Low-temperature dielectric relaxation in polyethylene

Abstract

The technique due to Vincett has been developed to extend previous measurements of the dielectric loss tangent of polyethylene at cryogenic temperatures upwards in frequency to 100 MHz. A second relaxation peak, centred at 4 MHz, found in oxidized polyethylene is both broader and larger than the well-documented peak at 3 kHz. The dielectric loss is changed by steaming in D 2 O and H 2 O vapour at elevated tempera­tures. In all cases the 3 kHz relaxation is removed. Steaming in D 2 O moves the 4 MHz peak to 30 kHz and subsequent steaming in H 2 O restores it, indicating that there is an interchange of labile hydrogen and deuterium atoms at the site of the dipoles. Subsequent drying in vacuo of polyethylene treated in D 2 O also partially restores the 4 MHz peak, suggesting a further secondary diffusion process. The magnitude of the loss in Rigidex 50 (high-density polyethylene) shows little temperature variation between 1.5 and 4.2 K, while the frequency of maximum loss varies linearly with temperature for the 30 kHz peak; there is, however, a small deviation from linearity at 1.5 K for the 4 MHz peak. The results are analysed in terms of phonon-assisted tunnelling of a particle between two nearly-equivalent potential wells, using exact solutions of Pauling’s cosine model. Tentative estimates of the parameters of the model indicate that the results can be explained by 180° rotation of the proton in a dipolar group of bond length 0.11 nm.

Low temperature dielectric loss spectra of polyolefins

“…The Vincett-Phillips relaxation [8,28] which reaches maximum strength early on, is due to a chemically and thermally unstable group whose occurrence and properties are recognisably those of a hydroperoxide group [16,34]. A peak similar to Carson's [29] resulted from adding a long-chain secondary alcohol to unoxidised polyethylene [35]. The group responsible for the third peak has not been identified, but polyethylenes containing longchain primary alcohols exhibited peaks [21] in roughly the same part of Fig.…”

“…A o i s related exclusively to V 2 and to a moment of inertia I, which is approximately the moment of the -OH group about its rotation axis. A o varies very roughly as exp -{a \JV 2 //#} where 2irfi is Planck's constant and a is a numerical coefficient of the order of one [29]. A o of phenol-OH and phenol-OD have been determined directly by microwave spectroscopy of the vapours and found to be 56 MHz and 220 kHz, respectively [26,30].…”

“…This is notably what happens when (antioxidant-free) polyethylene is allowed to oxidise in air. Study of the lowtemperature relaxations of oxidised polyethylene [28,29] in fact preceded study of the antioxidant-induced relaxations, but the responsible chemical groups were only recently identified [12,16,34,35]. Fig.…”

Dielectric loss of polyethylene below 4.2°K arising from proton tunneling

“…(5), we must first calculate 2A0, the energy splitting of the ground state for the double-minimum potential in Figure 9. (9) where fm are angles at the minima [see Fig. 7(a)].…”