The wear-out approach for predicting the remaining lifetime of materials

Failure models based on the Palmgren-Miner concept that material damage is cumulative have been derived and used mainly for fatigue life predictions f or metals and composite materials. We review the principles underlying such models and suggest ways in which they may be best applied to polymeric mat erials in temperature environments. We first outline expectations when poly mer degradation data can be rigorously time-temperature superposed over a g iven temperature range. For a step change in temperature after damage has o ccurred at an initial temperature in this range, we show that the remaining lifetime at the second temperature should be linearly related to the aging time prior to the step. This predicted linearity implies that it should be possible to estimate the remaining and therefore the service lifetime of p olymers by completing the aging at an accelerated temperature. We refer to this generic temperature-step method as the "wear-out" approach. We next ou tline the expectations for wear-out experiments when time-temperature super position is invalid. Experimental wear-out results are then analyzed for on e material where time-temperature superposition is valid and for another wh ere evidence suggests it is invalid. In analyzing the data, we introduce a procedure that we refer to as time-degradation superposition. This procedur e not only utilizes all of the experimental data instead of a single point from each data set, but also allows us to determine the importance of any " interaction effects". (C) 2000 Elsevier Science Ltd. All rights reserved.