Space charge characteristics in polymer materials and their relation with ageing

Abstract

Space charge formation in polymeric materials can cause serious concerns for design engineers as the electric field may severely be distorted, leading to part of the material being overstressed. This may give rise to accelerated material degradation and possibly premature failure in the worst conditions.

As space charge dynamics in polymeric materials can be directly related to the changes in their composition, structure and morphology, a quantitative model is highly desirable for a better understanding of charge injection, trapping and detrapping in the polymeric materials.

This dissertation reviews some well-known ageing models of insulating materials and shows their related derivation procedures. Most of these models treated ageing phenomenon as a chemical kinetics process but explained it using different approaches microscopically. Moreover, these models hold different perspectives on space charge effect on ageing. According to Crine’s model and Lewis’ model, space charge is just one of the consequences of material degradation whereas in the model established by Dissado-Montanari-Mazzanti, it is intended to prove that space charge can also act as a cause of ageing.

In the present research, the space charge characteristics of both normal and gamma-irradiated low density polyethylene (LDPE) films during voltage-off periods have been investigated by PEA technique. Irradiation treatment can alter the physical structures of materials and bring chemical changes into normal samples. By analyzing those parameters estimated by Chen’s trapping/detrapping model, it shows the increment in trap sites and changes in trap depths, indicating that chemical reactions brought by irradiation have some changes in trapping characteristics of LDPE.

However, there are some drawbacks in the previous trapping/detrapping model. By considering Schottky injection process and a modified Pool-Frenkel conduction mechanism, an improved trapping/detrapping model is established based on Chen’s model. Such model can not only estimate the trapping parameters of polymeric materials like previous models but also evaluate some new parameters like mobile charge escape rate coefficient and injection barrier for both electrons and holes.

By using such a model, trapping parameters of aged samples under nine different ageing conditions are estimated. Moreover, breakdown tests were performed on these samples. An increment in breakdown strength can be found for samples from the early stage of ageing. This can be explained by the surface oxidation process which concentrates the charges near the both electrodes and therefore reduce the field between metal-insulator interface. When the ageing condition reach certain circumstance (applied field and temperature sufficiently high), the breakdown strength begin to decrease due the generation of traps in the sample bulk.

With the experimental data of dc breakdown tests for different aged samples, a new model to depict the relationship between breakdown strength and ageing conditions (electric field, temperature and ageing time) has been built based on Simoni’s model. Such a new model mimic the initial rise of the breakdown strength by introducing an inverse reaction rate during the ageing. Moreover, through the implementation of such a model, the electron trap density can be used to quantitatively reflect the stage of ageing. Several XLPE extruded cable sections (12, 11, 10 and 8-year) retired from HVAC service are peeled into films by using the microtome. Preliminary experimental works were done by using the 10-year cable section: the heterocharges accumulation near both electrodes can be observed. To see homocharge injection, degas treatment is thereby required to eliminate the byproducts, which can give rise to ionic dissociation process inside the sample. Thereafter, based on the improved trapping/detrapping model, trapping parameters of the XLPE peelings are found for 12, 11 and 8-year operated cable sections. The estimated injection barrier, trap density and overall trap depth can be clearly related to the dc breakdown performance of XLPE samples. Moreover, from the estimated trapping parameters and the breakdown tests, they both indicate that the inner layer is most seriously aged part across the radial direction of the cable section.

The threshold field (dc) of charge injection can be evaluated through three approaches for LDPE materials. Subtraction method on voltage-on measurement data and direct measurement on charges with multiple voltage-off tests can both give values of the threshold field. And the subtraction method is more sensitive way to obtain a relatively lower threshold field at 8 kV/mm. Another method taking account of maximum peak heights at both electrodes is not suitable to measure threshold field for total charge injection but can give an inception field of injection respectively for positive or negative charges.

Taking advantage of plasma technology using gas CF4, fluorination treatments are performed on LDPE samples to produce fluorinated samples at two different discharge voltage levels and two different exposure times. For both normal and fluorinated LDPE samples, the space charge measurements are performed.
The results indicate that the injected homocharges can be effectively suppressed with fluorination process but the heterocharges which come from ionization of the lower molecular weight specifies become prominent in the sample. The sufficient fluorination on normal LDPE samples can lower the dc conductivity but when the fluorination is not enough, the heterocharges can lead to even higher dc conductivity than the original one.

More information

Identifiers

Catalogue record

Export record