Polymerized soybean oils of different molecular weights were used as plasticizers in NR/SBR compositions. The oils of different molecular weights and viscosities were synthesized by cationic polymerization using a proprietary technology. Because vegetable oils have double bonds, they are not only viscosity depressants but also active participants in cross-linking reactions. Properties of elastomers extended with different concentrations of mineral oil or pure soybean oil were compared with elastomers extended by polymerized oils of different molecular weights at the same concentrations. It was found that polymerized soybean oil could be substituted for naphthenic process oil with minimal differences in mechanical and dynamic properties.
Plasticizers play some important roles in formulating rubber. They can be used to improve processing, tackiness, dispersion, and flow of the unvulcanized compound. They can lower the modulus or improve the low temperature performance of the vulcanized compound, or they can be used simply to lower the cost of a rubber compound by extending the polymer with less expensive oil and filler. Primary plasticizers solubilize the rubber and assist in Brownian motion of the polymer chains, thus reducing the viscosity of the compound. Secondary plasticizers do not solubilize the elastomer, but they can act as lubricants between the polymer chains to improve flow. In choosing a plasticizer, it is important not to exceed the level of compatibility of the oil with the elastomer. Primary plasticizers (those that swell the polymer) can be used in large quantities, whereas secondary plasticizers are used in much more limited quantities. The choice of a primary plasticizer must be matched to the polymer; that is, nonpolar plasticizers generally find use in nonpolar elastomers, whereas polar elastomers use polar plasticizers.
Plasticizers can be generally classified as mineral oils (petroleum based), synthetic plasticizers, or vegetable oils and other natural products. Mineral oils are further classified by their content of aromatic, paraffinic, and naphthenic hydrocarbons. All mineral oils are nonpolar and relatively cheap, so they are widely used in nonpolar general-purpose elastomers such as NR, polyisoprene (IR), polybutadiene (BR), styrene butadiene (SBR), butyl (IIR), and ethylene propylene (EPR or EPDM). In small quantities, perhaps 5 to 20 parts per hundred rubber (phr), mineral oils are used to improve flow and processing. Quantities of 10 to 40 phr are often used to soften the rubber and reduce the hardness and modulus. To extend the rubber and reduce its cost, quantities of oil from 40 or 50 phr to more than 100 phr may be used.
One major drawback with the petroleum-based oils is that they are a nonrenewable resource. In some cases, the aromatic content of certain mineral oils may label them as carcinogenic. Replacing mineral oil with vegetable oils may be advantageous as a solution to these problems. The current price of soybean oil (SBO) is about 50 cents/lb (i.e., less than that of mineral oil).
Vulcanized vegetable oils or factices have a long history of use in the rubber industry. Factices are unsaturated vegetable oils that have been cross-linked, typically with sulfur, isocyanates, or peroxides. Factice can be used to extend rubber, and although it is a valuable processing aid, it is not a plasticizer, strictly speaking. Factice can improve the green strength and dimensional stability of extrusions, improve the processability of very-low-durometer compounds, and increase the plasticizer absorption.
Pure vegetable oils have been used as plasticizers in rubbers with different degrees of success.1–5 U.S. patent 7,335,692 reveals the use of vegetable oils for tire tread to improve resistance to separation of the plies.6 The oils used in SBR and BR were synthetic triolein (triester of oleic acid and glycerin), rapeseed oil, high oleic sunflower oil, and normal sunflower oil, typically at 30 phr in a mixture with mineral aromatic oil (6–30 phr). Soluble cyclohexane fraction with vegetable oil compositions was lower than that with mineral oil, indicting that possible grafting of oil to rubber was taking place. Coconut oil was used as a plasticizer for NR in concentrations from 2 to 10 phr. Although the cure time was marginally lower, tensile strength, tear strength, and resilience were better that the naphthenic oil–based compounds.7 When castor oil was used as a plasticizer for NR, the tear strength and modulus were improved, whereas other mechanical properties were comparable with those of the mixes containing naphthenic oil.8 Linseed oil was studied as a plasticizer in nitrile rubber9 and rubber seed oil in NR and SBR.10 Ten different vegetable oils were used as plasticizers for NR at 8 phr.11 Properties varied in a narrow range to draw general conclusions. Generally, the amount of vegetable oils used in references 7 to 11 was rather low—up to 10 phr. In some cases, antiplasticization manifested in hardness increase.
SBO and palm oil are some of the more cost-effective candidates. Vegetable oils differ in unsaturation and in the presence of functional groups. Low-viscosity vegetable oils may present a problem when mixing with high-viscosity polymers at high concentration of oils. To overcome this difficulty, we have polymerized SBO to obtain high-viscosity oils or solids similar to factice but without sulfur and without the excessive cross-linking. Unlike mineral oils, vegetable oils can be covulcanized with rubber to become the part of the polymer network. This helps to prevent “blooming” of oil. Although solubility of a polymer decreases with the increasing molecular weight of the solvent, using polymeric plasticizers (polymerized oils) instead of the pure vegetable oil has some advantages, such as lower diffusion (bleeding) from the rubber, higher degree of covulcanization if it takes place, and better oxidative stability due to the consumption of some double bonds during polymerization. Also, higher viscosity oils, such as polymeric oils, mix better with rubber at very high concentrations than monomeric oils.
Vegetable oils can be polymerized thermally (heat-bodied oils),12 which involves heating at high temperature followed by a significant degradation, or by oxidation (blown oils), accomplished by passing air through the hot oil. The resulting product contains various oxidation products such as aldehydes, ketones, hydroxyls, and carboxylic groups. In this work, we have used our patented procedure for low-temperature cationic polymerization of SBO.13,14 Figure 1 shows a schematic representation of the process and the structure of a highly branched polymerized triglyceride.
Although vegetable oils have been previously used in rubbers to improve different properties, there has been no systematic study of the effect of the concentration and molecular weight of natural oils on overall properties. The objective of this work was to study the feasibility of replacing a naphthenic petroleum oil with SBOs of different molecular weights and to examine the effect of molecular weight and concentration of polymeric SBOs on properties of rubber. Because of its unique chemical and physical properties, NR is used in many dynamic applications. Being nonpolar, NR can be blended with a number of other nonpolar elastomers. Blends of NR with BR and with SBR are commonly used for tires, vibration isolators and dampers, and belts. SBR is a general purpose rubber that finds additional use in wire and cable, footwear, and general mechanical goods. To cover a range of possible end uses, a 50:50 blend of NR and SBR was used for these evaluations.
It was assumed that the polymerized vegetable oils might covulcanize with rubber and impart new qualities to the product. The effect of two polymerized SBOs of different molecular weights and viscosities were compared with the effect of pure SBO on properties of rubber. The reference materials were rubbers containing naphthenic rubber process oil. Concentrations of oil were 10, 20, 40, and 60 parts per 100 phr.
SBO refined, bleached, and deodorized was obtained from Cargill (Des Moines, IA). The viscosity was 0.06 Pa·s at 25 °C, and the iodine value was 131.
Two polymerized oils designated as polymerized SBO1 (PSBO1) and polymerized SBO2 (PSBO2) were prepared using the procedure described in the patent.13 Properties of the oils are given in Table I, and the rubber formulation is presented in Table II.
Rubber compounding was carried out in a BR1600 lab Banbury using the following mix procedure:
First Cycle. — 0 s: load NR; 30 s or 80 °C: load styrene butadiene and 1/2 black; 100 °C: load 1/2 black and chemicals; 115 °C: add oil; 125 °C: sweep; 135 °C: dump.
Second Cycle. — 0 s: load first cycle master batch and curatives; 30 s or 95 °C: sweep; 60 s or 105 °C: dump.
The final dispersion was accomplished on a 15- × 30-cm two-roll mill.
Properties of rubber plasticized with oils was carried out using the following methods: hardness, ASTM D2240; tensile strength, elongation, and 100% modulus, ASTM D412; tear die C, ASTM D624; compression set, ASTM D395 method B; oven age, ASTM D573; dynamic properties, ASTM D5992; G′ and tan δ were tested at ±10% shear strain, 10 Hz, 21 °C.
RESULTS AND DISCUSSION
EFFECT OF OILS ON PROCESSING PROPERTIES
High-molecular-weight elastomers behave as high-viscosity non-Newtonian liquids before they are cross-linked. After cross-linking, they behave as elastic solids. A rheometer with heated dies is used to record a cure curve of modulus versus cure time. The early stage of the curing process measures the viscosity and the scorch time. The low torque value is a measure of the minimum rubber viscosity, and it gives an indication of how easily the compound will flow during processing. The period of time before vulcanization starts is referred to as the scorch time, and the scorch time is measured to a rise (in this case, two units indicated as Ts2) above the minimum viscosity. The cured modulus is measured by the high torque after the vulcanization process is essentially complete. The cure time is the time required for the cure to reach a certain state, in this case, 95% of total cure or Tc95.
Processing properties and cure characteristics were measured on a moving die rheometer at a cure temperature of 153 °C. Figures 2 and 3 display the effect of oil concentration on low and high rheometer torque values. Addition of oils reduces the viscosity of the uncured rubber and the modulus of the cured rubber in a similar fashion for all oils. However, the plasticizing effect of SBO and two polymeric oils seems to be stronger than that of the reference mineral oil, particularly at higher concentrations. Plasticizing power is the highest with SBO followed by SBO1 and SBO2.
The effect of oil content on cure time (Tc95) and scorch time (Ts2) is given in Figures 4 and 5. The addition of plasticizer increases both the scorch time and the cure time, but there is minimal difference between the four oils.
PHYSICAL PROPERTIES OF RUBBERS WITH OILS
Good compatibility of oils with rubber is required. If the oil is not compatible with the polymer, it may bleed out and give poor physical properties, a sticky surface, and poor adhesion in bonded parts. Compatibility is greatest when the solubility parameters of the plasticizer and polymer are similar. Solubility parameters for SBR were reported15 to be 16.6 to 16.9 (MPa)1/2, for NR 16.4 to 16.8 (MPa)1/2, and for naphthenic oil generally 14.4 to 14.5 (MPa)1/2. The SBO solubility parameter calculated by the van Krevelen method16 is 17.1 (MPa)1/2 and for polymerized oils is 17.7 (MPa)1/2. Thus, SBO and polymerized SBO should be more compatible with rubber than the naphthenic oil plasticizer.
Table III shows the properties of the different formulations.
Hardness is strongly affected by the level of plasticizer, as shown in Figure 6. SBO is a particularly effective plasticizer, providing significantly lower Shore A hardness than the other oils. The polymerized SBOs and the naphthenic oil behave similarly.
Tensile strength, as expected, decreases somewhat with an increasing concentration of oil (Figure 7). The SBO exhibits a greater decrease at 60 phr than the other plasticizers. This could be because it is approaching its compatibility limit or because it is more difficult to obtain adequate dispersion of the carbon black with this high concentration of low-viscosity oil. The best retention of tensile strength is with mineral oil (9% decrease between 10 and 60 phr), followed by two polymerized oils displaying the loss in strength of about 25% at 60 phr. It should be emphasized that SBO tracks well with the reference mineral oil up to the 40 phr level.
Elongation at break increases significantly as a function of plasticizer content (Figure 8). At 40 phr, the SBO shows significantly higher elongation than the other oils, but it drops at 60 phr because of limited compatibility.
Tear resistance decreases with the addition of plasticizers (Figure 9). At lower oil content, SBO seems to give higher tear strength than other three oils, but it had the lowest value for 60 phr. The scatter in the results was fairly large, not allowing definition of clear trends.
Plasticizers typically have a negative influence on compression set, and that effect is the highest with SBO (61.4% at 60 phr), followed by the two polymerized oils (58.7% at 60 phr), whereas mineral oil gave only 54% at 60 phr (Figure 10). These differences are relatively negligible, however.
EFFECT OF HEAT AGING ON PROPERTIES
The samples were oven aged by heating at 100 °C for 70 h and were then retested for hardness, tensile strength, and elongation. Although hardness decreases with addition of oils, aging causes hardness to increase relative to the unaged samples. Figure 11 shows the increase in hardness with aging. In general, the naphthenic oil aged the poorest, and PSBO2 is the most consistent across the range of plasticizer content.
Tensile strength after aging displayed the same trends with increasing oil content as unaged samples (Figure 12). The decrease in strength is similar for all samples. The loss of elongation after aging is shown in Figure 13. Again, performance is similar within experimental error for all four oils.
DYNAMIC PROPERTIES OF RUBBERS PLASTICIZED WITH OILS
The dynamic characteristics (G′ and tan δ) were measured using a high response servohydraulic test machine. This machine subjects a double-lap shear specimen to a sinusoidal shear motion at frequencies from 1 to 40 Hz and strains from ±1% to ±25% amplitude while measuring the resulting force. From the two signals, motion and force, and the geometry of the sample, the moduli are calculated. Fourier analysis is used to extract the frequency components from both the measured motion and force signals. G′ is the dynamic elastic shear modulus, and tan δ or loss factor is G″/G′. (G″ is the loss modulus in shear.) The reported values for G′ and tan δ are measured at 23 °C, 10 Hz, and ±10% shear strain. The static modulus is measured at 25% in shear.
This drop is largest with SBO, whereas the other three oils had a comparable effect at higher concentrations. On the other hand, the effect of oil content on damping as measured by tan δ at 21 °C was not significant because values varied within the limits of experimental error (Figure 16).
The dynamic/static ratio was similar for all plasticizers (Figure 17).
It should be noted that all tests were carried out at the same level of sulfur. Because vegetable oils can covulcanize with rubbers, a certain amount of sulfur will be consumed for vulcanization of the oil, which will, in theory, reduce the amount available for rubber cross-linking. Independent of the potential for covulcanization, it is well known that different properties in rubber are optimized at different cross-link densities. For example, tear strength and tensile strength increase with increasing cross-link density up to a point and then fall off with further increases in cross-link density. If there was a significant amount of the sulfur consumed in cross-linking the oil, we might have seen a loss in strength relative to the control. Conversely, increasing part or all of the cure package might increase the probability of curing the polymeric oil chains into the rubber matrix. To investigate this issue, compounds 10 and 11 containing 40 phr polymerized SBO were remixed with additional sulfur. The sulfur level was increased by 50% or from 1.5 phr to 2.25 phr. Table IV shows a comparison of the original compounds 10 and 11 with compounds 17 and 18, which contain a higher level of sulfur. The loss of tensile strength and elongation upon further sulfur addition suggests that we are now over–cross-linking the rubber (i.e. the oil was not consuming an appreciable amount of the available sulfur). This is not to say that a little more sulfur might not be advantageous, but a 50% increase for 40 phr oil is beyond the level that is desired. We chose a 50% increase on the assumption that the oil might have a higher probability of being cross-linked into the network with significantly more sulfur. This does mean that the level of sulfur does not need to be increased when substituting polymerized SBO for naphthenic process oil in NR, SBR, or blends thereof.
Elastic recovery was measured at 100% and 200% elongation by extending the sample at 100%/min (50 mm/min) and immediately returning to the initial position at the same rate. The samples with 10 phr oil displayed the recovery of 93% at 100% deformation and 90% at 200% deformation. By increasing oil content to 60 phr, the recovery improved to 98% at 100% elongation and 96% at 200% deformation and was independent on the type of oil, except for SBO, which showed 98% recovery at 100% extension for all concentrations in rubber.
THERMAL PROPERTIES OF OIL-PLASTICIZED RUBBER
The glass transition (Tg) of NR (poly 1,4 isoprene) is reported to be −62 °C, whereas that of SBR with 23.5% styrene is −52 °C.17 Glass transitions determined by differential scanning calorimetry (DSC) decreased by about 3 °C with addition of 60 phr of each plasticizer, varying from −63 °C to −69 °C for different samples, with increasing oil content from 10 to 60 phr. The plasticizing effect of neat SBO was slightly higher than other oils. Tg of the pure polymerized SBO was −69 °C. All oils crystallize below room temperature to form greases. Endothermic peaks observed at about −20 °C were ascribed to melting of greases.
Dynamic mechanical curves obtained at the heating rate of 5 °C and 10 Hz displayed α-transitions on loss modulus curves at −58 °C for all samples with 10 phr oil except for SBO, which had −60 °C. This peak is probably due to polyisoprene. Increasing oil content decreased slightly the temperature of α-transition to −60 °C, except for SBO, which displayed −67 °C for 60 phr. A shoulder observed at about 20 °C higher temperature of the α-transition is possibly due to SBR.
Thermal degradation as measured by thermogravimetric analysis displayed three distinct weight loss regions: the first was at 200 to 300 °C, which was attributed to the oil loss, whereas rubber loss was complete between 350 and 500 °C in two steps. The residue of 30 to 40% above 500 °C corresponds roughly to the carbon black content. Weight loss in the first step was consistently higher for mineral oil–plasticized rubbers, particularly at higher concentrations (13% vs 3–4% for vegetable oil–extended samples).
Based on solubility parameters, SBOs are predicted to have better compatibility with SBR/NR than mineral oil.
Glass transition of the rubber/oil blends varied in a narrow temperature range.
Polymerized SBOs display better compatibility, especially at high concentrations; give better heat stability and somewhat better retention of properties than SBO; and give lower plasticization than SBO.
SBO has the highest plasticizing effect and excellent performance up to 40 phr. Beyond that level, the polymerized oils are preferred.
Increasing the molecular weight of SBO decreases plasticization power but allows higher loadings.
Compression set increases with higher levels of oils, but the differences between oils are relatively negligible.
Dynamic properties and heat aging characteristics are similar for mineral oil and vegetable oil plasticized rubbers.
All oils improve elastic recovery, particularly SBO.
The thermal stability of vegetable oil–plasticized rubbers was better than that of the samples with mineral oil.
SBO and polymerized SBO can be substituted for naphthenic process oil in NR, SBR, or blends thereof without the need to make other adjustments to the formulation.
Utilization of SBO is currently more economical than the petrochemical oils. Polymerized SBOs are not commercial products, but because of the simplicity of their preparation, they should be also very competitive materials.
We are indebted to United Soybean Board for funding this research.