Abstract
This paper investigates the suitability of the isothermal linear viscoelastic framework to describe the behavior of polymers observed during DMTA tests. A good interpretation of these tests is important because, in practice, they are used to construct master curves using the time-temperature superposition principle at small strain. These curves are then considered to predict the material behavior under experimentally unreachable thermal and/or loading frequency conditions. Currently, the DMTA protocol neglects the temperature variations induced by the deformation of polymers. We wonder if these temperature variations can have an influence on the measurement of dynamic moduli. To answer this question, quantitative infrared techniques were developed and used to assess small temperature variations of samples undergoing cyclic loadings during mechanical spectrometry tests. Thermal and mechanical data were used to quantify the viscous dissipated and the thermoelastic coupling energies that can be both associated with the hysteretic stress-strain response of polymers. Energy balances were then performed to quantify the relative importance of dissipative and thermoelastic coupling heat sources. From the energy standpoint, it is found that the thermoelastic energy rate was dozens of times higher than the dissipation. Especially at low frequencies, thermoelastic effects can have a greater influence on the loss modulus value than viscosity.