I started working on the shape memory property of polymers at CU Boulder thanks to Professor Ken Gall who introduced me to the topic. Focusing on amorphous polymer networks, with several co-workers we have been working on the characterization and the modeling of this property. This post is about the experimental characterization of the shape memory property with respect to time and temperature. A second post shows more experimental results on strain recovery and force recovery measured with uniaxial tension tests. A last post will present a model based on the time-temperature superposition property and the viscoelasticity of these materials that proves that the shape memory property is the mere result of these two intrinsic properties to amorphous networks.
First let us define the shape memory property for polymers. It is the material ability 1) to retain a temporary shape that has been applied at a higher temperature thanks to some mechanical loading and 2) to recover its original shape by a mere temperature change. For amorphous polymer networks, this property results from a phase change from the glassy to the rubbery states. At high temperatures, polymer networks are in the rubbery state. They can undergo large strain. Due to the network feature, the applied strain is reversible. At low temperatures, these materials are in the glassy state, the polymer chains mobility is significantly reduced. Therefore when submitting a polymer network to large strains at high temperatures and maintaining the stress while cooling the material to reach its glassy state, one may apply a temporary shape and store it by maintaining the material at low temperatures. By merely heating the polymer, the material state changing from glassy to rubbery, the chain mobility is activated and the polymer recovers its original shape.
This property has been illustrated in several contributions of the literature and online. For quantitative data, uniaxial tension is usually favored since strain and stress are easily recorded. For large deformation, qualitative data are usually presented in terms of good or not so good recovery during bending or torsion tests. In order to study quantitative shape recovery involving large deformation, it was decided to apply some torsion to slender polymer specimens. For this purpose a device was built (Fig. 1) that allowed applying a torsion angle and maintaining it while cooling the specimen. The thermomechanical shape memory cycle consists in (1) heating the sample, (2) apply a torsion angle (usually 360 degrees) (3) maintaining the torsion angle while cooling the temperature (3), releasing the stress and finally heating the sample (4) and measuring the angle recovery with respect to time and temperature (Fig. 2). The test aimed at measuring the dependence to time and temperature of the shape recovery. Note that amorphous polymer networks are known to show excellent temporary shape fixure and shape recovery.
Fig. 2. Thermomechanical shape storage and recovery cycle.
Tests were applied to an epoxy network that had a convenient glass transition temperature around 50 °C. Excellent angle fixture larger than 99% was recorded once stress was release. In order to record the torsion angle during heating, two dots were painted on the edge of the specimen and the mirror reflexion of the sample edge was recorded thanks to a camera (Fig. 3).
Fig. 3. Click on the picture to see a video of the torsion angle recovery during heating.
The angle recovery with respect to the temperature heating ramp has been recorded (Fig. 4). It shows that the shape recovery is not only a function of temperature but also a function of time. In order to go further sample angle recovery was recorded with respect to time for a given temperature in the glass transition temperature rage (Fig. 4).
Fig. 4. Top: torsion recovery with respect to tempearture for three different heating ramp. Bottom: Torsion recovery with respect to time for a temperature stop at 42 °C within the glass transition temperature range.
This work was co-authored by Dr. Pierre Gilormini, Dr. Ingrid Rousseau and Master student Carole Frédy.
If interested you may find more information in the following papers:
J. Diani, C. Fredy, P. Gilormini, Y. Merckel, G. Regnier, I. Rousseau, 2011. A torsion test for the study of the large deformation recovery of shape memory polymers, Polymer Testing, 30, 335-341.
J. Diani, P. Gilormini, C. Frédy, I. Rousseau, 2012. Predicting thermal shape memory of crosslinked network polymers from linear viscoelasticity, International Journal of Solids and Structures, 49, 793-799.