Static and Dynamic Thermo Mechanical

Characterization of a Bio-Compatible Shape Memory

Polymer
Pauline Butaud, Anne Maynadier, Nicolas Brault, Vincent Placet, Morvan
Ouisse, Emmanuel Foltête, Cécile Rogueda-Berriet

To cite this version:

Pauline Butaud, Anne Maynadier, Nicolas Brault, Vincent Placet, Morvan Ouisse, et al.. Static and
Dynamic Thermo Mechanical Characterization of a Bio-Compatible Shape Memory Polymer. ICAST
2013, The 24th International Conference on Adaptive Structures Technologies, Jan 2013, Aruba,
Caribbean Netherlands. pp.1 - 10. �hal-00993450�

HAL Id: hal-00993450

https://hal.archives-ouvertes.fr/hal-00993450
Submitted on 20 May 2014

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est

archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.
 ICAST2013: 24th International Conference on
Adaptive Structures and Technologies
October 7-9, 2013, Aruba

ICAST2013 #paper number

Static and dynamic thermo mechanical characterization of a bio-compatible

shape-memory polymer

Pauline Butaud1*, Anne Maynadier2, Nicolas Brault, Vincent Placet1, Morvan Ouisse1,
Emmanuel Foltête1, Cécile Rogueda-Berriet1

1
FEMTO-ST Applied Mechanics Department, Besançon, FRANCE
2
LaMCoS Laboratoire de Mécanique des Contacts et des Structures, Lyon, FRANCE

Abstract

Shape memory polymers encounter a growing interest over the past ten years particularly because their
eventual bio-compatibility leads to many bio-medical applications. They also present many benefits for
the design of micro-adaptive systems for deployment or controlled damping materials. Indeed, the SMPs
are polymeric smart materials which have the remarkable ability to recover their primary shape from a
temporary one when submitted to an external stimulus.
The present study deals with the synthesis and the thermo-mechanical characterization of a thermally-
actuated SMP. The polymer considered hereafter is a chemically cross-linked thermoset. It is synthesized
via photo polymerization (UV curing) of the monomer tert-butyl acrylate (tBA) with the crosslinking
agent poly(ethylene glycol) dimethacrylate (PEGDMA) and the photoinitiator 2,2-dimethoxy-2-
phenylacetophenone (DMPA).
A mechanical characterization has been performed using three kinds of tests: quasi-static tensile tests,
tensile dynamic mechanical analysis (DMA) and modal tests. The Young’s modulus and the Poisson ratio
are determined at ambient temperature using the first technique. The DMA is used to determine the
evolution of viscoelastic properties as a function of the temperature and the frequency under harmonic
loading. The modal analysis is employed to identify the viscoelastic properties of the material at higher
frequency. A comparison of the results obtained by these three experimental methods highlights their
complementarity.

1 INTRODUCTION

Shape memory polymers (SMPs) belong to the class of smart materials. These materials have found
growing interest throughout the last few years. Indeed, in some aspects, SMPs are more efficient than the
well-known metallic shape memory alloy (SMA’s). SMPs particularly offer recoverable strain that can
reach 400% (to be compared to 8% for SMAs), light weight, easy manufacturing and cheapness compared
to SMA [1]. They can be used in a large variety of applications, e.g. actuators, electromechanical systems,
clothing manufacturing, morphing and deployable space applications, control of structures, self-healing,
biomedical devices and so on [2, 3]. The SMPs have the ability of changing their shape in response to an

*
[email protected]

1
ICAST2013: 24th International Conference on
Adaptive Structures and Technologies
October 7-9, 2013, Aruba

external stimulus [4], most typically thermal activation. When the SMP is heated above the glass transition
temperature , it is soft and rubbery and it is easy to change its shape. If the SMP is subsequently cooled
below , it retains the given shape (shape fixing characteristic). When heated again above , the material
autonomously returns to its original permanent shape [5]. The shape-memory effect is in principle a
behavior inherent to all polymers. However, polymers that exhibit truly exploitable shape-memory effect
must demonstrate a sharp transition temperature and a rubbery plateau, along with relatively large strain
capacity without local material damage [6]. The memory effect of SMP is quite depending on the level of
external constraint. For example if the stiffness of the constraining material is equal to the stiffness of the
shape memory polymer in the rubbery state, the polymer will only recover half of its total strain [7].
The increasing use of SMP for quasi-static and dynamic applications, under various temperature ranges,
has made necessary the characterization of these materials over wide frequency bands [8]. Three kinds of
tests have been performed in this study. First, quasi-static tensile test are used at ambient temperature and
at a controlled loading rate permit to identify the Young’s modulus. Then, a dynamic mechanical analysis
(DMA) is used to determine the evolution of viscoelastic properties as a function of the temperature and
loading frequency. Finally, a modal analysis has been performed to reach higher loading frequencies. In
this last technique, a finite element model is employed to identify the viscoelastic properties of the
material from the dynamic measurements. The main purpose of this work is to check the validity of the
time-temperature equivalence [9] obtained from the DMA measurements, and more exactly to check the
capacity of DMA on a restricted frequency range and at various temperature levels to predict accurately
the viscoelastic properties on a wider frequency range. The predicted viscoelastic properties are compared
to the measurements collected using quasi-static and modal test. This is an important query considering
that measurement combining several experimental methods is particularly time-consuming.

2 MATERIALS AND EXPERIMENTAL METHODS

2.1 Material and specimen preparation

As a representative thermally-actuated shape-memory polymer, a chemically-crosslinked thermoset

polymer recently studied by Yakacki et al. [10] and Srivastava et al [11] is chosen. Following the
procedure described by these authors, the SMP is synthesized by manually mixing 95 wt% of the
monomer tert-butyl acrylate (tBA) with 5 wt% of the crosslinking agent poly(ethylene glycol)
dimethacrylate (PEGDMA) (with typical molecular weight g/mol). The photoinitiator, 2,2-
dimethoxy-2-phenylacetophenone (DMPA), is added to the solution at a concentration of 0.5 wt% of the
total weight. The liquid mixture is then injected between two glass slides separated by a 3 mm spacer. The
polymerization is initiated by exposing the solution to UV light during 10 minutes and achieved by
heating the polymer at 90°C for 1 hour.
One of the obtained plates, measuring mm, was machined with a CNC machine. Dog-bone
shaped specimens were produced from the polymer plate for quasi-static tensile tests. The gauge length is
35 mm, and the rectangular cross section is . . mm. In the same plate, samples for DMA testing
were cut to . mm. Another plate was produced with the exact same protocol for the modal
analysis measurements.

2
 ICAST2013: 24th International Conference on
Adaptive Structures and Technologies
October 7-9, 2013, Aruba

2.2 Mechanical tests

2.2.1 Quasi-static tests

The quasi-static tensile test have been performed using a universal commercial testing machine (Instron
6025) with a capacity of ± 100kN. The specimen, clamped using wedge action grips, was subjected to a
displacement control feedback at a rate of 1 mm/min. The test was performed at ambient temperature
21°C (± 1°C). The engineering longitudinal strain was measured by a laser extensometer (EIR LE05) with
a displacement accuracy of approximately 1 µm. The gauge part of the specimen for the strain
measurement is 33 mm in length. The longitudinal Young’s modulus was determined by the slope of the
stress-strain curve at 0.1% of engineering strain (tangent modulus technique).

2.2.2 Dynamic mechanical analysis

Viscoelastic properties (storage modulus, loss modulus and loss factor) are measured using a Bose
Electroforce 3200 apparatus every 5°C in isothermal conditions. Temperature varies between -28°C and
30°C, the frequency of the solicitation from 0.1 Hz to 10 Hz. The temperature is stabilised during 2
minutes before measurements to ensure a homogeneous temperature inside the specimen. The heating rate
is about 0.1°C/sec between each plateau temperature. The temperature is measured using a thermocouple
positioned in a reference sample put in the heating chamber close to the tested specimen.
The compliance of the test set-up was taken into account for the determination of the viscoelastic
properties.
A sinusoidal tensile load was applied on the sample with a mean amplitude of 2 N and a peak-to-peak
amplitude of 3 N, in order to test the specimen in the linear viscoelastic range. Loading-unloading tests
were also made on this equipment, at temperature below and above the glass transition. A tensile load
from 0.1 N to 5 N was applied on the sample with a constant rate of 0.2 N/s, between -15 °C and 30 °C
every 15 °C. After the glass transition, between 70 °C and 90 °C (every 10°C), a tensile ramp of 0.05 N to
0.15 N was applied on the sample at a rate of 0.005 N/s. At high temperature, a more sensitive load cell
was used considering the high decrease in rigidity of the tBA/PEGDMA and to ensure accurate
measurements. It should be emphasized that at this temperature level the specimen stiffness is only of 0.1
N.mm-1.

2.2.3 Modal analysis

The studied structure is a tBA/PEGDMA rectangular plate with dimensions . . . mm (±
0.1 mm). The plate is suspended in order to realize free-free conditions. Contactless actuators and sensors
are used. The external force is applied thanks to a voice-coil actuator with a permanent magnet glued in a
corner of the plate. This assembly is placed in a thermal chamber with a temperature control between 22
and 36 °C every 7°C (± 2°C). A broadband random excitation is applied between 150 and 4000 Hz. The
plate response is measured by using a laser vibrometer (Polytec OFV-505) successively focused on 28
reflecting stickers spread on the plate. The modal frequencies, the modal damping and the mode shapes
are then identified by using Modan© (homemade modal analysis software). At the same time, a finite
element model is established on PatranTM2012 under the hypothesis that the material is homogeneous and
isotropic, with a Poisson’s ratio . (determined from quasi-static tests), a mass density
/ (determined by measuring the plate’s mass and volume) and a presumed Young’s modulus
. The magnet is included in the numerical model ( , . ,
/ ). Finally a model-test correlation is performed on AESOP software and a model updating
procedure is conducted to obtain the mechanicals parameters of the tBA/PEGDMA at the modal
frequencies.

2.2.4 Infrared thermography

Thermal field was monitored by infrared thermography using a CEdip Jade III MWIR camera. It provides
320 240 pixels images. The acquisition frequency is set to 25 Hz and the integration time is 430 µs. The

3
ICAST2013: 24th International Conference on
Adaptive Structures and Technologies
October 7-9, 2013, Aruba

measuring range was 5-90°C. The measurements were performed during quasi-static tests and DMA to
visualize the possible overheating in the sample.

2.2.5 Kinematic fields measurement

An optical camera (Ueye 1Mpixel 8bit) was used to assess the displacements fields in the gauge part of
the specimen using digital image correlation (DIC). The camera was equipped of a telecentric lens (0.20x
Techspec Silver series telecentric lens - 63073 - Edmund Optics) that prevents measurements from optical
distortions. It also ensures a large depth of field (of approximately 6 mm) which guarantees the sharpness
of images even in case of necking. Image acquisition was triggered using a Labview application at a
frequency of 1 frame per sec. DIC was performed with a free home-made global correlation code,
implemented in Octave and described by Maynadier et al. [12]. Quadratic elements were chosen for
meshing the area of interest. The strain fields were computed from the displacement fields by pixel-wise
difference gradient.

3 RESULTS

3.1 Quasi-static characterization

Figure 1 shows the stress-strain curve obtained from the quasi-static test. The apparent Young’s modulus
is approximately MPa, the ultimate stress (32 MPa) is reached at a strain of 1.8% and the elongation
at rupture is about %.
40

30
Stress [MPa]

20

10

0
0 10 20 30 40 50 60
Strain [%]
Figure 1. Stress – strain curves in tensile test at 1 mm/s, for
tBA/PEGDMA, at ambient temperature

Infrared thermography was used for visualization of the temperature field at the surface of the specimen
during this test. As shown in figure 2, the thermo-mechanical behavior of the tBA/PEGDMA in quasi-
static is quite similar to the classical polymer’s behavior: there is a cooling during the elastic phase, then a
warming during the viscoelastic phase. Heat release is localized first in the necking zone and then heat
propagates along the sample. Finally, rupture occurs with high heat release localized in a very sharp area
around the crack tip.

4
 ICAST2013: 24th International Conference on
Adaptive Structures and Technologies
October 7-9, 2013, Aruba

Figure 2. Temperature variation on the centerline of the sample

compare to the ambient temperature at 1 mm/s
The Poisson’s ratio was determined using the strain fields determined using DIC. The value is
approximately 0.37 (Figure 3).

0.4
-(tranverse strain) [%]

0.3

0.2

0.1 Experimental data

Polynomial curve fitting
0
0 0.2 0.4 0.6 0.8 1 1.2
longitudinal strain [%]
Figure 3. Identification of the Poisson’s ratio around 0.37 at 1 mm/s

Figure 4 shows the high influence of temperature on the Young’s modulus. Its value is almost constant,
around 2060 MPa from -20°C to 40°C and also constant but 2500 times lower (0.8 MPa) above 60°C.
This abrupt change in the elastic modulus reveals a phase transition between 40°C and 60°C.
4
10
Young's modulus [MPa]

2
10

0
10

-2
10
-20 0 20 40 60 80 100
Temperature [°C]
Figure 4. Stress – strain curves in tensile test at several temperature

5
ICAST2013: 24th International Conference on
Adaptive Structures and Technologies
October 7-9, 2013, Aruba

The temperature variation registered by infrared thermography was not significant during this kind of
tests; the variation level is lower than the noise level. The hypothesis of isothermal test is verified, there is
not self-heating of the sample under mechanical solicitation.

3.2 Dynamic mechanical analysis

The results of the DMA tests are shown in Figure 5. The storage modulus decreases with the temperature
and increases with the frequency. On the contrary, the loss factor increases with the temperature and
decreases with the frequency. These trends are coherent with the classical behavior of a polymer without
measurements on the transition area. Moreover the values of ’ and tan at 1 Hz are consistent with the
results on the tBA/PEGDMA of Ortega et al. [13].

2000 -28°C
-23°C
E' [MPa]

-19°C
1500
-14°C
-7°C
1000 -1 0 1
-3°C
10 10 10
2°C
0.2
7°C
0.15 11°C
tan 

0.1 16°C
20°C
0.05
25°C
0 -1 0 1
30°C
10 10 10
Frequency [Hz]

Figure 5. E’ and tan delta according to the frequency and for several temperatures

Curves of E’ vs. frequency at one temperature can be shifted to overlap with adjacent curves (Figure 5).
The time-temperature superposition shift factors used for E’ must be the same for tan to get time-
temperature equivalence [14]. In our case, the shift factors are obtained through an optimization
procedure (classical least square method) and are equivalent for both E' and tan , therefore the time-
temperature equivalence is validated in a specified temperature and frequency ranges for the
tBA/PEGDMA. The master curves of the storage modulus and tan are given in Figure 6.

6
 ICAST2013: 24th International Conference on
Adaptive Structures and Technologies
October 7-9, 2013, Aruba

1700

1600
E' [MPa]

1500

1400
11°C
1300 -3 -2 -1 0 1 2 3 4 16°C
10 10 10 10 10 10 10 10
0.2 20°C
25°C
0.15
30°C
tan 

0.1

0.05

0 -3 -2 -1 0 1 2 3 4
10 10 10 10 10 10 10 10
f.a [Hz]
Figure 6. Master curves of E’ and tan according to the reduced frequency with
a reference temperature of 20°C

3.3 Modal analysis

Depending on the temperature, 11 to 13 modes are identified after the model-test correlation (Figure 7).
(b) 1
7
9
0.8
Finite element model

1 11
10
13
Amplitude

-1 0.6
10 15
-3
17
10 0.4
19
-5 21
10
0 1000 2000 3000 4000 5000 0.2
23
Frequency [Hz]
25
0
1 3 5 7 9 11 13 15 17 19
Experimental

Figure 7. (a) FRF and (b) MAC Matrix for 22°C experimental data

Just as for DMA, the obtained Young's modulus and tan of the tBA/PEGDMA depend on the
temperature and the frequency as shown in Figure 8.

7
ICAST2013: 24th International Conference on
Adaptive Structures and Technologies
October 7-9, 2013, Aruba

Young's modulus [MPa]

4000

3000

2000
22°C
1000 29°C
36°C
0
2 3 4
10 10 10

0.04

0.03
tan 

0.02

0.01

0
2 3 4
10 10 10
Frequency [Hz]

Figure 8. E’ and tan according to the frequency and for several temperatures

As expected, the loss factor increases with the temperature and decreases with the frequency. Concerning
the storage modulus, it decreases as planned with the temperature but its variation with the frequency is
unusual, as it decreases too. This trend is not well understood yet, even if some hypotheses are considered.
For example, the Poisson’s ratio used in the finite element model was determined in quasi-static regime
whereas it could fluctuate with the frequency like the storage modulus. Other hypotheses like the isotropic
character of the material are currently under investigation.

4 DISCUSSION

This study intends to compare the master curves obtained with the DMA and the results collected by the
modal analysis and the quasi-static tests. Based on the time-temperature equivalence determined by DMA,
the storage modulus predicted by this equivalence should be equal with the quasi-static and modal
modulus, likewise for the loss factor. The DMA master curves and the results of quasi-static tests and
modal analysis in temperature beforehand translated thanks to the shift factor are plotted on a same
graphic (Figure 9). To define the frequency of the quasi-static test, the loading-unloading test period was
taken in account, thus the frequency of this test at low temperature is 0.02 Hz.

8
 ICAST2013: 24th International Conference on
Adaptive Structures and Technologies
October 7-9, 2013, Aruba

3000

2000
E' [MPa]

1000
Master curve DMA
0 -3 Modal tests
-2 -1 0 1 2 3 4
10 10 10 10 10 10 Quasi-static tests
10 10

0.2

0.15
tan 

0.1

0.05

0 -3 -2 -1 0 1 2 3 4
10 10 10 10 10 10 10 10
f.a [Hz]

Figure 9. Comparison of experimental methods on E’ and tan

The order of magnitude of E’ at low frequency is respected. The time-temperature equivalence data are
consistent with the quasi-static results. At high frequency, there is a gap of about 50 % between the
storage modulus prediction and the modal results. This is not observed for the loss factor, for which the
comparison is quite conclusive.
There are some options to explain this gap between the modal tests and the master curve. First of all, the
hypothesis of isothermal test could be unverified i.e. the sample could undergoes self-heating under
mechanical solicitation. This possibility was eliminated for DMA tests with infrared thermography but not
for the modal analysis. Secondly, as it was mentioned previously, the Poisson’s ratio could be not
correctly defined in the finite element model. Finally, the most important distinction between modal
analysis and the two others experimental methods is the kind of solicitations. Indeed, whether for quasi-
static or DMA, the test is in tensile on a beam sample, contrary to the modal analysis for which the test is
in bending on a plate and for smaller strain amplitude. It therefore appears necessary to study the behavior
of the tBA/PEGDMA in tensile and compression, and check whether a distinction. Lastly, a post-treatment
is required for the modal data thanks to a finite element model, contrary to DMA and quasi-static data
which are used unaltered.

5 CONCLUSION

The objective of this study was to highlight the time-temperature equivalence by comparing three
experimental methods allowing the identification of the storage modulus over frequency and temperature.
This equivalence has been checked on a sample tested with the DMA. Quasi-static tests and modal
analysis have been carried out to explore a wider frequency band. Even though the results are rather
consistent, the time-temperature equivalence is not yet established. Further investigations are needed to
clarify the tBA/PEGDMA behavior. The correlation between quasi-static, DMA and modal analysis
remains to be carried out.

9
ICAST2013: 24th International Conference on
Adaptive Structures and Technologies
October 7-9, 2013, Aruba

References

1. Pretsch Thorsten, “Review on the functional determinants and durability of shape memory polymers”, Polymers, Molecular
Diversity Preservation International, 2010, Vol. 2, pp. 120-158
2. Behl, M. and Lendlein, A., “Shape-memory polymers”, Materials Today, 2007, Vol. 10, pp. 20-28
3. Dietsch, B. and Tong, T., “A review-: Features and benefits of shape memory polymers (SMPs)”, Journal of advanced
materials, 2007, Vol.39, pp. 3-12
4. Ratna D. and Karger-Kocsis J., “Recent advances in shape memory polymers and composites: a review”, Journal of
Materials Science, 2008, Vol. 43, pp. 254-269
5. S.A. Wilson, R.P.J. Jourdain, Q. Zhang, R.A. Dorey, C.R. Bowen, M. Willander, Q.U. Wahab, S.M. Al-hilli, O. Nur, E.
Quandt, et al, “New materials for micro-scale sensors and actuators: An engineering review”, Materials Science and
Engineering: R: Reports, 2007, Vol. 56, pp. 1-129.
6. B. Sillion, “Shape memory polymers”, Acta Chim, 2002, Vol. 3, pp. 182-188.
7. Liu, Yiping and Gall, Ken and Dunn, Martin L and Greenberg, Alan R and Diani, Julie, “Thermomechanics of shape
memory polymers: uniaxial experiments and constitutive modeling”, International Journal of Plasticity, 2006, Vol. 22, pp.
279-313.
8. A. Lendlein and S. Kelch, “Shape-memory polymers”, Angewandte Chemie International Edition, 2002, Vol. 41, pp. 2034–
2057.
9. N. Okubo, “Preparation of Master Curves by Dynamic Viscoelastic Measurements”, SII NanoTechnology Inc., 1990, Vol. 6.
10. Yakacki, Christopher Michael and Shandas, Robin and Lanning, Craig and Rech, Bryan and Eckstein, Alex and Gall, Ken,
“Unconstrained recovery characterization of shape-memory polymer networks for cardiovascular applications, Biomaterials,
2007, Vol. 28, pp. 2255-2263.
11. Srivastava, Vikas and Chester, Shawn A and Anand, Lallit, “Thermally actuated shape-memory polymers: Experiments,
theory, and numerical simulations, Journal of the Mechanics and Physics of Solids, 2010, Vol. 58, pp. 1100-1124.
12. A. Maynadier, S. Carbillet, V. Placet, F. Amiot, A. Seror, A. Longuet, C. Mary, “Fatigue behaviour of D.A. INCONEL 718
alloy under tension-torsion loading”. In PhotoMechanics, Montpellier, 2013
13. Ortega, Alicia M and Kasprzak, Scott E and Yakacki, Christopher M and Diani, Julie and Greenberg, Alan R and Gall, Ken,
“Structure--property relationships in photopolymerizable polymer networks: Effect of composition on the crosslinked
structure and resulting thermomechanical properties of a (meth) acrylate-based system”, Journal of Applied Polymer Science,
2008, Vol. 110, pp. 1559-1572.
14. Dealy, John and Plazek, Don, “Time-temperature superposition—a user’s guide”, Rheol. Bull, 2009, Vol. 78, pp. 16-31

10