Williams, A.T. & Simmons, S.L.
Fig. 3. LDPE degradation trial (4 months): Bank samples (lower set); Buried samples (upper set); Tensile strength (bar represent 95% confidence limits).
I – Io I
Where: Ep = elongation at break; I = elongation in gauge length at break (mm); Io = original gauge length (mm). Ep can be conveniently expressed as a percent- age.
Elongation retention is largely dependent on poly- mer type and composition. “Relative to tensile strength, the elongation appears more sensitive to changes occur- ring during photodegradation” (Wypych 1990, p. 251). Wypych (1990) also noted that no correlation existed between percentage elongation and tensile strength, elon- gation being more indicative of changes in the amor- phous phase, whilst tensile strength was more depend- ent on crystalline polymer regions.
Other parameters such as break factor – the maxi- mum load divided by the minimum width– and percent- age elongation at yield – the elongation produced in gauge length of the test piece at yield stress, which is first marked as an inflexion of the stress/strain curve, expressed as a percentage of the original gauge length - were also calculated., They were very similar to the more common measurements of tensile strength and elongation at break respectively. Sample means were calculated for each parameter, together with two-sided confidence limits of the mean. The confidence limits “give an interval in which the true mean is expected to lie with specified confidence” (Gilbert 1987, p. 137) in this case 95 %.
Results of tensile strength and percentage elonga- tion at break indicated that some degradation had oc- curred, with significant decreases in values occurring between control samples at point 0, and those tested later (Fig. 3). Subsequent weeks showed little further degradation. Confidence limits were predictably greater for exposed samples than controls, as a result of mate- rial variations brought about through exposure. Al- though mean values differed from the second to the sixth week, at all times confidence limits overlapped, making predictions of further degradation inconclu- sive. The slight rise in tensile strength and elongation values for suspended samples after an extended expo- sure period, raised the possibility of differing degrada- tion rates resulting from variations in stranding posi- tion. Consideration was given to the possibility that bank-side samples may have been subjected to greater physical weathering from contact with earth and veg- etation on the bank. Physical abrasion in this manner could be the cause of flaws in the material, resulting in a lowering of tensile property values.
It was noted during tensile testing that test pieces within samples showed a wide variation in their re- sponses. Even after exposure, some strips reacted in a similar manner to control samples, whilst others failed very prematurely. The premature failure seemed to ini- tiate from a defect within the strip. These two very different responses resulted in a very broad range of results for samples, as indicated in the confidence lim- its (Fig. 3).