the copper so tightly that the copper is only marginally effective as a viscosity stabilizer (a peak torque value of 1050 m-g versus 1200 m-g for the base resin). However, Versene AG™, a 46 percent diammonium copper complex of EDTA in aqueous solution (54 percent water, and containing 7.5 percent copper) is a very effective controlled metal deactivator (Sample N).

Samples C and D identify a good controlled metal deactivator. While the peak torque was not measured for Sample E, it was measured for a similar sample at 1000 ppm and this sample demonstrated both good stabilization and good control. Since Naugard™ XL-1 did not completely deactivate the copper as a viscosity stabilizer at 1000 ppm, then it clearly would not do so at 593 ppm.

Samples G, H and I report good to marginal deactivators that demonstrate good viscosity control, while Samples K, M and N report controlled metal deactivators that combine excellent metal passivity relative to hydroperoxide decomposition catalysts and excellent metal availability with respect to viscosity stabilization.

Preferably the metal/controlled metal deactivator system retains its resistance to reaction with hydroperoxide after heat treatment (such as extrusion).

Samples E and G demonstrate this behavior.

Copper and Cobalt

Copper and cobalt behave differently towards hydroperoxide and ethylene polymers. Copper causes scission, cobalt causes crosslinking (provided that some vinylic unsaturation is present). Other types of transition metals can be classified as "copper-like" (e.g. ruthenium, chromium) or "cobalt-like" (e.g.

manganese, rhodium). Of interest is that although both copper and cobalt cause hydroperoxide decomposition and result in increases in carbonyl formation, only one (cobalt) causes crosslinking and thus only one can be linked to viscosity increases during melt processing. This clearly shows that crosslinking and scission in the presence of transition metals cannot be directly related to carbonyl formation rates as suggested by Gugumus, supra. At 150 C, copper causes scission, not

crosslinking, when it reacts with hydroperoxide in an ethylene polymer. This behavior is clearly evidenced by the data reported in Table V.

Unhindered Phenol The same test procedure was used to test the melt viscosity stabilizing effect and reduction of discoloration effect of substituting unhindered phenols for hindered phenols as the antioxidant, except that the samples were removed after 300, 600, and 1000 seconds for coiorimetric analysis and torque was measured during the first 1000 seconds of melt processing to obtain data for melt viscosity stabilization over a processing time period corresponding to the coiorimetric data.

This data was generated using Dow's Dowlex™ 2045 Polyethylene LLDPE resin. Figure 18 shows the changes in torque due to crosslinking of LLDPE during mell processing and Table VI below shows the Yellowness Index for a sample containing PTBP as the antioxidant relative to the yellowness index for the same amount of Irganox 1076, a commercial hindered phenol antioxidant.

Stabilizer Concentrates

Stabilizer concentrates (masterbatches) were prepared in DOWLEX 2047A (an ethylene-octene copolymer containing 500 ppm DHT-4A hydrotalcite acid neutralizer, 500 ppm Irganox 1076 antioxidant and 800 ppm PEPQ). The concentrates were prepared by melt blending in a Haake torque rheometer mixer with Rheomix 3000 mixing bowl, roller style blades, and glass-filled TEFLON bushings. Resin was added gradually to the mixer at 20 r.p.m. and mixed until it was melted, then additive was added and the ram was lowered and rotor speed increased to 60 r.p.m. for 60 seconds. The speed was then reduced to 20 r.p.m. and the resin processed for 19 additional minutes. Melt temperature was maintained at approximately 150ºC. Concentrates were removed from the mixer, cooled, then granulated using a Colortronic M 102-L granulator. The following two concentrates were prepared: Stabilizer Concentrate A

0.0706 g copper stearate and 3.000 g PTBP were weighed out, then DOWLEX 2047A was added to make a total of 200.0 g of sample. This sample was mixed in the torque rheometer mixer as described above. Following melt blending, the molten polymer was removed from the mixer, cooled, and granulate to pellet form.

Stabilizer Concentrate B

0.0706 g copper stearate and 3.000 g Irganox 1076 were blended with DOWLEX 2047A as described for Stabilizer Concentrate A.

' Samples were tested for color changes during extrusion as follows:

Sample Q

180.10 g of Stabilizer Concentrate A was tumble blended with Dowlex™ 2045 LLDPE pellets from Dow Chemical Co. to prepare a total of 3.5 lbs. of extrudable mixture. The stabilizer mixture was placed in the hopper of a 1 inch diameter, 20:1 L/D laboratory scale MPM single screw extruder running at 156 r.p.m. and having a temperature profile of 350, 482 and 500ºF for the two extrude zones and die, respectively. The polymer composition was extruded through a nozzle die, then the extrudate was quenched in a water bath, passed through a compressed air strand drier, and pulled through a chopper for granulation. The granulated pellets were compression molded (150ºC) into 125 mil plaques for color determinations. Color was measured using a Hunter ColorQuest

Spectrocolorimeter (sphere). The Yellowness Index was 3.8. Sample R

180.10 g of the Stabilizer Concentrate B was tumble blended with Dowlex™ 2045 LLDPE from Dow Chemical Co. to prepare a total of 3.5 lbs. of extrudable mixture. The subject mixture was tested as in Sample Q. The Yellowness Index after extrusion was 8.9.

These results show that the combination of 40 ppm copper stearate and 1700 PTBP is highly effective for melt stabilization (i.e., no change in torque during processing as shown in Fig. 18) and yet the data in Table VI and described above for Sample Q and Sample R show that the PTBP formulation was nevertheless less discolored after melt processing or extrusion than the Irganox 1076 hindered phenol antioxidant formulation. Aluminum-Containing Polyolefin

A torque rheometer was used to evaluate the change in torque (e.g., viscosity) occurring in an ethylene-octene copolymer during melt processing at 250ºC, both in the presence and absence of hydrotalcite, aluminum hydroxide, aluminum acetylacetonate, and/or transition metal salts using the same equipment and procedures used above to test polyolefin compositions containing controlled metal deactivator and/or antioxidant. Irganox 1010 hindered phenolic antioxidant was optionally added as a co-stabilizer.

Torque versus time data were collected and plotted using a computer graphics program. Those data plots are shown in Figs. 19 A to 19H.

Figure 19A shows the torque increase (viscosity increase) due to thermooxidative crosslinking in additive-free LLDPE in the absence of any stabilizer. It also shows the torque reducing effects of copper stearate and how DHT-4A hydrotalcite has an antagonistic effect such that the torque is increased as compared to the copper-stabilized formulation free of DHT-4A. Figure 19B shows the excellent stabilization obtained with a combination of 10 ppm copper stearate and 1000 ppm Irganox 1010 antioxidant. as previously disclosed. It also shows how 500 ppm DHT-4A reduces the effectiveness of the stabilizer system.

Figure 19C shows that other aluminum containing compounds have similar effects, though not all. A hydrated alumina from AluChem did not have an antagonistic effect. Hydrated alumina (e.g., aluminum hydroxide) from Aldrich did. Figure 19D shows that aluminum acetylacetonate also had an antagonistic effect on copper stearate.

Figures 19E and 19F show that the antagonistic effect is not limited to copper, but also applies to other transition metal salts such as zinc stearate or molydenum neodecanoate compositions.

Figure 19G shows that ruthenium acetylacetonate is only slightly reduced in its activity when DHT-4A is present. Figure 19H shows that manganic acetylacetonate is not affected at all by

DHT-4A and Figure 191 shows that cobalt stearate had improved performance when used in the presence of DHT-4A.

Table VII below summarizes the effects observed with compounds of several different metals.

These data in Table VII show that certain transition metals which are generally less effective than copper as melt viscosity stabilizers are more effectiv than copper in the presence of hydrotalcite. This result cannot be predicted from the ionic radii of the metal ions as can be seen from Table VII.

Recycled Polyolefin

Two different recycle streams were used. Both recycled polyethylene samples were obtained from Midwest Plastic Materials, Inc., in Edgarton, Wisconsin. Midwest reprocesses HDPE bottles collected from throughout the upper Midwestern United States, with emphasis on Milwaukee and Chicago. Bot samples were in "flake" form, consisting of polyethylene bottles which had been separated, washed without detergent, and chipped into flakes. The chips were irregularly shaped, but a typical chip was approximately 5 mm wide and 10 mm long with an average weight of 0.1 g.

• Recycle Stream #1 ("Natural Flake"): This consisted mostly of clear flakes, with only an occasional pigmented one. These natural bottle flakes were mostly from HDPE milk bottles, believed to consist mostly of Phillips HDPE resin. The natural flakes had a composite average melt index of approximately 0.7 g/10 minutes, and a 0.960 g/cc density.

• Recycle Stream #2 ("Pigmented Flake"): This was predominantly natural flakes but with many pigmented flakes of a wide variety of colors also present. Based on a 7.68 g random sample, 27% of the flakes were pigmented and 73% were clear. The dominant colors were white, blue and red or orange. A smaller number of black, yellow, green, and off-white wer present. The pigmented flakes were derived from a wide assortment of household bottles. This recycle stream had a melt index of approximately 0.5 g/10 min. and 0.955 g/cc density. The melt processing stability of recycled polyethylene formulations was tested using the equipment and procedure used to test polyolefin compositions containing controlled metal deactivator and/or antioxidant. Additives were incorporated by weighing them into a small film envelope made from compression molded flakes of the relevant recycle stream, then adding this envelope right after the base resin flakes were loaded into the mixer (e.g., approximately 20-30 seconds into the run). Samples totaling 40.00 g in weight were processed for either 1000 or 3000 seconds. Torque versus time data were collected and plotted by a computer graphics program.

Samples collected at the end of the 1000 second experiments were analyzed by High Temperature Gel Permeation Chromatography.

Natural Flakes

Figure 20A shows the torque curves for the natural flakes processed at 250 C for 1000 seconds. The torque increase with time is due to thermo-oxidative crosslinking. Addition of 1600 ppm Irganox 1076 reduced the rate of torque increase significantly, but did not completely eliminate it. The additional incorporation of 10 ppm copper acetate resulted in an approximately 30 m-g lower torque after 1000 seconds. Copper stearate was also tested. Although the copper metal content was significantly lower in the polymer composition containing copper stearate than in the polymer composition containing copper acetate, the torque curves were similar. Copper stearate is preferred because of its improved dispersibility.

Because the differences in torque were so subtle, gel permeation

chromatography (GPC) testing was done to confirm the reduction in crosslinking suggested by the torque curves. In addition, 3000 second experiments were conducted for better discrimination of performance between the various additive systems.

GPC data are summarized in Table VIII. The molecular weight of the starting material is not known; no GPC data were obtained on the unprocessed flakes because they were not homogeneous. Thus, comparisons must be made to the other processed samples. The GPC data show that the copper containing formulations reduced crosslinking (lower Mw) without significant evidence of accelerated scission (lower Mn). This confirms the crosslinking inhibition mechanism of these stabilizers, and that the torque-lowering effect of copper is n simply due to increased rates of competitive scission.

Figure 20B shows the torque curves for 3000 second experiments. While 3000 seconds is severe compared to the conditions experienced by the bulk of polyethylene passing through typical extrusion processes, small amounts of polyethylene in stagnant zones of extruders (where gels are typically formed) might experience thermo-oxidative conditions this severe. The 3000 second experiments more clearly differentiated the performance of Irganox 1076 alone a compared to its combinations with copper carboxylates. There was not a significant difference in performance of copper stearate as compared to copper acetate.

Pigmented Flakes

Figure 20C shows 1000 second torque curves for the pigmented flakes. The rate of torque increase was somewhat lower for this resin than in the natural flakes (Figure 20A). The effects of copper acetate and copper stearate were similar in pigmented flakes as in natural flakes.

As with the natural flake, the subtle differences in torque shown in Figure 20C were confirmed by GPC (Table VIII). Lower weight average molecular weights without concurrent reductions in number average molecular weights were obtained using the copper-containing formulations. Figure 20D shows the results of 3000 second experiments. Copper stearate performed slightly better than copper acetate over the course of 3000 seconds, but otherwise the results were not much different than those for natural flakes.

Controlled Metal Deactivator

Certain metal deactivators such as Irganox MD 1024 may be used to increase the long-term stability of polyolefins by inhibiting the catalytic decomposition of hydroperoxides caused by the catalyzing effects of certain metals such as copper. Adding metal deactivators to polyolefin compositions containing a melt viscosity stabilizer that causes such catalyzing effects, such as copper carboxylate, may thus increase long term stability of the polymer product. Figure 20E shows the effect of 1000 ppm Irganox MD1024 alone and as a complex with copper. The latter formulation was based on 47.3 ppm copper sulfate pentahydrate, or approximately 12 ppm copper. Because of its hindered phenolic functionality, Irganox MD1024 alone provided stability not unlike Irganox 1076 (see Figure 20D). However, the complex with copper was much higher in effectiveness and, due to the relatively high copper content or protection of the copper by the controlled metal deactivator from antagonistic materials postulated to be in the pigmented flakes, the torque curve was flat. This effect was surprising given the known metal-deactivating effects of Irganox™ MD1024. The copper was not deactivated with respect to its ability to stabilize polyolefins at elevated melt processing temperatures because Irganox™ MD1024 is a controlled metal deactivator.

Although the invention has been described in considerable detail through the preceding specific embodiments, it is to be understood that these

embodiments are for purposes of illustration only. Many variations and

modifications can be made by one skilled in the art without departing from the spirit and scope of the invention.

Claims

WHAT IS CLAIMED IS:

1. A process for stabilizing an ethylene polymer against changes in viscosity due to crosslinking while under melt processing conditions which include a temperature in excess of about 175 C, the process comprising adding to the polymer a viscosity-stabilizing amount of a transition metal other than nickel and, optionally, a controlled metal deactivator and/or an anti-oxidant.

2. The process of Claim 1 in which the transition metal is at least one metal selected from the group consisting of vanadium, chromium, manganese, cobalt, copper, molybdenum, ruthenium, palladium, platinum, rhodium and iridium.

3. The process of Claim 1 in which the antioxidant is selected from the group consisting of hindered phenols, aryl amines, phosphonites and phosphites.

4. The process of Claim 1 in which the controlled metal deactivator is selected from the group consisting of derivatives of oxamide, hydrazines, and EDTA.

5. The process of Claim 4 in which the controlled metal deactivator has hindered phenolic antioxidant functionality.

6. The process of Claim 1 in which the ethylene polymer is an ethylene interpolymer having pendent polar functionality.

7. A process for stabilizing an ethylene polymer against changes in viscosity due to crosslinking while under melt processing conditions which include a temperature in excess of about 175ºC, the process comprising adding to the polymer a viscosity-stabilizing amount of a transition metal other than nickel and an unhindered phenol and, optionally, a controlled metal deactivator.

8. A process for stabilizing an aluminum-containing ethylene polymer against changes in viscosity due to crosslinking while under melt processing conditions which include a temperature in excess of about 175ºC, the process comprising adding to the polymer a viscosity-stabilizing amount of at least one transition metal selected from the group consisting of chromium, manganese, ruthenium, cobalt, rhodium, iridium, rhenium, and osmium and, optionally, a controlled metal deactivator and/or an antioxidant.

9. A process for stabilizing a polyolefin polymer comprising a recycled ethylene polymer against changes in viscosity due to crosslinking while under melt processing conditions which include a temperature in excess of about 175ºC, the process comprising adding to the polymer a viscosity-stabilizing amount of a transition metal other than nickel and, optionally, a controlled metal deactivator and/or an antioxidant.

10. A process for stabilizing an ethylene polymer against changes in viscosity due to crosslinking while under melt processing conditions which include a temperature in excess of about 175 C and against long term heat degradation, the process comprising adding to the polymer a viscosity-stabilizing amount of molybdenum.

11. A crosslink-resistant, substantially polymeric composition which is viscosity stable at temperatures in excess of about 175 C, the composition comprising:

(A) An ethylene polymer,

(B) a viscosity-stabilizing amount of a transition metal other than nickel, and, optionally,

(C) a controlled metal deactivator and/or antioxidant.

12. A crosslink-resistant, substantially polymeric composition which is viscosity stable at temperatures in excess of about 175 C, the composition comprising:

(A) An ethylene polymer,

(B) a viscosity-stabilizing amount of at least one transition metal selected from the group consisting of chromium, manganese, ruthenium, cobalt, rhodium, iridium, rhenium, and osmium, and, optionally,

(C) an aluminum compound,

(D) a controlled metal deactivator and/or antioxidant.

13. A stabilizer concentrate useful for introducing a viscosity stabilizing amount of a transition metal other than nickel into a base ethylene polymer such that the base polymer is stabilized against changes in viscosity due to crosslinking while under melt processing conditions which include a temperature in excess of about 175º C, the concentrate comprising the transition metal and, optionally, a controlled metal deactivator and/or antioxidant, substantially uniformly dispersed throughout a concentrate ethylene polymer, the transition metal present in the concentrate polymer in an amount sufficient to stabilize the base polymer after the concentrate is blended with the base polymer.

14. A stabilizer concentrate useful for introducing a viscosity stabilizing amount of at least one transition metal selected from the group consisting of chromium, manganese, ruthenium, cobalt, rhodium, iridium, rhenium, and osmium into a base polymer such that the base polymer containing an aluminum

compound introduced into the base polymer via the base polymer and/or the concentrate is stabilized against changes in viscosity due to crosslinking while under melt processing conditions which include a temperature in excess of about 175º C, the concentrate comprising the transition metal and, optionally, a controlled metal deactivator, antioxidant, and/or an aluminum compound

substantially uniformly dispersed throughout a concentrate polymer, the transition metal present in the concentrate polymer in an amount sufficient to stabilize the base polymer after the concentrate is blended with the base polymer.

15. A gel-free predominantly ethylene polymer prepared by melt processing an ethylene polymer at a temperature in excess of about 175" C, the ethylene polymer containing a viscosity-stabilizing amount of a transition metal other than nickel and, optionally, a controlled metal deactivator and/or antioxidant.

16. A gel-free predominantly ethylene polymer prepared by melt processing an ethylene polymer at a temperature in excess of about 175ºC, the ethylene polymer containing an aluminum compound and a viscosity-stabilizing amount of at least one transition metal selected from the group consisting of chromium, manganese, ruthenium, cobalt, rhodium, iridium, rhenium, and osmium and; optionally, a controlled metal deactivator and/or antioxidant.

17. A recycled polyolefin polymer prepared by melt processing a polyolefin polymer comprising a recycled ethylene polymer at a temperature in excess of about 175º C, the ethylene polymer containing a viscosity-stabilizing amount of a transition metal other than nickel and, optionally, a controlled metal deactivator and/or antioxidant.