Physical properties and their meaning.
Virtually all rubber products are crosslinked (vulcanized before being placed into end-use applications. Prior to crosslinking, a number of additives (often in large amounts) are incorporated into rubber. Frequently rubber is the minor component in the resulting composition or compound. Different types of rubber, in combination with different types and levels of additives, cause wide-ranging property variations in unvulcanized rubber compounds.
Important properties of unvulcanized rubber and rubber compounds include flow behavior (rheology) and tack. Compounds must possess sufficient tack for successful fabrication of some products, like tires. Proper flow behavior is needed to obtain good dispersion of additives in rubber during mixing, and also for subsequent shaping operations like extrusion and molding.
A number of test have been designed to characterize rheology and tack. Some of these tests are used because they provide fundamental information; others are widely used because they are convenient and economical to perform. An example of the latter is the Mooney viscosity test.
This test, described by ASTM D 1646, determines viscosity of either raw or compounded rubber. In this test a rotor rotates in the rubber to be tested as shown in figure 4. All surfaces in contact with rubber are roughened to minimize or prevent rubber slippage during testing. Include are the surfaces in the cylindrical cavity formed by the upper and lower platens, as well as the rotor surfaces.
In operation, the rotor speed is 2 rpm and the test temperature is normally 100 [degrees] C (212 [degrees] F). Usually rubber is warmed for one minute prior to starting rotation of the rotor, after which a reading is taken at four minutes. Mooney viscosity is given by a number which is a measure of rotor torque (lbf-in). One Mooney unit is equal to 0.083 N*m (0.735 lbf*in). Results for a typical test are reported as:
50-ML 1 + 4 (100 [degrees] C)
Here, 50-M is the Mooney viscosity number, L indicates the use of the large rotor, 1 is the time in minutes the specimen is heated prior to starting the rotor, 4 is the time in minutes after starting the rotor, and 100 [degrees] C is the test temperature.
There are several advantage is that the machine operates at a shear rate of about 1 [s.sup.-1]. Shear rate can be visualized by sliding the cards in a deck of cards across one another (ref. 5). The shear rate is the relative velocity of the bottom and top cared (cm/sec.) divided by the thickness of the deck (cm), to yield units of reciprocal seconds ([s.sup.-1]). The shear rates encountered in rubber processing, e.g. extrusion and molding, are often considerably higher than 1 ([s.sup.-1]). Figure 5 shows an example for two molding methods (ref. 5). Rates for compression molding are about 1 [s.sup.-1] to 10 [s.sup.01],while those for injection molding are 1,000 [s.sup.-1] and higher. Over this shear rate range, the viscosity changes by more than two decades. Hence, the Mooney reading should predict rubber behavior better during compression molding than during injection molding.
Even with roughened metal surfaces in contract with rubber (figure 4), slippage sometimes occurs at rubber/metal interfaces. Slippage can be minimized by increasing test temperature or reducing rotor speed (ref. 6). Reduced speed requires machine modification and therefore is not as practical as increasing temperature. Hence for some rubbers like EPDM (ethylene propylene diene monomer) and IIR (isobutylene-isoprene), a higher test temperature of 125 [degrees] C (257 [degrees] C) is used. Figure 5 shows that the temperature increase from 100 [degrees] C (212 [degrees] F) to 125 [degrees] C (257 [degrees] F) lower viscosity significantly for a typical rubber (note that viscosity is on a log scale). This lower viscosity decreases shear stress at the rubber/metal interface, which in turn decreases slippage at the interface.
Further efforts to reduce wall slippage include providing a fresh rubber surface to the rubber/metal interface, and controlling cavity pressure (ref. 7). A transfer pot and ram, located directly above the Mooney cavity, provide fresh rubber to the cavity. The ram controls pressure in the Mooney cavity. This special machine is equipped with a variable speed rotor and several advantages are claimed for it.
The Mooney test described by ASTM D 1646 has been the major method for determining viscosity in the rubber industry for several decades. It will no doubt remain so for some time to come because it is convenient and economical to run. Its single rate-of-shear capability is a limitation. A capillary rheometer overcomes this limitation because it measures rubber viscosity over a range of shear rates.
A capillary rheometer consists basically of a cylinder and piston (ref. 8). For testing, rubber is placed in the cylinder between the piston and the base of the cylinder. As the piston advances, rubber flows through an opening (capillary) provided in the base of the cylinder. The three major test variables are:
* length and diameter of the capillary
* temperature of the cylinder
* piston speed
Capillary rheometers provide a range (ref. 8) of shear rates of [10.sup.-1] through [10.sup.-5][s.sup.-1]. This range includes the rates generally encountered in rubber processing.
In use, a transducer measures the force necessary to push rubber through a capillary. The raw data consist of force as a function of piston speed. When these data are properly treated (ref. 8), a graph is obtained like the one in figure 5. With a capillary rheometer, viscosity can be measured at the same shear rates as those encountered in processing, e.g. injection molding. Hence, test conditions are more relevant to processing conditions and this is an advantage.
The capillary rheometer test is more complex and time-consuming than the Mooney test. Thus the capillary rheometer is ordinarily used in research and development (R&D) activities. The Mooney test is used extensively for control testing purposes as well as in R&D activities. Both the Mooney viscometer and most capillary rheometers are constant rate devices. Now a constant load test is described.
This test is described by ASTM D 926. The apparatus consists of two metal plates, one of which is free to move vertically and parallel to the other plate on a common axis. A dial indicator follows the distance traveled by the plate free to move. Normally the plates are heated to an elevated temperature. For testing, a load is applied to a small sample of preheated, unvulcanized rubber of specified shape and volume that is placed between the plates. The load squeezes the rubber between plates for a specified time, after which the dial reading is noted in millimeters. This reading, multiplied by 100, is the "plasticity number."
Test temperature and the rubber viscosity mainly determine the shear rate during test. Shear rate is usually very low, below 0.1 [s.sup.-1]. This low rate is relevant to the slow flow of rubber that can occur in stacked rubber bales in storage. An example is polybutadiene with a high cis content and a narrow molecular weight distribution. The plasticity test should predict the flow tendency of this rubber because flow rate is low, and flow occurs under constant load.
Tack and green strength
The fabrication of many rubber products requires mating two or more rubber surfaces. Tack is a measure of the ability of these mated surfaces to resist separation after they have been lightly pressed together for a short time (ref. 9). Green strength is the resistance offered to deformation and fracture by unvulcanized rubber and rubber compounds. High green strength is a necessary, though insufficient, requirement for high tack.
Other requirements are intimate molecular contact between mated surfaces, and interdiffusion of rubber molecules across their interface. For intimate contact to occur, it may be necessary to displace entrapped or adsorbed gases.
A high level of chain mobility favors interdiffusion of molecules. Chain mobility, and consequently tack, are low for some rubbers. Acrylonitrile butadiene rubber (NBR) with a relatively high acrylonitrile content is an example. To increase tack in NBR, the surface can be wiped with an appropriate solvent like methyl ethyl ketone (ref. 10). The solvent temporarily plasticizes the NBR surface and increases mobility of the molecules near the surface. The increased mobility favors interdiffusion and entanglement among molecules, thus increasing tack.
Rubber molecules must be sufficiently long for entanglement to occur, as they are in most typical rubbers. Entanglement is a major contributor to tack for amorphous rubbers like NBR and SBR. As temperature increases, disentanglement increases as shown in zone IV-A of figure 1 (RW April, 1996, p. 18). Therefore, green strength decreases with increased temperature.
Crystallization is also an important factor because it can either decrease or increase tack. For example, some types of EPDM are partially crystalline inn the unstrained state (ref. 9). While this crystallinity generally improves green strength, it reduces chain mobility. The net effect is reduced tack.
Crystallization has a different effect in a strain-crystallizing rubber like NR. Unstrained NR is normally amorphous, where there are no crystallites present to reduce chain mobility and tack. When the surfaces of amorphous NR are mated and strained, crystallization occurs; crystallites reinforce the NR and increase its green strength. Hence, high tack in NR is due to high molecular mobility when surfaces are mated, followed by subsequent crystallization when the mated surfaces are strained.
It is possible for the tack to be undesirably high, for instance in the fabrication of tires (ref. 11). Many individual rubber components must be placed in proper alignment with one another for tires to perform satisfactory. Realignment of these components is necessary if the initial alignment was unsatisfactory. Realignment causes distortion of uncured rubber components if their tack is too high. Hence the tack level should be high, but it should not be high enough to cause realignment problems during fabrication. Tack level should be stable with time so that fabricated composites don't separate and distort during storage (prior to crosslinking).
Tests for tack are as simple as separating two mated rubber surfaces by hand. The quantitative measurements described by Hamed can be used (ref. 9).
Unvulcanized rubber generally goes through some shaping operation like extrusion or molding before it is vulcanized or crosslinked. Premature crosslinking or scorch must be avoided if these shaping operations are to be successful. Scorch raised rubber viscosity sharply and causes rough extrusions; scorch also prevents molds from filling and causes rough moldings. The effect of increasing molecular weight and of scorch on rubber viscosity is considered next.
Most rubber used industrially is considerably above its entanglement molecular weight ([M.sub.b]). For example, [M.sub.b] for NR (ref. 12) is 17,000; the molecular weight of NR and other commonly used rubbers is generally ten or more times this value. For molecular weights greater than [M.sub.b], the viscosity increases as the 3.5 power of the molecular weight (ref. 12). This relationship demonstrates the extremely sharp dependence of viscosity on molecular weight.
To illustrate this dependency, the effect of increased molecular weight is now considered for a rubber that consists of linear equal-length molecules above their entanglement molecular weight. If they are endlinked in pairs, rubber viscosity increases by a factor of [2.sup.3.5] or 11.3; for four endlinked molecules, viscosity increases by [4.sup.3.5] or 128 times. These simple calculations illustrate the extremely sharp increase in viscosity with endlinking. Upon crosslinking these endlinked molecules, viscosity continues to increase sharply.
Because premature crosslinking (scorch) in rubber is such an important consideration, several methods are available to assess it. The Mooney scorch machine and the oscillating disk rheometer are most commonly used to measure scorch.
This scorch test, described by ASTM D 1646, measures the onset of vulcanization or the time for crosslinking to cause a given increase in torque. An optimum test temperature will yield the required increase within a period of 10 to 20 min. A typical test temperature is 121[degrees]C (250 [degrees]F).
The scorch test is normally run with a smaller rotor than that illustrated in figure-4. As with the Mooney viscosity test, rubber is heated between platens for one minute before starting rotation (figure 6). Viscosity is high initially because rubber requires time to reach platen temperature. As the rubber heats up, viscosity decreases and eventually viscosity reaches a minimum. With continued heating, scorch (crosslinking) occurs and the scorch time is typically reported as [t.sub.3] or [t.sub.5]. The value of [t.sub.3] is the time required for a 3-Mooney unit increase above the minimum value (15 min. in figure 6); [t.sub.5] is the time for a 5-Mooney unit increase.
With additional test time, the curve continues upward as crosslinking increases. Increased crosslinking eventually causes considerable tearing and slippage of rubber. The crosslinked rubber can no longer accommodate the continuous rotation of the rotor. The test is normally stopped shortly after reaching the level of crosslinking associated with the [t.sub.5] value.
Useful information at higher levels of crosslinking (cure characteristics) can be obtained with the oscillating disc rheometer.
Oscillating disc rheometer (ODR)
The ODR (ASTM D 2084) differs from the Mooney in several important ways. The Mooney rotor rotates continuously about its axis. The rotor of the ODR oscillates about its axis through a small arc (typically 1[degree] as shown in figure 7). For rubbers which produce low torque, arcs of [+ or -]3[degrees] and [+ or -]5[degrees] can be used to increase the torque value. Thus in contrast to the Mooney machine, the ODR measures properties of both partially-and fully-crosslinked rubber.
The Mooney rotor is a disk (figure 4) and therefore does not provide a uniform shear rate. The biconical ODR disk (figure 7) provides a uniform shear rate over its entire surface (ref. 13).
Another difference is the higher temperature typically used with the ODR, relative to the Mooney machine. The higher ODR temperature shortens the time to obtain scorch times and curing curves.
A typical ODR curve is illustrated in figure 8 where torque is shown as a function of time. The initial part of the curve is very similar to the Mooney scorch curve. For an ODR used with 1[degree] arc, the scorch value is the time in minutes to a 1 dN[multiplied by]m (lbf[multiplied by]in rise ([t.sub.s]1) above the minimum torque value. In this vicinity rubber viscosity contributes mainly to the torque. As crosslinking increases, the rubber becomes more elastic and eventually maximum torque is reached. The time to reach 90% of maximum torque, t(90), is often taken as the cure time.
In figure 8 the solid line to the right of t(90) shows that rubber has cured to an equilibrium torque. Sometimes there is no equilibrium torque as shown by the top dashed line. Here, crosslinking continues at a lower rate and no torque maximum is evident. At other times, torque decreases after the maximum is reached as shown by the bottom dashed line; this behavior is called reversion. Reversion is often seen in NR cured at high temperatures with sulfur.
The relationship has been examined between maximum torque attained in the ODR and stress measured at 300% elongation (ref. 6). The effect of contamination of the ODR rotor on this relationship was established. It was found that best results were obtained when the rotor arc was [+ or -]1[degree], rather than [+ or -]3[degrees] or [+ or -]5[degrees]. Therefore, [+ or -]1[degree] arc is now specified as standard by ASTM D 2084 to reduce slippage.
Scorch by capillary rheometer
While the Mooney machine and the ODR are most often used to determine scorch, specialized tests are sometimes used. For example, computer models require scorch data to simulate mold filling and vulcanization during injection molding (ref. 14). It was found that Mooney scorch and ODR scorch values were not reliable data for computer simulation of injection molding. Only scorch values determined by capillary rheometry were reliable.
This result is probably related to the similarities between the capillary rheometer and injection molding. Geometries are similar for capillary rheometers and injection molding machines, and shear rates can be made similar. Effective use of test equipment and molding machines aids development of desired vulcanized properties.
[Figures 4 to 8 ILLUSTRATION OMITTED]
5. J.G. Sommer, "Rubber molding - equipment and methods," presendted at the Nineteenth Annual Lecture Series (Akron Rubber Group) at The University of Akron, May 3, 1982. 6. G.E. Decker and R.D Stiehler, "Standardization of Mooney viscometer and oscillating-disk cure meter," in ASTM Special Technical Publication 553, 1974, p.19. 7. D.M. Turner and M.D. Moore, Plastics and Rubber: Processing, 5, 81 (1980). 8. G.P. Colbert, Akron Rubber Group Technical Symposium, 1978-1979, p.66. 9. G.R. Hamed, Rubber Chem. Technol. 54, 576 (1981). 10. J.H. Daly, W.A. Hartz, D.A. Meyer and J.G. Sommer, Journal of Applied Polymer Science: Applied Polymer Symposium 25, p.261 (1974). 11. J.R. Beatty, "Physical testing," Presented at the Tenth Annual Lecture Series (Akron Rubber Group), Akron, Ohio, February 5, 1973). 12. F. Bueche, "Physical properties of polymers," Interscience Publishers, New York, 1962. 13. F.S. Conant, "Physical testing of vulcanizates," Chapter 5 in Rubber Technology, M. Morton, Ed., Van Nostrand Reinhold Company, Second Edition, 1973, p.114. 14. J.D. Byam and G.P. Colbert, Plastics and Rubber: Processing, 5, 95 (1980).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||part 2; testing the properties of unvulcanized rubber and rubber compounds|
|Author:||Sommer, John G.|
|Date:||Jun 1, 1996|
|Previous Article:||SR consumption up 5% in 1995, 1996 increase forecast to be 3.2%.|
|Next Article:||Use of precipitated silica to improve brass-coated wire-to-rubber adhesion.|