A comparative study of step curing and continuous curing methods.

The optimum state of cure which is quite often referred to as the technical cure for any rubber product signifies the best balance of properties that can be technically achieved for that product. The main advantage of the step cure method over the continuous vulcanization method is that the former allows the flexibility in choosing the optimum vulcanization time depending on certain specific properties, like tensile, modulus, tear, etc. It is quite often found that all properties do not develop at the same rate during vulcanization and the property-maxima do not occur at the same point on the time scale. The maximum tensile strength generally occurs at a shorter vulcanization time compared to the maximum modulus (ref. 1). This flexibility in determining the optimum cure time with respect to specific property becomes important, when the property concerned is the most critical one for some specific application. However the inherent testing errors of the step curing method limit its application for detecting smaller variation in curing behavior. The step curing method based on the measurement of mechanical property like tensile, modulus, etc., are well known. Recently the step curing and continuous curing techniques involving measurement of electrical properties like dielectric constant, loss factor, are also reported in the literature (refs. 2-7).

In this article step curing methods based on mechanical and electrical properties are determined for an EPDM-bromobutyl blend and compared with that of a continuous vulcanization method using a Monsanto rheometer. This particular blend system is chosen because it is expected to exhibit non equilibrium curing, that is a marching modulus curve against cure time. One of the objectives of this study is to check the efficacy of different step curing methods for determining the optimum cure time, especially for a system which exhibits marching modulus against cure time in a rheometer. For such a system optimum cure time is difficult to measure precisely from a rheometer.

Experimental

The elastomers were mixed with different ingredients according to the given formula (table 1) in a two-roll laboratory mill. The compound was cured at [160, 170 degrees] and [180 degrees C] for various times ranging from 10 to 60 minutes.

The ambient dielectric measurements were taken on a circular pallet like samples (dia 2 cm. thickness = 0.15 cm) punched out from vulcanized sheet. The instrument used for dielectric measurement was a Hewlett-Packard L.F. Impedence Analyzer (model 4192A) operative over the frequency range of 104-107 HZ.

The modulus at three different strains (300%, 400% and 500% elongation) were determined for different samples using an Instron universal testing machine (model 1095) according to ASTM D-412.

The cure characteristics were determined from the rheograph obtained at different cure temperatures using a Monsanto rheometer (R- 100).

Results and discussion

The mechanical probing Mechanical properties like modulus at different strains, tensile strength, tear strength and elongation at break are all functions of degree of crosslinking. However, the dependency of these properties on crosslink density is more pronounced in the case gum vulcanizates of strain-crystallizing rubbers, compared to that of non-strain-crystallizing ones.

Modulae at different strains namely 300, 400 and 500% elongations for samples cured for different time durations at three curing temperatures namely [160, 170, 180 degrees C] are presented in figures 1-3 respectively.

As shown in figure 1, the modulus at 300% elongation increases initially with cure time, attains the maximum and then decreases reflecting a clear tendency of reversion. The increase in modulus with cure time is a clear indication of the introduction of more and more chemical crosslinks in the rubber matrix, whereas the decrease in modulus reflects the destruction of some of these crosslinks as well as modification of the main chain due to prolonged exposure to elevated temperature. The reversion is well pronounced in moduluscuretime curves for [170 and 180 degrees C]. However this tendency of reversion is absent when the blend is cured at [160 degrees C]. This is mainly because the reversion reaction is nothing but an accelerated aging occurring at the curing temperature. The EPDM-BIIR blend system is appreciably age resistant at [160 degrees C] but it reflects some reversion at higher temperatures like [170 and 180 degrees C]. The optimum cure time decreases as the curing temperature increases. The optimum cure time for [170 and 180 degrees C] cure temperatures corresponds to the time at which the maximum modulus is attained. But for a temperalure. of [160 degrees C] the optimum time corresponds to the time when modulus becomes invariant with respect to cure time. The modulus at 400% elongation against cure time is presented in figure 2. The three plots in the figure correspond to the three different temperatures of curing. The sharp maxima are observed in modulus vs. cure time curves for temperatures Of [170 C and 180 C] whereas a broad maximum exists for a cure temperature of [160 degrees C]. The very similar set of curves are obtained for modulus at 500% elongation versus cure time at three different temperatures [160 C, 170 C and 180 degrees C]. It is noted that the rates of both forward (i.e. the development of modulus) and backward reactions (reversion) increase as the cure temperature is increased. Moreover, the extent of property development at any cure time increases as the cure temperature increases, thus the magnitude of maximum modulus is found to be the highest for [180 degrees C].

These step curing methods based on the measurement of modulae at different strains against cure time give almost similar information about the kinetics of curing reaction. However for the present system somewhat better indications for- cure are obtained from modulus-cure time plots for 400% and 500% elongations compared to that for 300% elongation when the latter is the most commonly used for the indication of property development for rubber compounds during cure.

The electrical probing

The dielectric analysis technique utilizes polar groups that are present in the elastomer mix. A dynamic electrical stress (an AC voltage) is applied to the system and dielectric constant and loss tangent or loss factor (product of dielectric constant and loss tangent) are measured as functions of cure time, cure temperature and also as functions of frequency of the applied electric field.

The curing process can not in general be described as a simple chemical reaction but rather as a mixture of complex reactions involving various polar and ionic mechanisms (ref. 8). When the polymer is initially polar or if the crosslinking agent introduces some polar groups in the elastomer network, the total concentration of dipoles is likely to change during the curing reaction. This is reflected as a change in both dielectric constant (e') and loss tangent (tan 8 ). The unsaturated rubber chains when vulcanized with sulfur plus accelerator system give rise to a three dimensional network containing different types of constituents as shown in figure 4.

The sulfur is combined in the vulcanization network in a number of ways; as crosslinks it may be present in the form of a monosulfide (-s-), disuifide (-52-) polysulfide (-sx-). However, only a part of the total sulfur added to the system is accountable in the form of different sulfide crosslinkages. Different side reactions leading to the formation of pendent sulfide and hetrocyclic sulfide also proceed simultaneously. Hence another part of the total sulfur is combined as heterocyclic and pendent sulfide; the unreacted sulfur may just be present as a filler. A few carbonyl > C=0 groups are also formed due to oxidation of some [CH sub.2] groups during vulcanization at elevated temperature. The possibilities of formation of hydroxyl (-OH) aldehide (-CHO) and other polar groups involving oxygen and hydrogen are also there. A few halogenated groups may be present in the halobutyl phase, with some sulfur containing crosslinks and substituents in the pendent diene portion of EPDM (not shown in the figure). However, the concentration of these groups are quite small. The curing reaction of an elastomer also yields some extra network materials which do not have any link with the main polymer network, for example unreacted curative, activators, accelerators and also reaction products of vulcanization like zinc sulfide and zinc salt of accelerator and unreacted ZnO etc. (refs. 9 and 10). During the course of the vulcanization reaction the relative concentration of these polar groups is always changing. Prior to the final formation of the thermally stable monosulfidic or disulfidic links the polysulfidic crosslinks may undergo thermal reaction involving formation and destruction of cyclic and pendent sulfide (refs. 7 and 8).

However, each different molecular species has its own dielectric response and that gets manifested in the variation of magnitude of dielectric constant and loss. The contribution of the sulfidic crosslinks especially mono- and di- to,the dielectric response is small, though sulfur crosslinks introduce some permanent dipoles to the polymer backbone, but these crosslinks simultaneously impose restrictions on the mobility of the main chain, i.e. on the orientation of dipoles. The main contribution of dipole response generally stems from hetrocyclic sulfide and pendent sulfide groups (refs. 11 and 12).

The dielectric response measured in terms of dielectric constant and loss tangent depends on two parameters (ref. 13):

* The dipole density of the total number of polar groups present in the system per unit volumes.

* The mobility of these dipoles to follow the reversal of polarity of the applied electric field.

Thus a curing process may be visualized as the consequence of two opposite effects simultaneously operative in the system; one, the introduction of polar linkages which increase the dielectric response; two, introduction of crosslink which imposes some restriction on the mobility of the polar group and thereby decreases the dielectric response.

The figures 5, 6 and 7 represent the variation of dielectric constant (e') with cure time at three different temperatures namely [160, 170 and 180 degrees C]. As observed from these figures the dielectric constant increases with the increase of cure time, then attains the maximum value at a definite cure time which is often referred to as the optimum cure time with respect to dielectric constant measurement (ref. 5). It is also found that the optimum cure time is independent of frequency of applied electric field. However, the magnitude of dielectric constant increases with the decrease of frequency of the field. This is due to the fact that as the measurement frequency is lowered more and more dipoles can orient themselves (the higher extent of dipolar polarization is possible) with the reversal applied electric field. Moreover, the contribution of Maxwell-Wagner interfacial polarization due to heterogerfiety present in the system progressively increases as the measurement frequency is decreased (ref. 13). The lowest optimum cure time is registered for the highest curing temperature, [180 degrees C]. It is also noteworthy that the dielectric constant at different cure time increases as the cure temperature is increased. This is due to the fact that at an elevated temperature vulcanization reaction occurs with faster speed as well as to a higher extent. The rate of change of dipoledensity increases with the increase of the curing temperature. In fact, the rate of increase of both forward and backward reaction increases with the increase of cure temperature. Thus a strong tendency of reversion is observed at temperatures of [170 degrees and 180 degrees C]. It was observed that the variation of loss tangent (tan 6) against cure time for different curing temperatures, that unlike dielectric constant the variation of tan 6 is relatively much less sensitive to cure time. At any fixed cure time and temperature, tan e increases with the decrease of measurement frequency, due to the increased contribution of interfacial polarization at lower frequencies.

A slow increase of tan a with cure time is observed for the sample cured at [160 degrees C] for all frequencies of measurement except 3.98 x 104 Hz, whereas a very weak tendency of reversion and attainment of invariance is observed in tan versus cure time plots at different frequencies for the sample cured at 170 degrees C]. But a more clearcut tendency of reversion is observed for corresponding plots for the sample cured at 180 degrees C]. The optimum curetime calculated from the variation of tan (e) against cure time decreases with the increase of cure temperature.

Continuous vulcanization probing

In the rheometer curves of samples cured at different temperatures, a tendency of marching modulus is observed for the sample cured at [150 and 160 degrees C] whereas the torque-time curve for the sample cured at [170 degrees C] exhibits a plateau effect. A distinct reversion tendency is observed for the samples cured at [180 degrees C]. It is difficult to calculate precisely the optimum cure time from marching modulus curves. A first order kinetics may be applied for these rheometric curves and the following kinetic equation is proposed:

ln(Rm-Rt) = -k t where Rm represents the maximum torque or (invarient torque) and it is proportional to the maximum crosslink density that can develop in the matrix at a given condition. Rt represents torque at any time instant t, and k is the rate constant of the proposed first order curing reaction. The plots of ln(Rm-Rt) against t for different cure temperatures are found to be straight lines, and rate constant k can be calculated from the slope of the plot. The rate constant k increases with the increase of cure temperature.

Calculation of energy of activation 

The activation energy of the curing process can be calculated from modulus measurement, dielectric and rheometric studies (refs. 2-6).

The relative reaction rate constant is inversely proportional to optimum curetime or time corresponding to maximum property development (t sub.max). The semilog plot of [t sub. max] against the reciprocal of cure temperature expressed in absolute scale is found to be a straight line.

* The activation energy of vulcanization reaction using different techniques can be calculated from the slope of the plots.

* The activation energy for rheometric study can be calculated in two different ways.

* From the slope of the semi4og plot of rheometric optimum cure time (t sub.90) against reciprocal of cure temperature.

* From the slope of the semi-log plot of different rate constant k against reciprocal of cure temperature.

* The activation energies thus calculated from different techniques are presented in table 2.

It is found that the value of activation energy obtained from modulus measurements are very close to those obtained from dielectric and rheometric measurements. However Eact obtained from first order reaction rate constants are somewhat higher. So it may be concluded that the applicability of first order reaction kinetics for vulcanization reaction may not be exact, though the method has some degree of practical competence.

Conclusion

The results show that there exists a correlation among the state of cure of a vulcanizate, its mechanical and dielectric properties.

Both electrical and mechanical step curing methods give better indication of optimum cure time especially at lower cure temperature for this blend elastomer system which exhibits marching modulus in torque-time plot.

The value of activation energy Eact calculated from mechanical, electrical stepcuring and continuous rheometric probing are found to be almost equal. But higher Eact is obtained when first order reaction kinetics are applied for rheometric data.

[TABLES OMITTED]

References

1. A.B. Sullivan and R. W. Wise "Rubber technology" Edt. M. Morton p. 117, Chapter 4. Van Nostrand ReinhoM Co. (New York) (1987).

2. D. Khastgir, P.K. Ghoshal and C.K. Das, Kautshuk. Gummi Kunststoffe 36, 277 (1983).

3. D. Khastgir et al. Elastomerics 116 (2) 26 (1984).

4. D. Khastgir, P.K. Ghoshal and C.K. Das, Rubber World 188 (3) 34 (1983).

5. D. Khastgir, P.K. Ghoshal and C.K. Das, Polymer, 24, 617 (1983).

6. Z.N. Sanjana, J.T. Siemen and J.P. Meir, IRC 1980, Proceeding, HT Kharagpur (India), p. 110.

7. Sture Persson, IRC 1986, proceeding, Gottenberg (Sweden) p. 150.

8. N.J. Morrison, M. Porter, Rubb. Chem and Tech. 57, 63 (1984).

9. E.R. Rodger, Gummi Asbest Kunstoffe 34, 124 and 300 (1985).

10. H.L. Curtis, A.T. McPherson and A.H. Scot. J. Res. Nat Bur. Hand, 11, 173 (1933).

11. A . Schaliamach Trans. Inst. Rub. Ind. 27, 40 (1961)

12. D. Khastgir, H.S. Maiti and P.C. Banerjee "Kautschuk Gummi and Kunststoffe, 40 (4) 333 (1987).

Acknowledgements

"New rubber friction testing machine" is based on a paper presented at the May, 1993 Rubber Division meeting. "Optimal measurement, use of dynamic properties from the MDR for compound analysis" is based on a paper presented at the May, 1992 Rubber Division meeting.

[TABLES OMITTED]