E.M. Galimova, I.G. Akhmetov, V.N. Borisenko, R.R. Galimov, and A.G. Sakhabutdinov

Transcription

1 Kauchuk i Rezina, No. 3, 2014, pp. 4 7 The influence of the molecular weight and viscosity of solution-polymerised styrene butadiene rubber on the properties of rubber mixes and vulcanisates E.M. Galimova, I.G. Akhmetov, V.N. Borisenko, R.R. Galimov, and A.G. Sakhabutdinov OAO Nizhnekamskneftekhim, Nizhnekamsk Selected from International Polymer Science and Technology, 41, No. 12, 2014, reference KR 14/03/04; transl. serial no Translated by P. Curtis Synthetic styrene butadiene rubbers currently occupy second place worldwide in terms of volume of production, behind only natural rubber. Thus, according to McGraw and Petrovic [1], the capacities for the production of different types of rubber based on styrene and butadiene in 2013 amounted to over 6 million t. The main consumer of these products is the tyre industry. The constant increase in requirements concerning the safety of motor transport, including ecological requirements, has inevitably affected the tyre sector. In a number of leading countries of the world this has led to the development and adoption of directives concerning the marking of tyres, showing at a glance the main properties of the tyres, their wet road holding characteristics, fuel economy, and noise level. A certain improvement in the service properties of tyres has become possible through the use of tread vulcanisates based on solution-polymerised styrene butadiene rubbers (RBSK) with an increased content of vinyl units and a statistical distribution of styrene in the polymer chains [2]. Styrene butadiene rubbers with a structure of this kind are generally produced by the copolymerisation of the monomers in solution under the action of anionic initiators. Of special importance in this case is the composition of the initiating system, the conditions of the copolymerisation process, and the method of its implementation. It is known [3] that rubbers of similar composition and Mooney viscosity can have considerable differences in molecular structure. It was therefore of interest to study the influence of the molecular weights and viscosity of trial specimens of RBSK rubbers on the properties of rubber mixes and vulcanisates. In addition, the expediency of such an investigation was borne out by the absence of any systematic data on this question, the viscosity of the rubber mix for non-plasticising polymers, including RBSK rubbers, being determined by the molecular structure of the initial rubber [3]. The production of RBSK rubbers was carried out on a pilot unit simulating all stages of the industrial production of synthetic rubbers by continuous polymerisation in solution. The copolymerisation of styrene and butadiene was carried out in a hexane solvent, with n-butyllithium used as the initiator. The modifier of the initiator comprised a mixture of sodium and magnesium alcoholates and sodium magnesium alcoholate [4] and was supplied by the Voronezh Branch of the Scientific Research Institute for Synthetic Rubber as a commercial product under the trade name Lapramolat M-3 according to the TU specifications. The molecular weight of the produced rubber was controlled by changing the concentration of the comonomers in the reaction medium. IR spectra of the RBSK rubbers were recorded on a Spectrum 100 spectrometer in the wave number range cm 1. Quantitative analysis of the isomeric composition of the studied polymer specimens was based on measuring the optical densities of the analytical absorption bands at 724, 967, 910, and 1491 cm 1 for cis-1,4-, trans-1,4-, and 1,2-units and styrene respectively [5]. Data on the content of different isomeric structures were reproduced within limits not exceeding ±0.5%. The molecular weight characteristics were determined on a Breeze chromatograph on high-resolution styrogel columns with a range of measurement of weights from 2015 Smithers Information Ltd. T/7

2 to Tetrahydrofuran was used as the solvent at a temperature of 40 C. The eluent flow rate was 0.3 ml/min. Gel chromatography data were processed using a Waters applied program package. The glass transition temperature (T g ) of RBSK specimens was determined on a DSC 204 F 1 Phoenix differential scanning calorimeter in accordance with ASTM E The viscosity of the rubber specimens and rubber mixes based on them was determined on a 2000 E Mooney viscometer according to the GOST standard. Mixes were prepared to the standard formulation according to ISO 2322 (ASTM D3185) in two stages. At the first stage, mixing was carried out in a RheoMix 3000 OS mixer of the Polylab OS laboratory system at a charging temperature of 60 C and a rotor speed of 50 rpm. The mixing time was 6 min. Sulphur and accelerating agent were introduced at the second stage of mixing on a laboratory mill at a roll temperature of 40 ± 5 C. Investigations of the vulcanisation characteristics were conducted on an MDR-2000 instrument according to ASTM 5289 at 160 C for 30 min. The viscoelastic properties were determined on an RPA-2000 rheometer. The hysteresis losses and the mechanical loss tangent (tg δ) at a temperature of 60 C were assessed in accordance with ASTM D6601. In this work, the reduction in the storage modulus (G ) with increase in strain from 0.98 to 49.94% was adopted for assessment of the filler dispersion (the Payne effect). Rubber mixes were prepared for vulcanisation and vulcanisation was carried out according to ASTM D3185. The rubber mixes were vulcanised on a LAP-100 vulcanisation press at 145 C for 50 min. Specimens for physicomechanical tests were manufactured according to GOST The elastic strength properties were determined according to GOST The Shore A hardness was determined on a Zwick/Roell instrument according to GOST Rebound resilience tests of specimens were conducted on a Schob instrument according to GOST For the investigation, a series of RBSK rubbers with different Mooney viscosities was prepared. Table 1 gives the results of analysing the micro- and macrostructure of the investigated rubber specimens. It can be seen that the presented series covers a fairly wide range of molecular weights with a similar microstructure and a practically unchanged polydispersity factor. Figure 1 presents the relationship between the numberaverage (M n ), weight-average (M w ), and sedimentationaverage (M z ) molecular weights and the Mooney viscosity. The given data show a linear dependence of the viscosity of the rubber on the molecular weights (MWs). A proportional dependence of Mooney viscosity Table 1. The characteristics of the RBSK specimens Specimens Processing properties a M t (nominal units) Microstructure (content of units, wt%) Styrene , ,4-cis ,4-trans MW b M n M w M z M w /M n Fractional composition (%) a M t Mooney viscosity (100 C) b M n, M w, M z number-average, weight-average, and sedimentation-average molecular weights T/8 International Polymer Science and Technology, Vol. 42, No. 3, 2015

3 Figure 1. The relationship between the number-average (1), weight-average (2), and sedimentation-average (3) molecular weights and the Mooney viscosity of specimens of RBSK rubber on M n and M w in the region of low molecular weights is typical of many polymers [6]. At the same time, once a certain molecular weight is reached, a fluctuation network arises in the polymer, formed by nodes of entanglements, and subsequent increase in the Mooney viscosity is connected with increase in the number of mobile nodes or linkages per macromolecule [7]. These effects can be seen particularly clearly with increase in M z. Therefore, the linear nature of the dependence of Mooney viscosity on the sedimentation-average molecular weight indicates that in the studied range of molecular weights the macromolecules have mainly a linear structure, while the contribution of the fluctuation network to the properties of the elastomer is minimal. The glass transition temperature of the investigated specimens lies in the range from 14 to 16 C, which is in fairly good agreement with the empirical equation of the dependence of the glass transition temperature on the content of styrene and vinyl units [8]. RBSK specimens were then tested in the composition of rubber mixes. Tests of the standard formulation showed a consistent change in the processing and physicomechanical properties of rubber mixes as a function of the molecular weight and accordingly the Mooney viscosity of the investigated rubber specimens (Table 2). The viscosity of the rubber mixes increases with an increase in the viscosity of the rubber. In respect of the influence of the latter on the vulcanisation properties of the rubber mixes, the following must be pointed out. The minimum torque (M L ) of the rubber mix increases with increase in the viscosity of the rubber. A similar Table 2. The properties of rubber mixes and vulcanisates based on the RBSK specimens Specimens Processing properties a M t (nominal units) Rheometric properties b (MDR 2000, 160 C, 30 min) M L (dn m) M H (dn m) t s1 (min) t 50 (min) t 90 (min) High-elastic properties c G (kpa) tg δ Properties of vulcanisates d (145 ± 1 C, 50 min) f 300 (MPa) f t (MPa) e b (%) H A (nominal units) E (%) a M t Mooney viscosity (100 C) b M L, M H minimum and maximum torques; t s1 scorch start time; t 50, t 90 time of achievement of 50 and 90% degrees of vulcanisation respectively c tg δ mechanical loss tangent at 60 C d f 300 nominal stress under 300% elongation; f t nominal tensile strength; e b breaking elongation; H A Shore A hardness; E rebound resilience at 20 C 2015 Smithers Information Ltd. T/9

4 dependence is observed for values of the maximum torque, which characterises the degree of vulcanisation of the rubber mix. Of note is the absence of any influence of viscosity either on the duration of the induction period or on the time of achievement of 50 and 90% degrees of vulcanisation. The reduction in the degree of dispersion of carbon black in the rubber can be judged from the increase in the storage (real part) of the complex modulus (G ) with increase in the amplitude of dynamic strain, and here, for specimens with a Mooney viscosity of 52.1 nominal units and above, the increase in the Payne effect is more pronounced. It is obvious that, for linear, non-plasticising RBSK rubbers of high viscosity, more time is needed for better filler dispersion [9]. As the molecular structure of the initial rubber in many ways determines the behaviour during processing, increase in the molecular weight has an influence on the increase in viscosity and on the reduction in quality of mixing of the obtained mixes. To assess the elastic hysteresis properties of tread vulcanisates, wide use is made of the mechanical loss tangent, which comprises the ratio of the loss modulus G to the real component of the complex modulus G. In a study of the dependences of the output characteristics of tread vulcanisates, which influence the main service properties of the tyres, it was established that the mechanical loss tangent, determined at C, characterises the rolling losses. Reduction in tg δ at 60 C indicates a reduction in rolling losses [10]. The influence of the molecular weight on the mechanical loss tangent is presented graphically in Figure 2a. According to the obtained results, an increase in M n is accompanied with a linear reduction in tg δ. In the opinion of Pichugin [11], the reduction in tg δ is connected with reduction in the number of free ends of the polymer chains on account of increase in the number-average molecular weight. It is known that for polydisperse polymers the mechanical properties of vulcanisates are a function of the number-average molecular weight. In the region of low molecular weights, the strength and elasticity of vulcanisates initially increase, and then the rate of change in properties slows down, and, on achievement of a certain limiting (critical) value of the molecular weight (M cr ), subsequently they hardly change [6]. The magnitude of M cr depends primarily on the nature of the rubber. Furthermore, with a fixed microstructure of the polymer, M cr is determined by the linearity of the polymer chains of the rubber. For example, for SKB, M cr amounts to , and for linear SKD-L produced over organolithium catalysts the critical molecular weight is much lower and roughly equal to [12]. For emulsion-polymerised styrene butadiene rubbers, M cr amounts to [13]. Figures 2b and c give the dependences of the tensile strength and rebound resilience of filled vulcanisates based on the studied specimens of styrene butadiene rubber on the number-average molecular weight. With M n values of over , which corresponds to specimens of rubber with a Mooney viscosity of 57.0 nominal units and above, the molecular weight tensile strength curve plateaus out. The rebound resilience increases steadily in the entire MW range studied. The strength properties of vulcanisates are adversely affected by the presence in the rubber of low-molecularweight fractions [13, 14], which, for solution-polymerised styrene butadiene rubber, include fractions with MW Reduction in the low-molecular-weight fraction in the rubber, lowering the active proportion of vulcanisation network on account of increase in the content of passive material, leads to a sharp increase in physicomechanical properties. Here there is an improvement in the elastic and hysteresis properties of vulcanisates, and also an increase in the stresses under tensile strains [14]. In fact, for specimens 1 to 3, the proportion of macromolecules with a molecular weight of over 10 6 amounted to 3 5% (Table 1), and the content of fraction with MW 10 5 to Figure 2. The dependence of the mechanical loss tangent tg δ at 60 C (a), the tensile strength (b), and the rebound resilience at 20 C (E) (c) of filled vulcanisates based on RBSK rubber on the number-average molecular weight of the initial rubber M n T/10 International Polymer Science and Technology, Vol. 42, No. 3, 2015

5 30%. For specimens with a Mooney viscosity of 57.0 nominal units and higher, the proportion of fraction with MW 10 6 was practically 3 times higher, which was reflected by a sharp increase in strength properties. Thus, the nominal tensile strength and nominal stress under 300% elongation amounted to ~20.0 and ~17.0 MPa respectively. The use of RBSK rubbers with an increased content of vinyl units in the formulation of tread mixes makes it possible to considerably reduce the rolling resistance of the vulcanisates with a simultaneous improvement in wet road holding and retention of wear resistance [2, 8]. The introduction of carbon black into a rubber mix based on RBSK rubber gives the vulcanisate better wet road holding [2, 8]. In Kuperman [15], a very simple method was proposed for assessing the grip of vulcanisates in terms of their hardness and elasticity that does not require the use of any special equipment. Approximate assessment of wet road holding by means of the calculated empirical equation showed that, with increase in M n within the investigated range of molecular weights of the polymer, the friction coefficient decreases from 0.72 to 0.69, which corresponds to the level of experimental values of the given index for vulcanisates based on RBSK rubber [8, 15]. For indirect estimation of the wear resistance, most convenient is the Ratner Mel nikova equation, which links the abrasion of the vulcanisates with their tensile strength and rebound resilience. For the investigated specimens of rubber with a Mooney viscosity of over 57.0 nominal units, the calculated values of abrasion lie in the range cm 3 /m 10 3, which is similar to experimental data of vulcanisates based on RBSK rubber on a Schopper Schlobach instrument according to GOST [8]. Thus, for solution-polymerised styrene butadiene rubber with a mainly linear structure, increase in the molecular weight and Mooney viscosity leads to a proportional reduction in the mechanical loss tangent and the friction coefficient against wet asphalt, and to an increase in the elastic strength properties of the vulcanisates. A Mooney viscosity of RBSK rubber of nominal units is optimal for the achievement of the necessary level of service and strength properties of carbon-black-filled vulcanisates, and makes it possible to design for further improvement in the properties of vulcanisates when highly disperse silica fillers are used. ACKNOWLEDGEMENTS The authors are grateful to the team of the Research Laboratory of Instrumental Methods of Analysis of the Nizhnekamskneftekhim Scientific and Technical Centre for kindly providing the results of measurements by IR spectroscopy and gel permeation chromatography. REFERENCES 1. McGraw J.L. and Petrovic R.B., World Rubber Statistics IISRP, Houston, TX, 91 pp. (2013). 2. Kuperman F.E., New Rubbers for Tyres. Solutionpolymerised Rubbers with an Increased Content of Vinyl Units as Alternatives to Emulsionpolymerised Styrene Butadiene Rubber. Transpolymers and Copolymers of Isoprene and Butadiene. NIIShP, Moscow, 367 pp. (2011). 3. Garmonov I.V., Synthetic Rubber. Khimiya, Leningrad, 752 pp. (1976). 4. Russian Patent , MKI C 08 F 36/06, C 08 F 236/10, C 08 F 4/ Klauzen N.A. and Semenova L.P., Atlas of Infrared Spectra of Rubbers and Certain Ingredients of Rubber Mixes. Khimiya, Moscow, 40 pp. (1965). 6. Grechanovskii V.A., Synthetic Rubber, ed. by Garmonov I.V. Khimiya, Leningrad, 560 pp. (1983). 7. Kulezne V.N. and Shershnev V.A., Polymer Physics and Chemistry. Vysshaya Shkola, Moscow, 313 pp. (1988). 8. Kuperman F.E., New Rubbers for Tyres. Priority Requirements. Methods of Assessment. NIIShP, Moscow, 329 pp. (2005). 9. L vova T.M. et al., Abstracts of Papers of 10th Russian Scientific and Practical Conference The Rubber Industry. Feedstock. Materials. Technology, Moscow, p. 42 (2003). 10. Pichugin A.M. et al., The relationship between the temperature dependence of the mechanical loss tangent and the output characteristics of tread vulcanisates of different composition. Kauch. i Rezina, (4):2 (2008). 11. Pichugin A.M., Materials Science Aspects of the Development of Tyre Rubbers. Mashinostroenie, Moscow, 383 pp. (2008). 12. Kuperman F.E. et al., The properties of polybutadiene rubber produced by mixing highmolecular-weight and low-molecular-weight cispolybutadienes. Kauch. i Rezina, (8):3 (1971). 13. Poddubnyi I.Ya. et al., The influence of molecular weight on certain physicomechanical properties of vulcanisates. Kauch. i Rezina, (2):6 (1958). 14. Krol V.A. et al., The relationship between the properties of filled vulcanisates and the structure of the network of unfilled vulcanisates based on linear polybutadienes. Kauch. i Rezina, (8):3 (1973) Smithers Information Ltd. T/11

6 15. Kuperman F.E., Assessment of the holding properties and wear resistance of tread vulcanisates in terms of hardness, elasticity, and strength. Promyshl. Proizvod. Ispol z. Elast., (4):20 (2001). Received T/12 International Polymer Science and Technology, Vol. 42, No. 3, 2015