Method for preparing rubber formulations using silanized silica nanofiller.
Previous studies have shown that precipitated silica nanofiller and TESPT coupling agent improved properties such as hardness, tear strength, tensile strength and cyclic fatigue life (ref. 2). However, the surface of precipitated silica contains silanol or hydroxyl groups, which make the filler polar and moisture adsorbing (ref. 3). This causes long cure times and slow cure rates in sulfur-cured rubber compounds. To remedy these deficiencies, bifunctional organosilanes are added with the filler. Precipitated silica is often treated with bis(3-triethoxysilylpropyl)-tetrasulfide (TESPT), known also as Si69 coupling agent. This silane chemically adheres silica to rubber and also prevents silica from interfering with the reaction mechanism of sulfur cure in rubber (ref. 4). In addition, the tetrasulfane groups are rubber reactive and react in the presence of accelerator at elevated temperatures, i.e., 140 - 260[degrees]C, without elemental sulfur being present, to form crosslinks in unsaturated rubbers, for instance SBR, NR and BR. The ethoxy groups react with the hydroxyl groups on the surfaces of these fillers and this leads to the formation of stable covalent filler/ rubber bonds via TESPT.
The aim of this study was to use precipitated silica nanofiller pre-treated with TESPT to reinforce the mechanical properties of SBR, NR and BR rubbers, and at the same time, to address the major issues of health and safety in the workplace related to the excessive use of rubber chemicals. Since many rubber chemicals are hazardous, a reduction in their use in rubber compounds will be desirable. The specific aim of this work was to reduce the use of these chemicals in rubber compounds without compromising good mechanical properties of the rubber vulcanizates, which are essential for long durability and life in service.
Materials and mixing
The raw rubbers used were standard Malaysian natural rubber grade L (SMR-L), styrene-butadiene rubber (SBR: Intol 1712, Enichem, oil-extended, styrene 23.5% wt.%) and high-cis polybutadiene (BR: Buna CB 24, Lanxess, not oil-extended, containing 98% cis-1,4 content). The reinforcing nanofiller was Coupsil 8113 supplied by Degussa AG. Coupsil 8113 is precipitated amorphous white silica (type Ultrasil VN3), surfaces of which had been pre-treated with TESPT bifunctional organosilane. The filler has 11.3% by weight TESPT silane, 2.5% by weight sulfur (included in TESPT), 175 [m.sup.2]/g surface area (measured by [N.sub.2] adsorption), and 20-54 nm particle size.
In addition to the raw rubbers and filler, the other additives were N-t-butyl-2-benzothiazole sulphenamide (Santocure TBBS, a safe-processing delayed action accelerator), N-cyclohexyl-2-benzothiazole sulphenamide (Santocure CBS, accelerator), zinc oxide (activator), N-(1,3-dimethylbutyl)-N'-phenyl-p-phenylenediamine (Santoflex 13, antidegradant) and heavy paraffinic distillate solvent extract aromatic processing oil (Enerflex 74).
Mixing was carried out in a Haake Rheocord 90, a small size laboratory mixer with counter-rotating rotors. The rotors and the mixing chamber were maintained at ambient temperature (-23[degrees]C) and the rotor speed was 45 rpm. The volume of the mixing chamber was 78 [cm.sup.3] and it was half full. Before mixing started, the ram was raised to introduce the filler into the mixing chamber, and then the raw rubber was added. The ram was lowered to keep the rubber in the mixing chamber during mixing. The mixing time was increased to 22 minutes in order to measure the time needed to disperse the silica particles fully in the rubbers. The Haake software, Version 1.9.1., was used for controlling the mixing condition and storing data. The temperature of the rubber compounds during mixing was 50-79[degrees]C. Twenty-four hours after mixing ended, the rubbers were examined in a scanning electron microscope to assess the filler dispersion.
Assessement of the silica particles dispersion in the rubber
Dispersion of the silica particles in the rubber was assessed by a LEO 1530 VP field emission gun scanning electron microscope (SEM). The degree of dispersion of the silica particles in the rubber was subsequently studied from some SEM photographs. After the SEM photos were examined, a suitable mixing time was selected for adding the filler and curing chemicals to the rubbers.
Selection of TBBS and CBS
To activate the rubber-reactive tetrasulfane groups of TESPT, TBBS and CBS were added. The loadings of TBBS in SBR and BR rubbers, and CBS in NR, were increased progressively to 11 parts per hundred rubber by weight (phr) and 9.6 phr, respectively, to measure the amounts needed to optimize the chemical bonding between the rubber and filler, and to increase the crosslink density in the rubbers. The formation of covalent sulfur bonds or crosslinks between the rubber and TESPT strengthened the rubber/filler interaction (ref. 4). In total, 52 compounds were prepared.
Selection of zinc oxide and stearic acid
The loading of zinc oxide in the SBR and BR rubbers filled with silica and TBBS was increased to 2.5 phr, and in the NR rubbers filled with 10, 30 and 60 phr silica, to 6 phr, respectively, in order to determine the amounts needed to maximize the efficiency of TBBS, CBS and cure. In total, 32 compounds were made.
To improve the efficiency of cure in the SBR and BR rubbers filled with silica, TBBS and zinc oxide, up to 2.5 phr stearic acid was also added. In total, 13 compounds were prepared. For further details on the SBR rubber see (ref. 5).
Finally, five rubber compounds were mixed for this study (table 1). After mixing ended, the rubber was recovered from the mixer and milled to a thickness of about 6 mm. The compounds were kept at 23[degrees]C for at least 24 hours before their cure properties were measured.
Cure properties of the rubber compounds
The scorch time and the optimum cure time were determined from cure traces generated at 140 [+ or -] 2[degrees] C by an oscillating disc rheometer curemeter (ODR) at an angular displacement of [+ or -] 3[degrees] and a test frequency of 1.7 Hz (ref. 6). From the cure traces, the Atorque, which is the difference between the maximum and minimum torque values, was calculated (figure 1). The cure rate index was calculated using the method described previously (ref. 7). The rheometer tests ran for up to two hours. Results from these experiments are summarized in table 1.
[FIGURE 1 OMITTED]
Test pieces and test procedure
After the ODR measurements were completed, the compounds were cured in a compression mold at 140[degrees]C with a pressure of 11 MPa. For measuring the mechanical properties of the rubbers, sheets 23 cm by 23 cm by approximately 2.8 mm thick were used, from which various samples for further tests were cut. Mechanical property results are listed in table 1.
For determining the hardness of the rubber, cylindrical samples 12.5 mm thick and 28 mm in diameter were cured. The samples were then placed in a durometer A hardness tester, and the hardness of the rubber was determined at 25[degrees]C after a 15 second interval. This was repeated at three different positions on the sample, and the median of the three readings was subsequently indicated (ref. 8).
Trouser tear tests were performed at an angle of 180[degrees], at 23[degrees]C and at a constant crosshead speed of 100 mm/min. (ref. 9) in a Lloyd mechanical testing machine. The tears produced varied in length from approximately 22 mm to 75 mm. In each experiment, the tearing force was recorded on a chart to produce a trace from which an average force was calculated (figure 2). Five test pieces were fractured and the tearing energies were calculated from equation (1) (ref. 10):
T = 2F/t (1)
where: F is the force, and t the thickness of the test piece. The median values of the tearing energies were subsequently shown in table 1.
The tensile stress, strain at break and stored energy density at break of the vulcanizates were determined in uniaxial tension in a Lloyd mechanical testing machine using dumbbell test-pieces 3.6 mm wide, with a central neck 25 mm long. These samples were die-stamped from sheets of the cured rubber. The tests were performed at 23[degrees]C and at a crosshead speed of 100 mm/min. (ref. 11). Lloyd DAPMAT computer software was used for storing and processing the data.
For determining the abrasion resistance of the rubbers, molded cylindrical test pieces, 8 mm thick and 16 mm in diameter, were cured. The tests were performed at 23[degrees]C in accordance with BS 903: Part A9 (Zwick abrasion tester 6102 and abrasion standard rubber S1) (ref. 12). For each rubber, three samples were tested to calculate the relative volume loss, [DELTA]v (table 1).
Cyclic fatigue life
The cyclic fatigue life of the rubbers was measured in uniaxial tension in a Hampden dynamic testing machine, using dumbbell test pieces. The test pieces were die-stamped from the sheets of vulcanized rubber. The tests were carried out at constant maximum strain amplitude of 100% (the central neck was stretched to 50 mm), and a test frequency of 1.4 Hz. The test temperature was 22[degrees]C, and the strain on each test piece was relaxed to zero at the end of each cycle. For each rubber, eight test pieces were cycled to failure and median values of the results were recorded (ref. 13). When the number of cycles exceeded 1,000 kc, the test was stopped (table 1).
Results and discussion
Filler dispersion in the rubber
After the SEM photographs were examined (figures 3 and 4), it was evident that the filler dispersion was largely affected by the mixing time. For SBR, a total mixing time of 10 minutes, and for BR, 16 minutes were sufficient to fully disperse the silica particles in the rubber matrix. For NR with 10 phr and 30 phr silica, 11 minutes, and for NR with 60 phr silica, 18 minutes were needed to fully disperse the silica particles in the rubber matrix. Full details of the mixing conditions of the NR rubbers containing 10, 30 and 60 phr silica were described previously (refs. 14-16).
[FIGURES 2-4 OMITTED]
Effect of TBBS, zinc oxide and stearic acid on the cure of the filled SBR and BR rubbers Figure 5 shows the [DELTA]torque versus TBBS loading. The [DELTA]torque is an indication of crosslink density changes in the rubber. For the SBR, the [DELTA]torque increased to 22 dNm as the loading of TBBS was raised to 3 phr. Further increase in TBBS had little or no benefit for the [DELTA]torque, which remained at 26 dNm. For the BR, the Atorque rose sharply to 87 dNm with 7.5 phr TBBS, and the increase slowed down substantially to about 94 dNm when the full loading of TBBS was added to the rubber. Evidently, 3 phr and 7.5 phr TBBS were sufficient to fully optimize the chemical bonding between the filler and SBR and BR rubbers, respectively. When zinc oxide was added to the filled SBR with 3 phr TBBS, there was a noticeable improvement in the [DELTA]torque (figure 6). It rose to 56 dNm when 0.5 phr zinc oxide was added, and it continued rising to 64 dNm when 2.5 phr zinc oxide was incorporated in the rubber. For the BR, the rise was even more significant. The [DELTA]torque rose to 130 dNm with 0.5 phr zinc oxide, and it remained unchanged when the loading of zinc oxide was raised to 1.5 phr.
[FIGURE 6 OMITTED]
Interestingly, when stearic acid was added to the filled SBR rubber with 3 phr TBBS and 0.5 phr zinc oxide (figure 7), the [DELTA]torque decreased as the loading of stearic acid was raised to 2.5 phr. Similarly for the filled BR rubber containing 7.5 phr TBBS and 0.5 phr zinc oxide, the addition of stearic acid up to 1 phr had little benefit, and in fact it was detrimental when the loading of stearic acid was increased to 2.5 phr.
[FIGURE 5 & 7 OMITTED]
Figure 8 shows the [DELTA]torque versus CBS loading for the filled NR. The rubber with 10 phr silica required 7 phr CBS to optimize the chemical bonding between the rubber and filler. However, as the loading of silica was raised to 30 phr and then to 60 phr, the amount of CBS needed to optimize the chemical bonding decreased to 4.4 phr and 4 phr, respectively. The inclusion of zinc oxide in the NR compounds was largely beneficial to the filler/rubber interaction. The [DELTA]torque increased to its maximum value when the amount of zinc oxide in the rubber was raised to 1 phr (figure 9). It was clear from figure 9 that 1 phr zinc oxide was sufficient to improve the efficiency of CBS and optimize to a greater extent the chemical bonding between the filler and rubber.
[FIGURES 8-9 OMITTED]
Effect of silica on the mechanical properties of the cured SBR and BR rubbers
The results summarized in table 1 show a substantial improvement in the mechanical properties of the rubber vulcanizates. The hardness of the SBR rubber was 62 durometer A, tensile strength 26 MPa and elongation at break 1,308%. The properties associated with fracture were also enhanced. The stored energy density at break was 140 MJ/[m.sup.3], and the rubber had a tearing energy of 75 kJ/[m.sup.2]. The abrasion resistance, which was measured by the relative volume loss, was 127 m[m.sup.3]/mg. Probably the most interesting results were for the cyclic fatigue life. The minimum fatigue life was 777.4 kc and seven samples lasted longer than 1,000 kc. For the BR containing the same loading of the filler, the hardness was 72 durometer A, which was noticeably higher than the SBR. However, the tensile properties were poorer. The tensile strength was 17 MPa, and elongation at break 606%. The fracture properties were also inferior to the SBR. The stored energy density was 49 MJ/[m.sup.3] and the tearing energy about 30 kJ/[m.sup.2]. The cyclic fatigue life was considerably shorter than SBR with a minimum fatigue life of 40 kc and four samples lasting longer than 1,000 kc. Probably the most interesting result for the BR was the substantial improvement observed in its abrasion resistance. The relative volume loss was 15.5 [mm.sup.3]/mg, which was eight times better than the SBR.
The results indicated that a substantial improvement in the mechanical properties of the rubbers was achieved with fewer and smaller amounts of accelerators and activators, which are normally used in carbon black filled industrial rubber compounds (ref. 1).
Effect of silica on the mechanical properties of the cured NR As the loading of silica in NR was increased progressively from 10 phr to 60 phr, the mechanical properties were affected in different ways. The hardness increased from 28 to 71 (table 1). Similarly, the tensile strength was improved from 18 MPa to 38 MPa as the loading of silica was increased to 30 phr; and then it decreased to 33 MPa when the full loading of the filler was reached, i.e., 60 phr. Elongation at break also decreased from 1,486% to 922% as a function of silica loading. The largest improvement was recorded for the fracture properties of the rubber. Stored energy density at break rose from 70 MJ/[m.sup.3] to 135 MJ/[m.sup.3] and tearing energy from 15 kJ/[m.sup.2] to 61 kJ/[m.sup.2.] It was not possible to make an accurate assessment of the effect of silica on the abrasion resistance of the rubber, since there were no data available for compounds 3 and 5. The cyclic fatigue life seemed to have deteriorated when 30 phr silica was added, and there was no evidence to suggest that adding more filler had any noticeable effect on the fatigue life. Therefore, the exact benefit of silica for this property remained unclear.
The findings suggested that for NR, some properties gained benefits and some did not from the increase in the loading of silica in the rubber.
From this study, it is concluded that:
* At 60 phr loading of silica, 3 phr TBBS for SBR and 7.5 phr TBBS for BR were needed to optimize the rubber/filler interaction. However, only 0.5 phr zinc oxide was needed to optimize the efficiency of TBBS in the rubber.
* The addition of stearic acid to the SBR rubber filled with silica, 3 phr TBBS and 0.5 phr zinc, and the BR rubber filled with silica, 7.5 phr TBBS and 0.5 phr zinc oxide, offered no additional benefit to the rubber/filler interaction, and in fact it was detrimental to it.
* For NR, the amount of CBS needed to optimize the rubber/filler interaction decreased from 7 phr to 4 phr as the loading of the filler was increased from 10 phr to 60 phr. However, only 1 phr zinc oxide was needed to optimize the efficiency of CBS in the rubber.
Two general rules for crosslinking and reinforcing SBR, BR and NR rubber compounds with silanized silica nanofiller can be made from this study:
For SBR and BR
For a given loading of silica, the requirement for TBBS depends on the composition of the rubber. However, the amount of zinc oxide needed to optimize the efficiency of TBBS is independent of the composition of the rubber and the loading of TBBS.
For a given rubber, the requirement for CBS depends on the loading of silica, but the requirement for zinc oxide is independent of the loading of silica and CBS.
It is clear from the results that this new method for preparing a rubber formulation helps to substantially reduce the use of curing agents in rubber compounds without compromising the mechanical properties of the rubber vulcanizates, which are essential in maintaining long life, good performance and durability in service. This will help to improve health and safety in the workplace. A reduction in the use of rubber curing chemicals will also reduce costs.
This article is based on a paper presented at RubberChem 2006, a Rapra Technology conference. (www.rapra.net/conferences)
(1.) "Natural rubber formulary and properties index," Re: EUR053, Archives, Tun Abdul Razak Research Centre, MRPRA, Brickendonbury, Hertford, UK SG13 8NL.
(2.) A. Ansarifar, R. Nijhawan, T. Nanapoolsin and M. Song, "Reinforcement effect of silica and silane fillers on the properties of some natural rubber vulcanizates," Rubber Chem. Technol., 76 (5), 1,290-1,310, 2003.
(3.) S. Wolff U. Gorl, M-J. Wang and W. Wolff, "Silane modified silicas--silica-based tread compounds," Eur. Rubber J., 16, 16-19, 1994.
(4.) S. Wolff "Chemical aspects of rubber reinforcement by fillers," Rubber Chem. Technol., 69, 325-346, 1996.
(5.) A. Ansarifar, Li Wang, R.J. Ellis and S.P. Kirtley, "The reinforcement and crosslinking of styrene butadiene rubber with silanized precipitated silica nanofiller," Rubber Chem. Technol., 79 (1), 39-54, 2006.
(6.) British Standard 12673: Part 10 (1977), "Methods of tests for raw rubber and unvulcanized compounded rubber. Measurement of prevulcanizing and curing characteristics by means of curemeter."
(7.) British Standard 903: Part A60: Section 60.1 (1996), "Methods of tests for raw rubber and unvulcanized uncompounded rubber: Measurement of prevulcanizing and curing characteristics by means of curemeter."
(8.) British Standard 903: Part A26 (1995), "Physical testing of rubber: Meihod for determination of hardness."
(9.) British Standard 903: Part A3 (1995), "Physical testing of rubber: Method for determination of tear strength--trousers, angle and crescent test pieces."
(10.) H.V. Greensmith and A.G. Thomas, "Rupture of rubber. III. Determination of tear properties, " J. Polym. Sci., 43, 189200, 1955.
(11.) British Standard 903: Part A2 (1995), "Physical testing of rubber: Method for determination of tensile stress strain properties."
(12.) British Standard 903: Part A9, "Method A. 1 (rotating test piece)", 1995.
(13.) British Standard 903: Part A51 (1986), "Methods of testing vulcanized rubber. Determination of resistance to tension fatigue."
(14.) A. Ansarifar, A. Azhar and M. Song, "A new design concept for natural rubber compounds using silanized precipitated silica," J. Rub. Res., 6 (3), 129-152, 2003.
(15.) A. Ansarifar, A. Azhar, N. Ibrahim, S.F. Shiah and J.M.D. Lawton, "The use of silanized silica filler to reinforce and crosslink natural rubber," Int. J. Adhesion & Adhesives, 25 (1), 77-86, 2005.
(16.) A. Ansarifar, S.F. Shiah and M. Bennett, "Optimizing the chemical bonding between silanized silica nanofiller and natural rubber and assessing its effects on the properties of the rubber," Int. J. Adhesion & Adhesives, 26 (6), 454-463, 2006.
A. Ansarifar and L. Wang, Loughborough University, and R.J. Ellis and S.P. Kirtley, Avon Automotive (Email: M.A.Ansarifar@lboro.ac.uk)
Table 1--recipes and results for the rubber compounds Formulation (phr) Compound no. 1 2 3 SBR 100 -- -- BR -- 100 -- NR -- -- 100 Silanized silica 60 60 10 TBBS 3 7.5 -- CBS -- -- 7 Zinc oxide 0.5 -- 1 Processing oil 5 -- -- Santoflex 13 1 1 1 ODR results (at 140[degrees]C) Minimum torque (dNm) 18 37 13 Maximum torque (dNm) 56 129 43 [DELTA]torque (dNm) 38 92 30 Scorch time, [t.sub.s2] (min.) 16 8 97 Optimum cure time, [t.sub.95](min.) 80 83 156 Cure rate index ([min..sup.-1]) 1.6 1.3 1.7 Mechanical properties (cured) Hardness (durometer A) 62 72 28 Tensile strength (MPa) 26 17 18 Elongation at break (%) 1,308 606 1,486 Stored energy density at 140 49 70 break (MJ/[m.sup.3]) Tearing energy (kJ/[m.sup.2]) 75 30 15 Relative volume loss in the 127 15.5 ** abrasion tests, [DELTA]v ([mm.sup.3]/mg) Cyclic fatigue life (kc) 777- 40- 92- >1,000 >1,000 142 Compound no. 4 5 SBR -- -- BR -- -- NR 100 100 Silanized silica 30 60 TBBS -- -- CBS 4.4 4 Zinc oxide 1 1 Processing oil -- -- Santoflex 13 1 1 ODR results (at 140[degrees]C) Minimum torque (dNm) 23 27 Maximum torque (dNm) 75 103 [DELTA]torque (dNm) 52 76 Scorch time, [t.sub.s2] (min.) 24 8 Optimum cure time, [t.sub.95](min.) 52 29 Cure rate index ([min..sup.-1]) 3.6 4.8 Mechanical properties (cured) Hardness (durometer A) 52 71 Tensile strength (MPa) 38 33 Elongation at break (%) 1,075 922 Stored energy density at 149 135 break (MJ/[m.sup.3]) Tearing energy (kJ/[m.sup.2]) 48 61 Relative volume loss in the 257 ** abrasion tests, [DELTA]v ([mm.sup.3]/mg) Cyclic fatigue life (kc) 59- 77- 78 121 ** No data available for these rubber compounds.
|Printer friendly Cite/link Email Feedback|
|Date:||Apr 1, 2007|
|Previous Article:||Mechanical properties of rubbers reinforced with porcelainite-derived precipitated silica.|
|Next Article:||Oil swell and low temperature properties of AEM and impact on seal performance.|