GB2392162A - Styrene-butadiene rubber cement - Google Patents

Abstract

A cement of living styrene-butadiene rubber which is comprised of an organic solvent and polymer chains that are derived from 1,3-butadiene and styrene, wherein the polymer chains are terminated with lithium end groups, wherein the polymer chains have a vinyl content of less than 10 percent, wherein less than 5 percent of the total quantity of repeat units derived from styrene in the polymer chains are in blocks containing five or more styrene repeat units, and wherein the molar amount of polar modifier in the cement of the living styrene-butadiene rubber is at a level of less than 20 percent of the number of moles of lithium end groups on the polymer chains of the living styrene-butadiene rubber.

Description

23921 62

- 1 SYNTHESIS OF STYRENE-BUTADIENE RUBBER

. Background of the Invention

It is desirable for a tire to exhibit good 5 traction characteristics on wet and dry pavements, and for the tire to provide good treadwear and low rolling resistance. In order to reduce the rolling resistance of a tire, rubbers having a high rebound can be utilized in making the tires' tread. Tires made with l0 such rubbers undergo less energy loss during rolling.

The traditional problem associated with this approach is that the tire's wet traction and wet skid resistance characteristics are compromised. This is because good rolling resistance which favors low 15 energy loss and good traction characteristics which favor high energy loss are viscoelastically inconsistent properties.

In order to balance these two viscoelastically inconsistent properties, mixtures of various types of 20 synthetic and natural rubber are normally utilized in tire treads. For instance, various mixtures of styrenebutadiene rubber (SBR), polybutadiene rubber, and natural rubber are commonly used in automobile tire treads formulations. Styrene-butadiene rubber is 25 included in tire tread formulations primarily to improve the traction characteristics of the tire without greatly compromising treadwear or rolling resistance. The versatility of solution SBR (SSBR) synthesis 30 relative to the synthesis of emulsion (ESBR), including control of molecular weight, microstructure, microstructure, and functionalization, is well established (see Hirao, A.; Hayashi, M. Acta. Polym.

1999, 50, 219-231, and references cited therein).

35 Performance advantages arising from this versatility have led to an acceleration of the replacement of

r - 2 - emulsion SBR in the tire industry, and an expansion in the market for random, low vinyl SBR for use in tire compounds (see Autcher, J.F.; Schellenberg, T.; Naoko, T. "Styrene-Butadiene Elastomers (SBR)," Chemical 5 Economics Handbook, SRI-International, November, 1997). These developments have stimulated interest in developing technology for commercial production of random, low vinyl solution SBR.

Although anionic initiated synthesis of random 10 medium vinyl solution SBB and random high vinyl solution SBR is easily accomplished by the addition of Lewis bases, these polar modifiers promote randomization at the expense of increased vinyl content (see Antkowiak, T.A.; Oberster, A. E.; Halasa, 15 A.F.; Tate, D.P. J. Polym. Sci., Part A-1, 1972, JO, 1319). Due to the large differences in monomer reactivity ratios of butadiene and styrene, measures must be taken to promote random incorporation of styrene into low vinyl solution SBR. In the absence 20 of such measures, the polymerization leads to a tapered block copolymer with inferior elastomeric performance characteristics (see United States Patent 3,558, 575).

British Patent 994,726 reports that it is 25 possible to produce random solution SBR by manipulating monomer polymerization rates via control of monomer concentrations throughout the polymerization process without the use of polar modifiers. For solution SBR, this requires that the 30 polymerization proceed in a styrene rich medium throughout the polymerization. In continuous polymerizations the issues associated with maintaining constant monomer concentration ratios while increasing conversion become quite complex.; 35 United States Patent 3,787,377 reports that alkali metal alkoxides (NaOR) can be used as polar

! - 3 modifiers in the copolymerization of styrene and butadiene to randomize styrene incorporation without significantly increasing the vinyl content of the rubber. However, alkali metal alkoxide modifiers are 5 so effective that they may actually increase the rate of polymerization of styrene to the extent that it is depleted before the polymerization is complete (see Hsieh, H.L.; Wofford, C. F. J. Polym. Sci., Part A-1, 1969, 7, 461-469). Furthermore, there is typically 10 some undesired increase in vinyl content over what would be expected from an unmodified polymerization (see Hsich, H.L.; Wofford, C. F. J. Polym. Sci., Part A-1, 1969, 7, 449-460).

15 Summary of the Invention

A method to prevent the formation of tapered block solution Son in unmodified polymerizations using standard continuous stirred tank reactors (CSTRs) has been developed. This method involves charging all of 20 the styrene and part of the 1,3-butadiene being polymerized into a first polymerization zone. The first polymerization zone is typically a continuous stirred tank reactor. The amount of styrene charged into the first polymerization zone will typically be 25 at least 5 percent more than the amount of styrene bound into the styrene-butadiene rubber being synthesized. It is important for a conversion within the range of 60 percent to 90 percent to be attained in the first polymerization zone.

30 Additional 1,3-butadiene monomer is charged into a second polymerization zone, such as a second continuous stirred tank reactor. Typically from 20 percent to 40 percent of the total amount 1,3-butadiene charged will be charged into the second 35 polymerization zone. It is also important for a 1,3-

butadiene conversion of at least about 90 percent to

- 4 - À. be attained ln the second polymerization zone and for the total conversion (styrene and 1,3-butadiene) to be limited to a maximum of about 95 percent in the second I polymerization zone. f S This invention more specifically discloses a process of synthesizing random styrene- butadiene rubber having a low level of branching and a low vinyl: content which comprises: (l) continuously charging 1,3-butadiene, styrene, an alkyl lithium initiator, 10 and an organic solvent into a first polymerization zone, (2) allowing the 1,3-butadiene and styrene to copolymerize in the first polymerization zone to total conversion which is within the range of 60 percent to 90 percent to produce a polymer IS cement containing living styrene-butadiene chains, (3) continuously charging the polymer cement containing; living styrene-butadiene chains and additional 1,3-: butadiene monomer into a second polymerization zone, I wherein from 20 percent to 40 percent of the total 20 amount of 1,3- butadiene changed is charged into the second polymerization zone, (4) allowing the copolymerization to continue in the second polymerization zone to a conversion of the 1,3-

butadiene monomer of at least 90 percent, wherein the 25 total conversion of styrene and 1,3-butadiene in the second polymerization zone is limited to a maximum of 95 percent, (5) withdrawing a polymer cement of random styrene-butadiene rubber having living chain ends from the second reaction zone, (6) killing the living chain 30 ends on the random styrenebutadiene rubber, and (7) I recovering the random styrene-butadiene rubber from the polymer cement, wherein the copolymerizations in: the first polymerization zone and the second: polymerization zone are carried out at a temperature 35 which is within the range of 70 C to 100 C, and wherein the amount of styrene charged into

- 5 - the first polymerization zone is at least 5 percent more than the total amount of styrene bound into the random styrene-butadiene rubber. The living chain ends on the random styrene-butadiene rubber can 5 optionally be killed by the addition of a coupling agent, such as tin tetrachloride.

The present invention also reveals a cement of living styrene-butadiene rubber which is comprised of an organic solvent and polymer chains that are derived 10 from 1,3-butadiene and styrene, wherein the polymer chains are terminated with lithium end groups, wherein the polymer chains have a vinyl content of less than 10 percent, wherein less than 5 percent of the total quantity of repeat units derived from styrene in the 15 polymer chains are in blocks containing five or more styrene repeat units, and wherein the molar amount of polar modifier in the cement of the living styrene-

butadiene rubber is at a level of less than 20 percent of the number of moles of lithium end groups on the 20 polymer chains of the living styrene-butadiene rubber.

Such cements of living styrene-butadiene rubber made by the process of this invention can be easily coupled because they contain very low levels of polar modifiers. Detailed Description of the Invention

The polymerizations of the present invention are carried out continuously in a first polymerization zone, such as a first reactor, and a second 30 polymerization zone, such as a second reactor. These copolymerizations of 1,3-butadiene and styrene are carried out in a hydrocarbon solvent which can be one or more aromatic, paraffinic or cycloparaffinic compounds. These solvents will normally contain from 35 4 to 10 carbon atoms per molecule and will be liquid under the conditions of the polymerization. Some

a' - 6 - representative examples of suitable organic solvents include pentane, isooctane, cyclohexane, methylcyclohexane, isohexane, n-heptane, e-octane, n-

hexane, benzene, toluene, xylene, ethylbenzene, 5 diethylbenzene, isobutylbenzene, petroleum ether, kerosene, petroleum spirits, petroleum naphtha and the like, alone or in admixture.

In the solution polymerizations of this invention, there will normally be from 5 to 30 weight 10 percent monomers in the polymerization medium. Such polymerization media are, of course, comprised of the organic solvent, monomers, and an initiator. In most cases, it will be preferred for the polymerization medium to contain from 10 to 25 weight percent 15 monomers. It is generally more preferred for the polymerization medium to contain 15 to 20 weight percent monomers.

In the polymerizations of this invention the styrene, l,3-butadiene, solvent, and initiator are 20 continuously charged into the first polymerization zone. All of the styrene and a portion of the 1,3-

butadiene is charged into the first polymerization zone. The amount of styrene charged into the first polymerization zone is at least 5 percent more than 25 the total amount of styrene bound into the random styrenebutadiene rubber being synthesized. In other words, at least 5 percent more styrene is charged into the first polymerization zone than will be polymerized during the polymerization in the first polymerization 30 and second polymerization zone. It is preferred for the amount of styrene charged into the first polymerization zone to be at least 7 percent more than the total amount of styrene bound into the random styrene-butadiene rubber being synthesized. It is 35 more preferred for the amount of styrene charged into the first polymerization zone to be at least 10

If - 7 percent more than the total amount of styrene bound into the random styrene-butadiene rubber being synthesized. The conversion attained in the first 5 polymerization zone will be within the range of 60 percent to 90 percent. It is preferred for the conversion attained in the first polymerization zone will be within the range of 75 percent to 85 percent. The polymer cement containing 10 living styrene-butadiene chains and additional 13-

butadiene monomer made in the first polymerization zone is continuously charged into a second polymerization zone. 20 percent to 40 percent of the total amount of 1,3-butadiene changed into the 15 first polymerization zone and the second polymerization zone is charged into the second! polymerization zone. Preferably from 25 percent to 35 percent of the total amount of 1,3-butadiene changed into the first polymerization zone and the second 20 polymerization zone is charged into the second polymerization zone. Most preferably from 27 percent to 33 percent of the total amount of 1,3-butadiene changed into the first polymerization zone and the second polymerization zone is charged into the second 25 polymerization zone.

It is critical for the total conversion (styrene and 1,3-butadiene) attained in the second polymerization zone to be held below about 95 percent and preferably below about 93 percent. However, the 30 1,3butadiene will be polymerized in the second reaction zone to a conversion of at least about 90 percent. The 1,3-butadiene will preferably be polymerized in the second reaction zone to a conversion of at least about 95 percent and will most 35 preferably be polymerized to a conversion of 98 percent.

- 8 - The copolymerizations of styrene and butadiene in the first polymerization zone and the second polymerization zone will be maintained at a temperature which is within the range of 70 C to S 100 C. At temperatures below about 70 C the polymerization is too slow to be commercially acceptable. On the other hand, at temperatures above 100 C thermal induced branching occurs to the extent that it adversely affects the hysteretic properties of 10 the styrene-butadiene rubber. For these reasons, the polymerization temperature will normally be maintained within the range of 75 C to 85 C, and will preferably be maintained within the range of 80 C to 90 C.

The styrene-butadiene rubber made utilizing the 15 technique of this invention is comprised of repeat units which are derived from 1,3butadiene and styrene. These styrene-butadiene rubbers will typically contain from 5 weight percent to 50 weight percent styrene and from 50 weight 20 percent to 95 weight percent 1,3-butadiene. The styrenebutadiene rubber will more typically contain from 10 weight percent to 30 weight percent styrene and from 70 weight percent to 90 weight percent 1, 3-butadiene. The styrene 25 butadiene rubber will preferably contain from 15 weight percent to 25 weight percent styrene and from 75 weight percent to 85 weight percent 1,3-butadiene.

In the styrene-butadiene rubbers of this 30 invention, the distribution of repeat units derived from styrene and butadiene is essentially random. The term "random" as used herein means that less than 5 percent of the total quantity of repeat units derived from styrene are in blocks containing five or more 35 styrene repeat units. In other words, more than 95 percent of the repeat units derived from styrene are

9 - in blocks containing less than five repeat units. A large quantity of repeat units derived from styrene will be in blocks containing only one styrene repeat unit. Such blocks containing one styrene repeat unit 5 are bound on both sides by repeat units which are derived from 1,3-butadiene.

In styrene-butadiene rubbers containing less than about 30 weight percent bound styrene which are made with the catalyst system of this invention, less than 10 2 percent of the total quantity of repeat units derived from styrene are in blocks containing five or more styrene repeat units. In other words, more than 98 percent of the repeat units derived from styrene are in blocks containing less than five repeat units.

15 In such styrene-butadiene rubbers, over 40 percent of repeat units derived from styrene will be in blocks containing only one styrene repeat unit, over 75 percent of the repeat units derived from styrene will be in blocks containing less than 3 repeat units and 20 over 95 percent of the repeat units derived from styrene will be in blocks containing less than 4 repeat units Normally less than 2 percent of the bound styrene in the styrene-butadiene rubber is in blocks of greater than 3 repeat units Preferably 25 less than 1 percent of the bound styrene in the styrenebutadiene rubber is in blocks of greater than 3 repeat units.

In styrene-butadiene rubbers containing less than about 20 weight percent bound styrene which are made 30 with the catalyst system of this invention, less than 1 percent of the total quantity of repeat units derived from styrene are in blocks containing 4 or more styrene repeat units. In other words, more than 99 percent of the repeat units derived from styrene 35 are in blocks containing less than 4 repeat units. In such styrene-butadiene rubbers, over 60 percent of

. repeat units derived from styrene will be in blocks containing only one styrene repeat unit and over 95 percent of the repeat units derived from styrene will be in blocks containing less than 3 repeat units.

5 Normally less than 2 percent of the bound styrene in the styrenebutadiene rubber is in blocks of greater than 3 repeat units. Preferably less than 1 percent of the bound styrene in the styrene-butadiene rubber is in blocks of greater than 3 repeat units.

10 The styrene-butadiene copolymers of this invention also have a consistent composition throughout their polymer chains. In other words, the styrene content of the polymer will be the same from the beginning to the end of the polymer chain. No 15 segments of at least 100 repeat units within the polymer will have a styrene content which differs from the total styrene content of the polymer by more than 10 percent. Such styrene-butadiene copolymers will typically contain no segments having a length of at 20 least 100 repeat units which have a styrene content which differs from the total styrene content of the polymer by more than about 5 percent.

The polymerizations of this invention are initiated by adding an organolithium compound to the 25 first polymerization zone containing the styrene and 1,3-butadiene monomers. The organolithium compounds which can be employed in the process of this invention include the monofunctional and multifunctional initiator types known for polymerizing the conjugated 30 diolefin monomers. The multifunctional organolithium initiators can be either specific organolithium compounds or can be multifunctional types which are not necessarily specific compounds but rather represent reproducible compositions of regulable 35 functionality, The choice of initiator can be governed by the

- 11 -

degree of branching and the degree of elasticity desired for the polymer, the nature of the feedstock and the like. With regard to the feedstock employed as the source of conjugated diene, for example, the 5 multifunctional initiator types generally are preferred when a low concentration diene stream is at least a portion of the feedstock, since some components present in the unpurified low concentration diene stream may tend to react with carbon lithium 10 bonds to deactivate the activity of the organolithium compound, thus necessitating the presence of sufficient lithium functionality so as to override such effects.

The multifunctional organolithium compounds which 15 can be used include those prepared by reacting an organomonolithium compounded with a multivinylphosphine or with a multivinylsilane, such a reaction preferably being conducted in an inert diluent such as a hydrocarbon or a mixture of a 20 hydrocarbon and a polar organic compound. The reaction between the multivinylsilane or multivinylphosphine and the organomonolithium compound can result in a precipitate which can be solubilized, if desired, by adding a solubllizing monomer such as a 25 conjugated diene or monovinyl aromatic compound, after reaction of the primary components. Alternatively, the reaction can be conducted in the presence of a minor amount of the solubilizing monomer. The relative amounts of the organomonolithium compound and 30 the multivinylsilane or the multivinylphosphine preferably should be in the range of about 0.33 to 4 moles of organomonolithium compound per mole of vinyl groups present in the multivinylsilane or multivinylphosphine employed. It should be noted that 35 such multifunctional initiators are commonly used as mixtures of compounds rather than as specific

t - 12 individual compounds. Exemplary organomonolithium compounds include ethyl lithium, isopropyl lithium, n-

butyllithium, sec-butyllithium, tert-octyl lithium, n-

eicosyl lithium, phenyl lithium, 2-naphthyllithium' 4 5 butylphenyl lithium, 4 -tolyl lithium, 4 -

phenylbutyllithium, cyclohexyl lithium and the like.

Exemplary multivinylsilane compounds include tetravinylsilane, methyltrivinylsilane, diethyldivinylsilane, di-n-dodecyldivinylsilane, 10 cyclohexyltrivinylsilane, phenyltrivinylsilane, benzyltrivinylsilane, (3ethylcyclohexyl) (3-n-

butylphenyl)divinylsilane and the like.

Exemplary multivinylphosphine compounds include trivinylphosphine, methyldivinylphosphine, 15 dodecyldivinylphosphine, phenyldivinylphosphine, cyclooctyldivinylphosphine and the like.

Other multifunctional polymerization initiators can be prepared by utilizing an organomonolithium compound, further together with a multivinylaromatic 20 compound and either a conjugated diene or monovinylaromatic compound or both. These ingredients can be charged initially, usually in the presence of a hydrocarbon or a mixture of a hydrocarbon and a polar organic compound as a diluent. Alternatively, a 25 multifunctional polymerization initiator can be prepared in a two-step process by reacting the organomonolithium compound with a conjugated diene or monovinyl aromatic compound additive and then adding the multivinyl aromatic compound Any of the 30 conjugated dienes or monovinyl aromatic compounds described can be employed. The ratio of conjugated diene or monovinyl aromatic compound additive employed preferably should be in the range of 2 to 15 moles of polymerizable compound per mole of 35 organolithium compound. The amount of multivinylaromatic compound employed preferably should

I, - 13

be in the range of 0.05 to 2 moles per mole of organomonolithium compound.

Exemplary multivinyl aromatic compounds include 1,2-divinylbenzene, 1,3divinylbenzene, 1,4 5 divinylbenzene, 1,2,4-trivinylbenzene, 1,3-

divinylnaphthalene, 1,8-divinylnaphthalene, 1,3,5-

trivinylnaphthalene, 2,4-divinylbiphenyl, 3,5,4'-

trivinylbiphenyl, m-diisopropenyl benzene, p-

diisopropenyl benzene, 1,3-divinyl-4,5,8 10 tributylnaphthalene and the like. Divinyl aromatic hydrocarbons containing up to 18 carbon atoms per molecule are preferred, particularly divinylbenzene as either the ortho, mete or pare isomer, and commercial divinylbenzene, which is a mixture of the three 15 isomers, and other compounds such as the ethyl styrenes, also is quite satisfactory.

Other types of multifunctional lithium compounds can be employed such as those prepared by contacting a sec- or tert-organomonolithium compound with 20 l,3-butadiene, at a ratio of to 9 moles of the organomonolithium compound per mole of the 1,3butadiene, in the absence of added polar material in this instance, with the contacting preferably being conducted in an inert hydrocarbon diluent, though 25 contacting without the diluent can be employed, if desired. Alternatively, specific organolithium compounds can be employed as initiators, if desired, in the preparation of polymers in accordance with the present 30 invention. These can be represented by R(Li)x wherein R represents a hydrocarbyl radical containing from 1 to 20 carbon atoms, and wherein x is an integer of 1 to 4. Exemplary organolithium compounds are methyl lithium, isopropyl lithium, n-butyllithium, sec 35 butyllithium, tert-octyl lithium, n-decyl lithium, phenyl lithium, 1-naphthyllithium, 4

- 14 butylphenyllithium, p-tolyl lithium, 4-

phenylbutyllithium, cyclohexyl lithium, 4-

butylcyclohexyllithium, 4-cyclohexylbutyllithium, dilithiomethane, 1,4dilithiobutane, 1,10 5 dilithiodecane, 1,20-dilithioeicosane, 1,4-

dilithiocyclohexane, 1,4-dilithio-2-butane, 1,8-

dilithio-3-decene, 1,2-dilithio-1,8-diphenyloctane, 1,4-dilithiobenzene, 1,4-dilithionaphthalene, 9,10-

dilithioanthracene, 1,2-dilithio-1,2-diphenylethane, 10 1,3,5trilithiopentane, 1,5,15-trilithioeicosane, 1, 3,5-trilithiocyclohexane, 1, 3, 5,8-tetralithiodecane, 1,5,10,20-tetralithioeicosane, 1, 2, 4! 6-

tetralithiocyclohexane, 4,4'-dilithiabiphenyl and the like. Some highly preferred functionalized 15 organolithium initiators are Nlithiopiperidine and 3-

pyrrolidine-l-propyllithium The organolithium compound will normally be present in the polymerization medium in an amount which is within the range of O.01 to 1 phm 20 (parts by 100 parts by weight of monomer). In most cases, from 0.01 phm to 0 1 phm of the organolithium compound will be utilized with it being preferred to utilize from 0 025 phm to 0.07 phm of the organolithium compound in the polymerization medium 25 Polar modifiers can be used to modify the microstructure of the rubbery polymer being synthesized. However, the amount of polar modifier employed should be limited to keep the vinyl content of the styrene-butadiene rubber being synthesized at a 30 low level. Ethers and amines which act as Lewis bases are representative examples of polar modifiers that can be utilized. Some specific examples of typical polar modifiers include diethyl ether, di-n-propyl ether, diisopropyl ether, di-n-butyl ether, 35 tetrahydrofuran, dioxane, ethylene glycol dimethyl ether, ethylene glycol diethyl ether, diethylene

r - 15 glycol dimethyl ether, diethylene glycol diethyl ether, triethylene glycol dimethyl ether, trimethylamine, triethylamine, N,N,N', N'-

tetramethylethylenediamine (TMEDA), N-methyl 5 morpholine, N-ethyl morpholine, N-phenyl morpholine and the like. Dipiperidinoethane, dipycrolidinoethane, tetramethylethylene diamine, diethylene glycol, dimethyl ether, TMEDA, tetrahydrofuran, piperidine, pyridine and 10 hexamethylimine are representative of highly preferred modifiers. United States Patent 4, 022, 95g describes the use of ethers and tertiary amines as polar modifiers in greater detail.

After the polymerization has reached the desired 15 level of conversion it is terminated using a standard technique. The polymerization can be texminated with a conventional noncoupling type of terminator (such as, water, an acid and/or a lower alcohol) or with a coupling agent 20 Coupling agents can be used in order to improve the cold flow characteristics of the rubber and -

rolling resistance of tires made therefrom. It also leads to better processability and other beneficial properties. A wide variety of compounds suitable for 25 such purposes can be employed. Some representative examples of suitable coupling agents include: multivinylaromatic compounds, multiepoxides, multiisocyanates, multiimines, multialdehydes, multiketones, multihalides, multianhydrides, 30 multiesters which are the esters of polyalcohols with monocarboxylic acids, and the diesters which are esters of monohydric alcohols with dicarboxylic acids and the like.

Examples of suitable multivinylaromatic compounds 35 include divinylbenzene, 1,2, 4-trivinylbenzene,

- 16 l,3-divinylnaphthalene, l,8-divinyluaphthalene, l, 3,5trivinylnaphthalene, 2,4-divinylbiphenyl and the like The divinylaromatic hydrocarbons are preferred, particularly divinylbenzene in either its ortho, mete 5 or pare isomer. Commercial divinylbenzene which is a mixture of the three isomers and other compounds is quite satisfactory.

While any multiepcxide can be used, liquids are preferred since they are more readily handled and form 10 a relatively small nucleus for the radial polymer.

Especially preferred among the multiepaxides are the epoxidized hydrocarbon polymers such as epcxidized liquid polybutadienes and the epoxidized vegetable -

oils such as epoxidized soybean oil and epoxidized 15 linseed oil. Other epoxy compounds, such as l,2,S,6,9,l0-triepaxydecane, also can be used.

Examples of suitable multiisocyanates include benzene-l,2,4-triisocyanate, naphthalene-l,2, 5,7-tetraisocyanate and the like.

20 Especially suitable is a commercially available product known as PAPIl, a polyarylpolyisocyanate having an average of three isocyanate groups per molecule and an average molecular weight of about 380.

Such a compound can be visualized as a series of 25 isocyanatesubstituted benzene rings joined through methylene linkages.

The multiimines, which are also known as multiaziridinyl compounds, preferably are those containing three or more aziridine rings per molecule.

30 Examples of such compounds include the triaziridinyl phosphine oxides or sulfides such as tri(l-ariridinyl)phosphine oxide, tri(2-methyl-lariridinyl)phosphine oxide, tri(2-ethyl-3-decyl-1-ariridinylphosphine sulfide and 35 the like.

The multialdehydes are represented by compounds - 17 such as 1, 4,7-naphthalene tricarboxyaldehyde, 1,7, 9-anthracene

tricarboxyaldehyde, 1, 1,5-pentane tricarboxyaldehyde and similar multialdehyde containing aliphatic and aromatic compounds. The 5 multiketones can be represented by compounds such as 1,4, 9,10anthraceneterone, 2,3-diacetonylcyclohexanone and the like. Examples of the multianhydrides include pyromellitic dianhydride, styrene-maleic anhydride copolymers and the like. Examples of the multiesters 10 include diethyladipate, triethyl citrate, 1, 3,5-tricarbethoxybenzene and the like.

The preferred multihalides are silicon tetrahalides (such as silicon tetrachloride, silicon tetrabromide and silicon tetraiodide) and the 15 trihalosilanes (such as trifluorosilane, trichlorosilane, trichloroethylsilane, tribromobenzylsilane and the like). Also preferred are the multihalogen-substituted hydrocarbons (such as, 1,3,5tri(bromomethyl)benzene and 20 2, 4,6,g-tetrachloro-3,7-decadiene) in which the halogen is attached to a carbon atom which is alpha to an activating group such as an ether linkage, a; carbonyl group or a carbonto-carbon double bond.

Substituents inert with respect to lithium atoms in 25 the terminally reactive polymer can also be present in the active halogen-containing compounds.

Alternatively, other suitable reactive groups different from the halogen as described above can be present. 30 Examples of compounds containing more than one type of functional group include 1,3-dichloro-2-propanone, 2,2-dibromo-3-decanone, 3, 5,5-trifluoro-4-octanone, 2,4-dibromo-3pentanone, 1,2,4,5-diepoxy-3-pentanone, 35 1,2, 4,5-diepoxy-3-hexanone, 1, 2, 11, 12-diepoxy-8-pentadecanone,

. Rae - 18 1,3,18,19-diepoxy-7,14-eicosanedione and the like In addition to the silicon multihalides as described hereinabove, other metal multihalides, particularly those of tin, lead or germanium, also can 5 be readily employed as coupling and branching agents.

Difunctional counterparts of these agents also can be employed, whereby a linear polymer rather than a branched polymer results. Monofunctional counterparts can be used to end cap the rubbery polymer. For 10 instance, trialkyl tin chlorides, such as tri-isobutyl tin chloride, can be utilized to end cap the rubbery polymer. I Broadly, and exemplary, in the case of tetraLunctional coupling agents, such as tin 15 tetrachloride, a range of 0.01 to 1 moles of-

coupling agent are employed per mole of lithium in the initiator. To attain a maximum level of coupling, it is preferred to utilize 0.1 to 2.5 moles of the coupling agent per mole of lithium in the 20 initiator. The larger quantities tend to result in I production of polymers containing terminally reactive groups or insufficient coupling. The coupling agent can be added in hydrocarbon solution (e.g., in cyclohexane) to the polymerization admixture in the 25 final reactor with suitable mixing for distribution and reaction.

After the copolymerization has been completed, the styrene-butadiene elastomer can be recovered from the organic solvent. The styrenebutadiene rubber can 30 be recovered from the organic solvent and residue by means such as Recantation, filtration, centrification and the like. It is often desirable to precipitate the segmented polymer from the organic solvent by the addition of lower alcohols containing from 1 to 35 4 carbon atoms to the polymer solution.

Suitable lower alcohols for precipitation of the

r - 19 segmented polymer from the polymer cement include methanol, ethanol, isopropyl alcohol, normal-propyl alcohol and t-butyl alcohol. The utilization of lower alcohols to precipitate the rubber from the polymer 5 cement also "kills" the living polymer by inactivating lithium end groups After the segmented polymer is recovered from the solution, steam-stripping can be employed to reduce the level of volatile organic compounds in the rubber.

10 This invention is illustrated by the following examples which are merely for the purpose of illustration and are not to be regarded as limiting the scope of the invention or the manner in which it can be practiced. Unless specifically indicated 15 otherwise all parts and percentages are given by weight. Examples

In order to best evaluate different methodologies 20 (split feed, alkali metal alkoxide modified, and unmodified), four target polymers were chosen with varying styrene levels and Mooney viscosity targets.

The target polymers are listed in Table I along with the desired styrene, vinyl, and Moony ML:+ viscosity 25 at 100 C for each sample. For the purposes of these examples each sample type (1-4) will be further classified as being synthesized via Split Feed (a), Sodium t-amylate (STA) modification (b), or unmodified polymerization(c).

- 20 TABLEI

TARGETED LOW V]NYLSSBR POLYMERS

Sample Styrene Vinyl Mooney Type Target (wt%) Target Viscosity (wt%) (ML14(1 00 C)) 1(a-c) 18 10 40+/-5 2(a-c) 25 10 40+/-5 3(a-c) 30 10 40+/5 4(a) 25 10 45+/-5 OF (-1 10 base) 5 All of the samples presented herein were prepared via anionic chain polymerization using n-butyllithium as the initiator except where noted. The polymers were prepared in a continuous two-reactor chain where each reactor was equipped with two axial flow turbines 10 (AFTs) and baffles. The agitation speed was 200- 250 rem with the AFT s pumping down. The polymerizations were controlled through use of a Foxhoro distributive control system. The polymerizations were terminated with rosin acid at a level of 1 pier, and Wingstay Ken 15 antioxidant was added at a level of 0.5 pier. In Sample Type 4, the highly aromatic oil (37.5 parts) was post added to the polymer cement before isolation by steam stripping, extruder dewaterlng, and oven drying. Preparations Styrene and 1,3-butadiene solutions were made up at 14 to 15 weight percent in hexanes (mixed hexane 25 isomers) with 1,2-butadiene added for gel suppression at a level of 100-150 ppm (based on total monomer).

The monomer solutions were purified by passing over molecular sieves and silica gel. The initiators, sodium alkoxide, rosin acid, and antioxidant were 30 diluted with hexanes.

- 21 Analytical Testing Reactor conversions were determined by 5 gravimetric analysis. A Hewlett Packard (RTM) 5890 Series: II gas chromatograph was used for residual monomer analysis. Mooney viscosities (ML+ (100 C)) were measured on a Flexsys (RTM) MV2000. Glass transition temperatures (onset on heat @ 10 C/min) were measured 10 on a TA 2910 DSC. Molecular weights were measured by multi-angle light scattering (MANS) GPC.

Microstructures were measured by FTIR (Nicolet(RTM) 510) and proton NMR (Varian (RTM) 300). Block styrenc was determined by an ozonolysis procedure derived from a 15 published procedure (see Tanaka, Y.; Sato, H.; Adachi, J. Rubber Chem. Technol. 1986, 59, 16). All polymers were evaluated in typical tread formulations with the; experimental SSBR constituting 100% of the elastomeric ingredients. Carbon black loadings were 50 parts for 20 the non-oil extended samples and 65 parts for the oil extended samples. Compound cures were optimized from rheometer data.

Split Feed SSBR Model and Preparation The Split Feed samples were prepared by splitting the total 1,3-butadiene monomer feed stream into each reactor in a series of CSTRs (see Tirrell, M; Galvin, R.; Laurence, R. L. In Polymerization Reactors in 30 Chemical Reaction and Reactor Engineering; Carberry, J. J. and Varma, A, Eds.; Marcel Dekker, New York, 1987; and Saldivar, E. PhD Thesis; Modeling and Control of Emulsion Copolymerization Reactors, University of Wisconsin-Madison, 1996). The reactor 35 model used to determine split ratios is based upon the first principal approach: the material and energy balance of each species in each reactor. For a two

- 22 reactor split feed process, the governing equations are described as follows: msm -FOUI} [St]l -V (kil [P]} [St]l +k21 [Q]I [SI]l)=0 5 (1) mBD, n (1 f) Fout l [BD]I - Vl (kl2 [P]l [BD]t + k2 [Q]IEBD]I)=0 (2) 10 mP inl t I [P]l -vl (kl2 [P]I [BD]' -k21 [Q]l [St]l)=0 (3) mp in + mg.n-FoYt 1 ( [P]l + [Q]l) = 0 (4) it.1 [St]l -Fo2z [st]2 -V2 (kil [P]2 [St]2 +k21 [Q] 2 [st]2)-o (5) Fo l [BD]' + mD ' f - Fo' 2 [BD]2 - V2 (kl 2 [P]2 [BD j2 ka [Q]2 [BD]2) = 0 20 (6)

FoY,1 [P]I - FO! 2 [P]2 - F2 (kl2 [P]2 [BD]2 - k2' [Q]2 [St]2) = 0 (7) 25 FOU! 1 ( [P]I+[QL)_FOUI2 (]2 + [Q]2)=0

() (mSt,tn - F(7U!,][SI]l) MW5, (mSr,n-FOur.1 [5]1) MWst + (mBD.m (1 f)FS I [BD]I) MWBD (FOU, I []I - FoUt'2 [St]2). Mws' (FOUt l [S{]l - F -, 2 [St]2) MWSI + (mBD,n f Jr F - [ l [BD]I - Fout 2 [BD]2) MVBD (9) In these equations, [P.] and [] are the concentrations of live polymers with the end group of 35 styrene and butadiene, respectively. [St] and [BD] are the concentrations of monomer styrene and butadiene in the reactor. FOueis the volumetric flow rate leaving the reactor.- m. in is the mole rate for

- 23 each species flowing into the reactor. Except where they appear in rate constants, the subscripts 1 and 2 correspond to the first and the second reactor. V is the reaction volume occupied by the reaction mass. f 5 is the percentage of butadiene fed into the second reactor that reflects the split ratio of butadiene to the two reactors. Ma is the molecular weight of monomer. kin stands for the propagation rate constants with the chain end of group i by adding monomer j on 10 it. Because of the association effects in anionic polymerization, the general forms of the rate constant for styrene (kin) and butadiene (ken) are given by: k,,fP]=(1- f)[l734 102exp(- 18RT4) [p]/2]+ f[2.878 10' exp(-] RTO)[P]] ,, un mod iced mod idled (10) k22[Q]=(l-g)[1994 10'2exp(-iRT6)[Q] "3]+g[7024 108exp(-]RT2)-[Q]] ,, un mod idled mod Wed (11) where fMR (=0. 5 MR) and Gem (=0.5 MR) are functions of 25 modifier ratio, MR (i.e. the ratio of modifier to the initiator). The cross-propagation rate constants \2 and k2, are determined from the homopolymerization rate constants and the reactivity ratios rig and r21. These 30 ratios were determined experimentally and found to be strong functions of modifier ratios. The detailed expressions were reported by Chang et al (see Chang, C; Miller Jr., J.; Schorr, G. J. App. Polym. Sci., 1990, 39, 2395; and Chang, C.; Halasa, A. F.; Miller

f - 24 - 3r.,,J.; Schorr, G.; Presented at the ACS Rubber Division Meeting, May, 1990).

Given the reactor operating conditions, the independent variables in the above equations are the 5 concentrations of monomer species and live polymer chains in two reactors and the amount of butadiene fed to the second reactor, f. The total of nine variables and nine equations was solved using a commercially available software package.

lOThe 1,3-butadiene solution was split between the first and second reactors as dictated by the above model. The styrene monomer and initiator solutions were added to the first reactor. The first and second reactor residence times for the non-oil extended 15 polymers were 37 minutes and 58 minutes, respectively.

The first and second reactor residence times used for the oil extended sample were 49 minutes and 78 minutes, respectively. The polymerization temperature used for all the Split Feed polymers was 90 C.

STA SSBR Preparation The sodium t-amylate (STA) modified samples were prepared by addition of the STA to the first reactor 25 at a 0.15 molar ratio to n-butyllithium. The n-

butyllithium and all of the 1,3-butadiene and styrene solutions were added to the first reactor. The first and second reactor residence times for the non-oil polymers were 29 minutes and 58 minutes, respectively.

30 The first and second reactor residence times used for the oil extended sample were 33 minutes and 66 minutes, respectively. The polymerization temperature used for the STA modified polymers was 95 C.

35 Unmodified SSBR Preparation The unmodified samples were prepared by adding the n-butyllithium and all of the 1,3-butadiene and

f - 25 styrene solutions to the first reactor. The ranges for the first and second reactor residence times for the non-oil polymers were 36-42 minutes and 72-84 minutes, respectively. The first and second reactor residence 5 times used for the oil extended sample were 39 minutes and 78 minutes, respectively. The polymerization temperature used for the unmodified polymers was 90 C.

Model Validation To verify the Split Feed reactor model, experimental data was analyzed from the four Split Feed SSBRs prepared as previously described. First and second reactor conversions of l,3-butadiene and 15 styrene, as well as the overall monomer conversions are shown in Table II.

. - 26 - TABLE IL

COMPARISON OF PREDICTED AND EXPERIMENTAL

CONVERSIONS

Conversions (wt%) by Polymer Sample 1 a 2a 3a 4a Butadien Reactor 1 Experimental 96 95 93 93 Model 94 94 94 94 prediction Reactor 2 Experimental 95 96 95 93 Model 95 96 96 96 prediction Total Experimental 99 99 98 98 Model 98 98 99 99 prediction Styrene Reactor 1 Experimental 54 56 54 50 Model 56 57 57 58 prediction Reactor 2 Experimental 60 55 61 59 Model 64 65 68 66 prediction Total Experimental 82 80 82 79 Model 84 85 86 86 prediction s From the data in Table II, it is observed that the model's predictions of butadiene conversions in each reactor match the experimental data quite well.

In addition, the predicted conversions of styrene are 10 in general agreement with the experimental data, but the model tends to over-predict styrene conversions as targeted bound styrene (3a) or targeted molecular weight (4a) of the polymer increases. Deviations of the model predictions from experimental data on 15 styrene conversion may be due to poor estimates of the styrene homopolymerization rate constant in hydrocarbon solvent resulting from poor Volubility of polystyrene.

- 27 The weight percent bound styrene for each polymer is shown in Figure 2 (a), (b) and (c) by reactor and final polymer. It is observed that the model slightly over-predicts polymer bound styrene in the 5 first reactor. In the second reactor, the model predictions are case dependent. This variability may be traced to magnification of deviations from the first reactor predictions. For the final polymer, STIR was used to analyze the average styrene content 10 in the polymer. In conclusion, analysis of the final

polymer indicated that the model predictions match the measurements quite well with the largest error at about 9% under the predicted value.

_ _ (a) Reactor 1 (b) Reactor 2 - 1 1 _ 35 in' ---| &B 301 |Me| alp | g i l l 2 4 Sample Type Sample Type (c) Final Polymer l E hat Bal l ICY. '-flj ='OO:'ll,-,---1 1. 2 3 4

Sample Type

- 28 Figure 2. Comparison of bound styrene levels as estimated by residual monomer analysis and as predicted by our model. Actual FTIR data is included for final polymer comparisons.

s Raw Solution SBR Properties The three methodologies were initially compared based on the raw polymer properties of the four targeted polymers. Bound styrene levels of all the 10 samples fell within reasonable variation (a/- 3) of targeted values with the exception of 2b which was high in styrene (30%). Aside from this sample, glass transition temperature variations seen in the polymers were largely determined based on the vinyl content of 15 the samples. Table III summarizes styrene levels, block styrene, vinyl content, Tg, and Mw of the samples prepared in this study.

- 29 TABLE m

RAW POLYMER PROPERTIES

Styrene Block Vinyl To Mw Styrene (%) (wt %) (ADS) (wt %) ( C) (K) 18% Styrene _ Split Feed 17.5 0 10.0 -73.5 222 (1a) Unmodified 19.5 0.8 10.6 77.5 285 (1b) STA (1c) 19.3 0 16.1 9.9 258 25% Styrene Split Feed 24. 1 0 8.5 -66.3 239 (2a) Unmodified 30.2 6.4 10.7 -59.8 268 (2b) STA (2c) 25.9 0.5 14.5 -62.6 268 30% Styrene __ Split Feed 27.2 0 8.2 0.9 236 (3a) Unmodified 29.8 10.4 9.1 6.6 301 (3b) STA (3c) 30.0 0.8 13.3 -56.1 247 25% Styrene Split Feed 24.8 0 9.4 0 3 477 (4a) Unmodified 22.1 0.9 9.8 49. 5 389 (4b) STA (4c) 24.5 0.S 16.4 -57.7 636 5 As may be seen from the tabulated data, the STA modified polymerizations do not lead to the low vinyl levels found in unmodified systems, producing SSBRs with vinyl levels that are - 30-80% higher relative to the unmodified and split feed systems. As expected 10 both the Split Feed and unmodified samples showed vinyl levels that centered around 9-10% of the total monomer. In addition, the molecular weights of the STA modified samples were generally higher for a given Mooney viscosity. Consistent with known effects of

- 30 NaOR type modifiers, this might indicate a higher degree of branching for the STA modified polymers relative to the unmodified samples.

The samples were analyzed for block styrene using 5 ozone to cleave the polymer followed by chromatographic analysis of the fragments. The results of this analysis, summarized in the second column of Table III, illustrate the effectiveness of the split feed methodology in producing random SSBR.

10 None of the split feed samples contained any block styrene sequences greater than 3 repeat units in length. Although the STA modified polymers did contain very small amounts of block styrene in the higher styrene samples, the randomization was quite effective 15 in that 3a contained <1% block styrene. As expected, the unmodified samples were especially blocky, containing as much as 10% block styrene in the 30% bound styrene sample, 3b. This, again, underscores the effectiveness of the split feed system in 20 preventing block styrene by manipulating monomer concentrations as a control of monomer polymerization rates. Physical Properties 25 Ultimately, the evaluation of our compounded samples in terms of tire performance properties was of paramount importance in evaluation of the three methodologies for producing random, low vinyl SSBR.

Given government mandated Corporate Average Fuel 30 Economy standards for the automobile industry, hysteresis is perhaps the most important factor in selecting elastomers for tire tread applications.

Hysteretic properties of tread compounds of elastomers prepared using the three methodologies were evaluated 35 by Goodyear-Healey Hot Rebound, Goodrich Flexometer

- 31 Heat Buildup, and Autovibron tan 6. Data from these evaluations are summarized below in Table IV.

TABLE TV

5HYSTERETIC ANALYSTS OFSSBRPOLY5IERS

. Goodyear- Goodrich Autovibro Healey Flex n Hot Rebound aT ( C) tan 18% Styrene | _; Split Feed (1a) 75.7 16.6 0.086 Unmodified 76.3 20.6 0.097 (1b) STA (1c) 73.4 18.9 0.096 25% Sfyrene Split Feed (2a) 78.6 17.2 0.080 Unmodified 71.2 18.9 0.116 (2b) STA (2c) 71.2 21.1 0.117 30% Styrene | =_ Split Feed (3a) 76.9 17.8 0.092 Unmodified 74 17.8 0.116 (3b) STA (3c) 71. 2 20.0 0.131 25Yo Styrene = Split Feed (4a) 76.9 18.3 0.144 Unmodified 74. 6 16.7 0.126 (4b) STA (4c) 71.2 17.8 0.124 As may be seen from the data, in nearly all of the sample types and analytical methods, the Split Feed methodology produced the polymer having the l0 lowest hysteresis. For example, in the case of sample type 3 (30% styrene) the Split Feed SSBR, 3a, was 30% lower in tan O than the STA sample, 3c, and 21% lower tan O than the unmodified SSBR sample, 3b. Heat buildup and rebound measurements of hysteresis were in 15 general agreement with the tan 6's for sample types 1 3. However, the oil extended sample, 4a, was an..DTD:

- 32 exception in that the advantages of the split feed polymer were not seen in the tan or heat buildup measurements. Although some of the differences may be attributed to molecular weight effects, it should be 5 noted that the rebound, heat buildup, and tan measurements are inconsistent across sample types 4a-

c, and we are suspicious of their validity. The hysteretic performance of these samples is currently being re-evaluated.

10 A combination of factors is likely to contribute to results found in our hysteretic analyses. First, randomization of styrene minimizes the formation of non-elastomeric block polymer which can contribute significantly to hysteresis. Second, from comparisons 15 of molecular weights and radii of gyration calculated from light scattering measurements, we have found that the STA samples are more branched than the Split Feed polymers. Branched polymers have more chain ends which are elastically ineffective and contribute to 20 increased hysteresis (see Flory, P.J. Principles of Polymer Chemistry; Cornell University: Ithaca, New York, 1953; p.461; and Aggarwal, S. L; Fabris, H. J.; Hargis, I G.; Livigni, R. A. Polym. Prepr. -Am. Chem. Soc. Div. Polym. Chem., 1985, 26 (2J, 3).

25 Thus, the hysteretic advantages of the Split Feed polymers may be traced back to their linear microstructure and lack of block styrene. The linearity of the Split Feed polymers is a direct result of the unmodified nature of the polymerization 30 which is carried out at relatively low temperature under isothermal operating conditions. Although the Split Feed process has the versatility to be operated adiabatically as well as isothermally, operating under isothermal conditions at lower temperatures minimizes 35 lithium hydride elimination and reduces thermal

- 33 branching resulting in a polymer with improved hysteretic properties. As expected, the unmodified sample suffers from the presence of nonelastomeric block polymer which contributes to hysteretic loss.

5 In light of the proliferation of high mileage warranty tires, wear performance is another important consideration for SSBR elastomers designed for tire tread applications. Though laboratory abrasion tests have notoriously poor correlation to actual tire wear 10 testing, when such tests are consistent across samples and methods, performance trends may be predicted with a fair degree of accuracy. DIN abrasion and Pico abrasion tests were used as predictors of tire tread-

wear performance. For each sample type the Split Feed 15 SSBRs were the most resistant to abrasion averaging l6% better in volume loss from DIN abrasion and 12% better in Pico abrasion index. Although wear performance may generally be correlated back to glass transition temperature (Tg) of the polymers, in these 20 cases the split feed polymers did not always have the lowest Tg in each series. The STA samples (having the highest Tg in Sample Types 1,3, & 9) performed the worst overall in abrasion resistance. Data are summarized below in Table V.

- 34 TABLE V

ABRAS10NPERFORAqANCEOFSSBPOLYMERS DIN Abrasion Pico Abrasion RelVol Index Loss(mm3) 18% Styrene _ Split Feed (1a) 83 144 Unmodified 87 143 (1 b) STA (1c) 94 123 ...,. _

25% Styrene _ Split Feed (2a) 98 136 Unmodified 116 121 (2b) STA(2c) 116 115 30% Styrene _ Split Feed (3a) 102 126 Unmodified 111 1 17 (3b) STA(3c) 120 112 25% Styrene _ OE _ -

Split Feed (4a) 87 97 Unmodified 109 95 (4b) STA (4c) 115 95 5 In our evaluations, the tensile properties were essentially equal for the samples, and no advantage could be detected for one methodology over another.

Whereas the Split Feed SSBRs might be expected to be deficient in tear properties due to their low vinyl 10 microstructure and linear microstructure, results of our tear test analysis are case dependent.

Variability in tear properties also prevents us from drawing any broad conclusions as to the benefits of

one methodology over another. The Split Feed SSBRs do, 15 however, suffer from the typical tradeoffs characteristic of linear low vinyl SSBRs in terms of

- 35 wet traction (as predicted by tan at 0 C) versus rolling resistance. Tensile, tangent (0 C), and tear data are summarized in Table VI below.

5 TABLE VI

MISCELANEOUS PHYSICAL DATA FROM SSBR POLYMERS

_ _ Tensile 300% Modulus Elongation Autovibron Tear Die (B) (kPa? (kPa) (%) tan (0 C) (kN/m) 18% Styrene Split Feed (1 a) 19200 13400 399 0. 119 43.4 Unmodified 21800 14000 442 0.124 45.7 (1 b) STA (1c) 21500 12800 455 0.128 37.6 25% Styrene Split Feed (2a) 21200 14500 408 0. 117 33.1 Unmodified 20300 14000 419 0.122 45.7 (2h) STA (2c) 20300 11600 496 0.138 47.6 30 /0 Styrene Split Feed (3a) 22700 14800 438 0.128 44.8 Unmodified 22100 13500 475 0.116 48.1 (3b) STA (3c) 22800 12300 518 0.151 47.3 25Yo Styrene _ Split Feed (4a) 20900 8700 609 0.168 45.2 Unmodified 20200 8212 628 0.163 48.1 (4b) STA (4c) 21100 8860 610 0.188 41.5 Additional Split Feed Studies 10 To summarize the results of our comparison of methodologies for synthesis of random, low vinyl SSBR, we have found that the Split Feed technology clearly performs best for producing SSBR intended for tire tread applications. Furthermore, the reproducibility 15 of the Split Feed technique has been verified in multiple preparations of sample type 2a. From our studies, it was decided to further investigate the

versatility of Split Feed technology in producing high styrene and functional SSBR. Although the results discussed below are preliminary, they reflect the direction of our current efforts in exploiting the 5 true value of the Split Feed technology.

High Styrene SSBR In order to test the upper limits of the ability 10 of Split Feed technology to randomize styrene, we synthesized a 35% styrene SSBR in hexane. Although the sample was approaching the upper limits of solubility for high styrene SSBR in hexanes, the polymerization proceeded well and the polymer remained 15 in solution. Analysis of the final product using 'H NOR revealed that of the 35% bound styrene, only 1% was block. Vinyl content of the polymer was only 8% by 1H NOR. It is likely that even higher levels of styrene could be randomized if the polymerization were 20 to be carried out in cyclohexane, a solvent better suited for high styrene SSBR.

Functional SSBRs Via Split Feed The use of functional groups to initiate, 25 terminate, and/or couple living anionic polymers has been studied for many years. Out of the many candidates available, several functional groups are reported to have beneficial effects in terms of improved processing, improved filler dispersion, and 30 reduced hysteresis of rubber compounds. However, the compatibility of the functional group with the polymerization process is sometimes an issue. For example, some polar modifiers have an adverse effect on coupling efficiencies, and high polymerization 35 temperatures can lead to thermally induced termination reactions. Because the Split Feed SSBRs are produced

- 37 without modifiers and under controlled temperature conditions, they are more amenable to functional termination and coupling than other processes we have; considered. 5 In several experiments, we have tin coupled split feed n-butyl lithium initiated SSBRs to extremely high coupling efficiencies as measured by Mooney viscosity jump. For example, a low Mooney (30 base ML1+4 (100 C)) 25t styrene SSBR was coupled to 90 ML1+ (100 C).

10 Coupling efficiencies of this degree are indicative of a highly living polymer - a characteristic that makes the Split Feed technology an extremely versatile tool for the preparation of coupled, partially coupled, or otherwise functionally terminated polymers.

15 It has been reported that secondary amine! initiated polymers suffer from loss of functional head groups when the living polymers are exposed to elevated process temperatures (see Lawson, D.F.; Brumbaugh, D.R.; Stayer, M.L.; Schreffler, J.R.; 20 Antkowiak, T.A.; Saffles, D; Morita, K. i Ozawa, Y.; Nakayama, S. Polym. Prepr.-Am. Chem. Soc. Div. Polym.

Chem., 1996, 37 (2), 728). In contrast to adiabatic processes, the lower temperature isothermal conditions seen in Split Feed polymerizations allow for higher 25 head retention in functionally initiated polymers. By controlling the temperature of the polymerization, we are better able to maximize the number of functional chain ends and consequently the effectiveness of the functional group in improving elastomer properties.

30 We have conducted preliminary investigations with the split feed technology using two functional initiators available on a developmental basis from FMC Corporation. These initiators, 3-tN,N-dimethylamino)- i 1propyllithium (DMAPLi), and 3-(t 35 butyldimethylsilyloxy)-1propyllithium (TBDMSPLi),:

- 38 -

H3C\ CH3 I H3

/NCH2CH2CH2Li H3C-si-ocH2cH2cH2Li] H3C,H3 CH3 3-(N,Nimethylamino)- 1 -propyllithium 3-(t-butyldimethylsilyloxy)-1 -

propyllithium 5 6 were effective in Initiating the synthesis of SSBR using the Split Feed technology.

In addition, the DMAPLi initiated sample was coupled using SnCl4. Raw polymer properties are listed 5 in Table VII. As may be seen from the data, although the functional initiators have some degree of Lewis; basicity, they produce low vinyl solution SBR. In; addition, 1H NOR showed no evidence of block styrene for these samples. Furthermore, the coupling 10 efficiency of the DMAPLi initiated SSBR was quite high; as measured by Mooney viscosity jump upon coupling.

TABLE VII

FUNCTIONAL SBR RAW POLYMER DATA

Initiator 5 6 Styrene Target 25 18 (%): Bound Styrene 23.5 16.9; (%) Vinyl (%) 11.6 11.3 L1+4 (base) 30 49 ML14 116 N/A

(coupled) TO ( C) 64.5 -75

Split Feed technology is clearly the best method tested for synthesis of random low vinyl SSBR for use; in tire tread applications. As evidence, Split Feed 20 SSBRs have the lowest block styrene and the lowest i vinyl of all sample types tested, allowing us to produce the lowest glass transition temperature at a given styrene content. This is reflected in superior

l f r - 39 abrasion resistance as measured by both DIN and Pico abrasion. In addition, due to the controlled polymerization temperature and lack of polar modifiers, the polymers do not suffer from branching 5 which would have adverse effects on hysteretic performance. As a result, the Split Feed SSBRs consistently displayed the lowest hysteresis of the samples tested. Furthermore, in minimizing thermal branching and termination reactions typical of high 10 temperature processes, the split feed methodology ultimately allows synthesis of polymers with higher degrees of functionality than would be possible via an adiabatic or polar modified process. And finally, by avoiding the introduction of polar modifiers in

15 commercial operations, environmental concerns and demands on recycle and recovery processes are minimized. Low levels or the absence of polar modifiers in polymer cements of living SBR also makes coupling much more efficient.

20 While certain representative embodiments and details have been shown for the purpose of illustrating the subject invention, it will be apparent to those skilled in this art that various changes and modifications can be made therein without 25 departing from the scope of the subject invention, as embodied in the appended claims.

Claims (6)

t l CLAIMS

1. A cement of living styrene-butadiene rubber which is comprised of an organic solvent and polymer chains that are derived from l,3-butadiene and styrene, wherein the polymer 5 chains are terminated with lithium end groups, wherein the polymer chains have a vinyl content of less than lO percent, wherein less than 5 percent of the total quantity of repeat units derived from styrene in the polymer chains are in blocks containing five or more styrene repeat units, and 10 wherein the molar amount of polar modifier in the cement of the living styrene-butadiene rubber is at a level of less than 20 percent of the number of moles of lithium end groups on the polymer chains of the living styrene-butadiene rubber.

15

2. The cement of living styrene-butadiene rubber as specified in claim l wherein said cement of the living styrene-butadiene rubber is void of polar modifiers.

3. The cement of living styrene-butadiene rubber as 20 specified in claim l wherein said polymer chains are derived from 5 weight percent to 50 weight percent styrene and from 50 weight percent to 95 weight percent l, 3-butadiene.

4. The cement of living styrene-butadiene rubber as 25 specified in claim l wherein said polymer chains are derived from lO weight percent to 30 weight percent styrene and from 70 weight percent to 90 weight percent l, 3-butadiene.

5. The cement of living styrene-butadiene rubber as 30 specified in claim l wherein said polymer chains are derived from 15 weight percent to 25 weight percent styrene and from 75 weight percent to 85 weight percent l, 3-butadiene.

-

6. A cement of living styrene-butadiene rubber substantially as hereinbefore described in any one of the foregoing Examples la, 2a, 3a and 4a.