Change in Dielectric Losses of Rubbers Based on Ethylene-Propylene Rubbers over Time during Ionization Aging

O. N. Sokolovskii, N. R. Gonchar, M. N. Treskina

(Presented by the scientific seminar of the Department of EIKT)

In the present work, the influence of the vulcanizing group and filler on the dielectric losses of rubbers based on ethylene-propylene rubbers, and the change in dielectric losses over the time of ionization aging, was studied.

The composition of the vulcanizates investigated is given in Table 1.

Table 1

The specimens were prepared by vulcanizing the rubber compound in a press and had the form of plates 0.3 mm thick. To measure the dielectric loss tangent $\operatorname{tg}\delta$ and the dielectric permittivity $\varepsilon$, a differential bridge described in [1] was assembled.

Measurements of $\operatorname{tg}\delta$ and $\varepsilon$ were carried out in the frequency range from 20 Hz to 200 kHz and at temperatures from $-80^\circ$C to $+120^\circ$C. This temperature and frequency interval makes it possible to obtain good resolution of the relaxation spectra of polymeric materials [2].

The relative error in determining $\operatorname{tg}\delta$ and $\varepsilon$ is $\pm 10\%$ and $\pm 2\%$, respectively. Based on the results of measurements of $\operatorname{tg}\delta$ and $\varepsilon$, dependences of these quantities on temperature and frequency were constructed for each of the vulcanizates studied. For vulcanizate ET-96 such a dependence is shown in Fig. 1. As can be seen from Fig. 1, for unfilled

vulcanizate based on SKEPT with a sulfur-containing vulcanizing group, the temperature–frequency behavior of \(\operatorname{tg}\delta\) exhibits one maximum in the region of negative temperatures, shifting with increasing frequency toward higher temperatures. In the unfilled vulcanizate ET-95, obtained with dicumyl peroxide, \(\operatorname{tg}\delta\) likewise passes through a maximum in the region of negative temperatures, but the value of \(\operatorname{tg}\delta_{\max}\) for them is almost 2.5 times smaller than for the ET-96 vulcanizate with a sulfur-containing vulcanizing group.

Fig. 1. Temperature–frequency dependence of \(\operatorname{tg}\delta\) and \(\varepsilon\) of the ET-96 vulcanizate.

Studies showed that the concentration of crosslinks in the ET-95 and ET-96 vulcanizates is of the same order (\(11 \cdot 10^{19}\ \text{cm}^{-3}\) and \(8 \cdot 10^{19}\ \text{cm}^{-3}\), respectively). The magnitude of \(\operatorname{tg}\delta_{\max}\) of these vulcanizates is apparently explained by the different type of crosslinks formed during vulcanization of SKEPT in the presence of sulfur (ET-96 vulcanizate) and dicumyl peroxide (ET-95 vulcanizate). During vulcanization with sulfur, polar sulfur bridges are formed, \(-\mathrm{C}-\mathrm{S}_x-\mathrm{C}-\) (where \(x=2\) and higher); in peroxide vulcanizates, crosslinks \(-\mathrm{C}-\mathrm{C}-\) are formed, which are less mobile and less polar [3].

From the temperature–frequency dependences, the activation energies of the relaxation processes were calculated, equal to \(51\ \text{kcal/mol}\)

for the peroxide vulcanizate ÉT-95 and 30 kcal/mole for the sulfur vulcanizate ÉT-96. Consequently, the values of $\operatorname{tg}\sigma$ of the vulcanizates in the given temperature-frequency interval are due to dipole-segmental losses associated with the relaxation of polar groups together with the nearest parts of the macromolecule (segments).

The relaxation time in the region of dipole-segmental losses for ethylene-propylene unfilled vulcanizates is $10^{-3}$ sec. The value of the dielectric constant is 2.5–3.0. Conductivity losses for unfilled vulcanizates are insignificant (Fig. 1).

The introduction of filler substantially affects the dielectric losses of the vulcanizates. Along with a significant increase in $\operatorname{tg}\delta_{\max}$ and conductivity losses in filled vulcanizates, the appearance of losses has been observed

Fig. 2. Change in $\operatorname{tg}\delta_{\max}$ at 20 Hz of ethylene-propylene vulcanizates with ionization-aging time

in the low-temperature region, which are analogous to dipole-group losses and are due mainly to polar impurities. The activation energy of the filled vulcanizates is somewhat lower than that of the unfilled ones. The relaxation time decreased by an order of magnitude. Probably the filler plays the role of a lubricant, allowing the groups to relax more freely.

With increasing ionization-aging time, the magnitude of the maximum of the relaxation losses, $\operatorname{tg}\delta_{\max}$, which characterizes the number of groups participating in relaxation, increases (Fig. 2). In this case the changes in $\operatorname{tg}\delta_{\max}$ are most substantial for rubbers vulcanized with sulfur (ÉT-96, ÉT-196). For vulcanizates based on the same rubber but with dicumyl peroxide as the vulcanizing agent (ÉT-95, ÉT-195), a slight change in $\operatorname{tg}\delta_{\max}$ with ionization-aging time is characteristic.

The conductivity losses of all the vulcanizates studied increase with increasing ionization-aging time (Fig. 3). The conductivity losses in filled vulcanizates are much greater than in unfilled ones.

filled ones. Filling the peroxide vulcanizate leads to an increase in conduction losses with increasing time of ionization aging.

The activation energy in the region of dipole-segmental losses for the unfilled vulcanizates ET-95 and ET-96 changes almost not at all with increasing time of ionization aging.

Fig. 3. Change in \(\tan \delta\) at \(70^\circ\)C and 20 Hz for ethylene-propylene vulcanizates with ionization aging time

If, for the sulfur vulcanizate ET-196, filling does not affect the change in activation energy during the aging process, then the activation energy of the peroxide vulcanizate ET-195 increases.

Fig. 4. Dependence of the parameter \(\alpha\) on temperature for vulcanizates ET-96 (a) and ET-95 (b) at different ionization aging times

To assess the nature of the relaxing groups, Cole–Cole diagrams were constructed from the temperature–frequency dependences of $\operatorname{tg}\delta$ and $\varepsilon$ of the vulcanizates; from these diagrams the distribution parameter of the relaxation-time spectrum, $\alpha$, was determined. As can be seen from Fig. 4, an increase in the relaxation-time spectrum with the duration of ionization aging is characteristic of the vulcanizates, to a greater extent for the sulfur vulcanizate.

The results of the studies carried out show that dielectric losses and their changes over the course of ionization aging depend to a considerable degree on the type of vulcanizing group and filler.

REFERENCES

1. B. I. Vorozhtsov. Measurement of the dielectric characteristics of electrical insulating materials. Pribory i tekhnika eksperimenta, 1959, No. 1.

2. B. I. Sazhin. Electrical Properties of Polymers. “Khimiya,” 1970.

3. R. Houwink, A. Staverman. Chemistry and Technology of Polymers. Vol. I, “Khimiya,” 1965.