V. COMPARISON OF SBRs IN TIRE COMPOUNDS

Figure 23 Effect of Tg on hot rebound in solution SBR. (x) Change in vinyl

content (15% styrene); (n) change in styrene content (30% vinyl). (From Ref. 9 with

permission of the copyright holders Rapra Technology Ltd.)



Figure 24 Effect of Tg on wear resistance in solution SBR (lower is better). (x)

Change in vinyl content (15% styrene); (n) change in styrene content (30% vinyl).

(From Ref. 9 with permission of the copyright holders Rapra Technology Ltd.)



Copyright © 2004 by Taylor & Francis



effect of polymer Tg (Table 12). A compound made with an SBR of 200,000

molecular weight and Tg of À65jC (polymer C) had the same loss tangent at

66jC as a compound made with a 133,000 molecular weight SBR and a Tg of

À96jC (polymer A) (12).

Day and Futamura also compared two solution SBRs and a high-vinyl

polybutadiene with the same glass transition temperature and similar macrostructures. Cure time was longer, and tensile strength, elongation, and tear

strength were poorer as the vinyl content of the polymers increased (Table 13)

(Fig. 25).

Finally, all rubber technologists should appreciate that solution SBRs

with similar Tg values and styrene and vinyl contents do not necessarily

process the same or show exactly the same compounded properties. This is

due to macrostructural differences and was illustrated by Kerns and Henning

(6) in their study of the effect of synthesis parameters on polymer structure.

Some common initiator systems, such as the alkali metal alkoxide–butyllithium system can behave as a ‘‘superbase’’ and abstract a proton from the

backbone of the polymer. This creates a site for branching and results in

higher hysteresis in compounds. They developed a method for measuring the



Table 12



Effect of Styrene Level on Compounded Properties

Polymer



Properties



A



B



C



D



E



Raw polymer properties

Styrene %

0

18

25

35

35

Vinyl %

12.0

9.8

9.0

7.8

7.8

Tg (jC)

À96

À75

À65

À53

À45

MWD

1.75

2.05

1.85

1.8

1.8

ML 1+4 (100jC)

55

100

110

148

148

Mn

133,000

200,000

200,000

225,000

225,000

Compounded properties

Rheometer at 150jC

TS2

8.0

9.5

10.5

10.3

9.0

T50

15.8

16.0

17.0

17.6

16.7

21.3

22.5

25.0

24.8

25.0

T90

300% Modulus (MPa)

6.99

9.30

8.27

10.68

11.02

Elongation (%)

515

515

550

535

510

Die C (kN/m)

36.9

41.1

44.8

42.9

42.9

Tan y at 20jC

0.196

0.207

0.224

0.238

0.213

Tan y at 66jC

0.175

0.161

0.177

0.158

0.140

Source: Ref. 12.



Copyright © 2004 by Taylor & Francis



Table 13 Effect of Styrene and Vinyl at Constant Tg

Polymer

Properties

Raw polymer properties

Styrene %

Vinyl %

Tg (jC)

ML 1+4 (100jC)

Mn

Compounded properties

Rheometer at 150jC

TS2

T50

T90

300% Modulus (MPa)

Elongation (%)

Die C tear (kN/m)

Tan y at 20jC

Tan y at 66jC



F

0

54

À52

97

195,000



9.5

20.5

32

8.96

465

40.1

0.226

0.159



G

17.6

31

À51

102

182,000



10

18

27.7

8.96

545

43.6

0.241

0.165



H

28.3

15

À51

97

165,000



10

17.0

23.3

8.96

555

45.7

0.247

0.171



Source: Ref. 12.



ability of different initiators to induce branching, then studied these initiators

in a series of high-vinyl SBRs. Branching was characterized by Mooney stress

relaxation, gel permeation chromatography, and crossover frequency of

elastic and storage shear modulus. The relative order of branching was the

same regardless of which technique was used. The data are shown in Table 14

and Figure 26 and 27.

The polymers in runs 1–3 are virtually identical in most ways that are

included in a typical specification sheet but would be expected to process

differently as evidenced by the Mooney force decay (T80, time to 80%

relaxation). The higher value indicates higher branching. Thus, the n-butyllithium/sodium t-amylate initiator produces a polymer that is more branched

than the n-butyllithium/sodium dodecylbenzenesulfonate initiator, which in

turn is more branched than the polymer produced by the dibutylmagnesium/

sodium t-amylate initiator. The same thing is seen in the root-mean-square

(rms) radius versus molar mass in that the polymer with the smaller radius at a

given molecular weight is more branched than a polymer with a larger radius

at that same molecular weight.



Copyright © 2004 by Taylor & Francis



Figure 25 Effect of vinyl content on solution polymers. (x) Tangent delta; (n)

Die C tear strength; (x) elongation; (n) tensile strength. (From Ref. 12.)



Table 14



Characterization and Stress Relaxation of SBR Prepared by Selected Initiators



Sample



Type



Mw



Mn



M w/

Mn



ML4



Tg

(jC)



Styrene

(%)



Vinyl

(%)



T80



n-BuLi/

SMTa

n-BuLi/

SDBSb

Bu2Mg/

SMTc



5.20E+05



2.42E+05



2.15



58



À14



26



53



0.033



3.29E+05



1.92E+05



1.71



54



À13



27



51



0.016



2.89E+05



1.82E+05



1.59



57



À12



26



58



0.013



4



n-BuLi/

SMT



3.78E+05



1.74E+05



2.17



70



À70



18



15



0.020



5



Dist. feed



3.15E+05



1.98E+05



1.59



68



À72



18



10



0.007



1

2

3



a



n-Butyllithium/sodium t-amylate.

n-Butyllithium/sodium dodecylbenzene sulfonate.

c

Dibutylmagnesium/sodium t-amylate.

Source: Ref. 6.

b



Copyright © 2004 by Taylor & Francis



Figure 26 Root-mean-square radius versus molar mass of solution SBRs (x) 1, (n)

2, and (E) 3. (From Ref. 6.)



Runs 4 and 5 (Fig. 27) illustrate the microstructural differences between

two low-vinyl SBRs. In run 4, the styrene is randomized by use of sodium tamylate/n-butyllithium, whereas in run 5 randomization takes place by controlling the monomer concentration by ‘‘distributing’’ the butadiene in such

a way that blocky styrene does not occur. No modifier is used in run 5. The T80

for the run with no modifier is only one-third that of the modified run, although



Figure 27 Root-mean square radius versus molar mass of solution SBRs (x) 4 and

(n) 5. (From Ref. 6.)



Copyright © 2004 by Taylor & Francis



the rest of the typical raw polymer properties are virtually identical. The

difference in branching is also seen in the rms radius versus molar mass curve.



VI.



ZIEGLER–NATTA POLYMERIZATION AND ZIEGLER–

NATTA POLYMERS



Ziegler–Natta catalysts are prepared from transition metal compounds

(halides, alcoholates, acetylacetonates, or long-chain carboxylic acid salts)

and an aluminum alkyl. Ziegler–Natta polymerizations are usually developed

for a specific polymer, and it is difficult to modify a specific catalyst to make

others. This is in contrast to anionic polymerization, where temperature,

modifier, and feed rate can be controlled to make a predictable variety of

polymers. Small changes in a Ziegler–Natta catalyst system can have major

unpredictable effects on polymer macrostructure and microstructure. Although there are general principles, much catalyst development work is still

empirical. The two most important general-purpose elastomers prepared by

Ziegler–Natta polymerization are high-cis polybutadiene and high-cis polyisoprene. The latter is discussed in another chapter in this volume.



A.



Mechanism of Butadiene Polymerization



Reaction of a transition metal salt with alkylaluminum in the presence of

butadiene can lead to a syn- or anti-k-allyl complex as shown in Figure 28. The

thermodynamically less stable anti form is the primary reaction product in

many systems (71–73) where a preformed p-allyl complex is reacted with a

diene substrate.

The k-allyl complex is in equilibrium with both the syn complex and

j complex, where the metal is directly bonded to one of the carbon atoms.

The j complex will coordinate with another butadiene molecule then a

new carbon–carbon bond will be formed to the carbon bonded to the

metal migrating to the terminus of the newly coordinated butadiene. The

microstructure of the polymer is set in this reaction. If the original

complex is anti, a cis double bond will be formed. If the original complex

is syn, a trans double bond will be formed. If the reaction of the anti form

with monomer is faster than the equilibrium between syn and anti, a

predominantly cis polymer will be formed. Ligands (anions from the

original transition metal salt, reaction products from alkylaluminum, or

added ligands) and solvents play a major role in determining the rate and

microstructure of the polymer by complexing with the metal and affecting

the various equilibria. This is illustrated in Table 15, where a preformed k-



Copyright © 2004 by Taylor & Francis



Figure 28 Mechanism of polymerization of butadiene with a Ziegler–Natta catalyst. (Adapted from Ref. 79, p. 93.)



allylnickel complex is used to polymerize butadiene. As the anion in the

complex varies from iodide to chloride, the rate of polymerization

decreases whereas the cis content increases from virtually none to 92%.

The solvent effect is shown with the trifluoroacetate ligand, where the cis

content is increased from 59% to 94% by changing the solvent from

toluene to heptane (74).



Copyright © 2004 by Taylor & Francis