LONG TERM STRESS RELAXATION BEHAVIOUR OF
POLYETHYLENE TEREPHTHALATE SUBJECTED TO THERMALLY
ACTIVATED RESTRAINED SHRINKAGE
T. Hazelwood1 , A. D. Jefferson1 and R. J. Lark1
1 Cardiff School of Engineering, Cardiff University, Queen’s Buildings, The Parade, CF24
3AA, Cardiff, Wales, United Kingdom – e-mail: hazelwoodt@cardiff.ac.uk;
jeffersonad@cardiff.ac.uk; lark@Cardiff.ac.uk
Keywords: concrete, self-healing, shape memory polymer, thermo-mechanical behaviour, long-term stress relaxation.
ABSTRACT
Research has been carried out with the aim of better understanding the relevant properties of materials to be used in a new self-healing cementitious composite material system. The system consists of shape memory polymer tendons embedded within a cementitious matrix, upon heat activation these tendons undergo a shrinkage process which closes any cracks present. After this heat activation, autogenous healing of the cracks is enhanced by the ongoing application of pressure to the cracked faces.
In a previous study the build-up of stress in a restrained poly(ethylene terephthalate) specimen upon heat activation was investigated. In the current study, the long term stress relaxation behaviour of such a restrained specimen has been investigated so that its potential for use in the material system described above can be better understood. Both experimental and numerical approaches have been undertaken. A specimen of poly(ethylene terephthalate) was restrained against longitudinal shrinkage and a heat of 90oC applied for a duration of 10 minutes to simulate the
conditions it would be subject to within the system. The stress in the specimen was continuously monitored during heat activation, cooling, and for approximately 6 months after this. This data was used to quantify the stress relaxation of the specimen.
The stress relaxation over the period studied was found to be very low; the peak stress was 32.80MPa, this fell to a minimum value of 30.84MPa after a period of approximately 97 days had elapsed. These low levels of stress relaxation are beneficial to the system since a larger shrinkage force will be applied for a longer period of time, which should also lead to improved autogenous healing of cracks.
1. INTRODUCTION
This study represents an extension of previous research carried out by this research group [1] into the behaviour of SMP tendons in the context of a new self-healing composite materials system. This system has been named LatConX and has been previously described by Jefferson et al [2].
Dunn et al [1] investigated stress development in pre-drawn PET during restrained shrinkage due to temperature activation using both experimental and numerical techniques. In the context of the LatConX system, the long term behaviour after stress development is also of significant importance as this will have an effect on the extent of any autogenous healing within the cementitious matrix. In light of this, the
current publication continues the work of Dunn et al [1] to include this long term aspect of the polymer’s behaviour.
2. MATERIALS AND METHODS
All experiments have been carried out using the drawn polymeric material Aerovac Shrink Tite, in 32 mm × 0.046 mm tape form [3].
All test specimens consisted of 25 strips of the polymer tape approximately 400 mm long, clamped at both ends between grips consisting of two flat metal plates measuring approximately 60 mm × 23 mm × 2 mm. In reality it was impossible to have each specimen exactly 400 mm long due to the intricacies involved with assembling the multiple strips in the grips; the length of the specimens actually ranged from 400 mm to 402 mm.
Test specimens were held in the grips in a position so that they were just taut, a heat of 90oC was then applied to the specimen for a period of ten minutes, thus activating
the shrinkage behaviour of the drawn polymer, and inducing a stress within the specimen. The stress and the ambient temperature were monitored and logged over time by a 2.5 kN load cell and 12 thermocouples respectively. The tests were carried out in a custom built rig consisting of an Instron oven fixed within a Mechtric reaction frame; fixings were constructed to connect the grips holding the specimen to the top and bottom of the reaction frame with the specimen running through the middle of the oven.
3. RESULTS AND DISCUSSION
Five of these tests have been carried out; a typical set of results is presented in Figure 1. From this it can be seen that the stress reaches a peak of 32.8 MPa at the start of the test before reducing over a period of approximately 100 days to an average plateau value 31.48 MPa. This value is an average as the stress continues to fluctuate between approximately 32 MPa and 31 MPa once the plateau is reached.
Figure 1: Stress vs time for heat activated stress relaxation test
The peak stress reached in the five tests has been observed to vary from a minimum of 25.95 MPa to a maximum of 32.80 MPa, one theory is that the value of the peak stress is closely linked to the age of the material as the locked in stress is thought to gradually release over time. The length of time that the material relaxed over was also seen to vary, the maximum period was approximately 100 days, and the minimum was approximately 1 day; this tended to be a shorter time in the older material. Finally the stress that the relaxation plateaued at also varied from a minimum of 96.4 % of the peak stress to a maximum of 98.6 % of the minimum. Due to the long time period required to carry out a test such as the one above, it was not possible to continue every test for this period. However four other shorter term tests were undertaken to confirm that a similar early trend was seen, these tests were found to show close agreement; they have been normalised to the peak stress and averaged to produce the results shown in Figure 2.
Figure 2: Normalised average stress vs time for 4 different heat activated stress relaxation tests
There was a concern regarding the tensile strength of the pre-drawn PET and whether any extension due to a crack crossing the material would induce a tensile stress close to failure. The authors have not carried out the relevant research themselves, however reference to other sources has yielded the following relevant values; the elongation at break ranges from 30 – 300 %, and the tensile strength at break ranges from 47 – 70 MPa [4], based on these figures we can be confident that the proposed application is comfortably within the safe operating limits of the material.
The set of experimental results presented above have good implications for the use of this material in the proposed LatConX system as they show very limited stress relaxation, so the force applied to the cracked faces of the cementitious material would remain high for an extended period of time giving more effective autogenous healing of the crack for a longer time.
4. CONCLUSION
An experimental study into the long term stress relaxation behaviour of the shape memory polymer Poly(ethylene terephthalate) has been presented here. Stress relaxation of this material has been monitored for periods of up to 6 months, following activation of the shape memory recovery process under restrained conditions. This study has been carried out in the context of the new LatConX material system. The findings of this study are considered to have good implications for the use of the material in the proposed LatConX system.
Further investigations into this and other related viscoelastic behaviour of this material have been undertaken in conjunction with the development of a numerical model capable of simulating this behaviour. The authors hope to present this further work in the form of a journal article in the near future.
REFERENCES
[1] S. Dunn, A. Jefferson, R. Lark, B. Isaacs, Shrinkage behaviour of poly(ethylene terephthalate) for a new cementitious-shrinkable polymer material system. Journal of Applied Polymer Science 120 (2011) 2516-2526.
[2] A. Jefferson, C. Joseph, R. Lark, B. Isaacs, S. Dunn, B. Weager, A new system for crack closure of cementitious materials using shrinkable polymers. Cement and Concrete Research 40 (2010) 795-801
[3] http://www.aerovac.com/ [Accessed: 17th May 2012]
[4] http://www.maropolymeronline.com/Properties/PET.asp#Tensile Strength at Break [Accessed 8th March 2013]
