polymers

Article
Shape Memory Behavior of PET Foams
Loredana Santo *, Denise Bellisario ID
and Fabrizio Quadrini
Department of Industrial Engineering, University of Rome “Tor Vergata”, Via del Politecnico 1,
00133 Rome, Italy; denise.bellisario@uniroma2.it (D.B.); fabrizio.quadrini@uniroma2.it (F.Q.)
* Correspondence: loredana.santo@uniroma2.it

Received: 14 December 2017; Accepted: 23 January 2018; Published: 25 January 2018

Abstract: Shape memory properties of PET (polyethylene-terephthalate) foams have been evaluated
for two different foam densities. Samples were subjected to multiple memory-recovery cycles
along three different directions to measure the effect of foam anisotropy on static mechanical and
shape memory properties. The memory cycle was performed by uniaxial compression tests at
room temperature. Despite these severe conditions, PET foams demonstrated very good shape
memory behavior with shape recovery always higher than 90%. Due to cycling, the mechanical
performance of foam samples is partially reduced, mainly along the extrusion direction of the
foam panels. Despite this loss of static performance, shape memory properties are only partially
affected by thermo-mechanical cycles. The maximum reduction is 10% for shape fixity and 3% for
shape recovery. The experimental results are particularly interesting considering that compression
tests were undertaken at room temperature. Indeed, PET foams seem to be optimal candidates for
self-repairing structures.

Keywords: shape memory behavior; PET foam; compression mode; shape recovery

1. Introduction
Shape memory polymers (SMPs) are a very interesting and well-known class of stimuli-responsive
materials which have attracted a great deal of attention due to their ability to restore an equilibrium
shape upon certain stimulus such as temperature, electricity, pH, ionic strength and light [1].
The non-equilibrium shape can be easily fixed by means of a memorizing step where elastic stresses
are frozen into the material structure. Generally, temperature is used to trigger the non-equilibrium
shape to the equilibrium shape. In the last decade, SMPs have found a wide range of application in
many fields due to the variety of shapes in which they can be manufactured, including sheets, bulks,
fibers, and foams. Foam structures improve the shape memory behavior of bulk polymers mainly in
terms of compression rate [2,3]. For this reason, SMP foams can be used in the aerospace industry in
light actuators [4], and in expandable and self-deployable structures such as wheels for exploration
rovers [5], and solar sails [6].
At present, numerous methods have been developed to produce SMP foams on the basis
of physical (syntactic foams, particulate leaching, phase separation) and chemical principles.
Foam structures strongly depend on the production method in terms of cell size, distribution,
and extension along the foaming direction. In order to evaluate these differences, the shape anisotropy
ratio R, defined as the average ratio of cell size in the rise and transverse directions, is considered.
Geometrical anisotropy of foam cells leads to anisotropic mechanical properties of foams, with higher
strength and stiffness exhibited in the rise direction [7]. The shape anisotropy ratio generally decreases
with increasing foam density [8]. In the case of closed-cell rigid PU foams, R decreases from 2.5 to
1.7 by increasing the foam density by 1.5 times [9]. By reducing the relative density down to 11–13%,
R is about 1.6, which is very close to the stiffness anisotropy (1.7), [10]. It would be expected that the
anisotropy ratio affects shape memory behavior as well. Polyurethane foams are the first and most

Polymers 2018, 10, 115; doi:10.3390/polym10020115 www.mdpi.com/journal/polymers

Polymers 2018, 10, 115 2 of 16

studied foams for their shape memory properties [11–13], although the effect of the foam structure
anisotropy on the shape memory behavior has been never investigated.
Apart from polyurethane-based foams which are mainly thermosetting, most foams are
thermoplastics. However, the shape memory properties related to the foam structure induced
by the production process, have not been fully investigated. In particular, the properties of the
memory of thermoplastic foams have been evaluated for polycaprolactone (PCL), microbial polyester
(poly-hydroxybutyrate-hydroxyvalerate, PHBV), and their mixture (PCL/PHBV) [14]. After training,
in the best case, a 100% shape recovery was achieved on 30% compressed PHBV foams. There was
a linear dependence of the recovery stress on the temperature, but no correlation with the compression
direction or the foam structure was studied. PCL, PHBV, and PCL/PHBV foams are soft at room
temperature, and fix the deformed shape by cooling. A different behavior was found in thermoplastic
PET (polyethylene-terephthalate) foams, which are rigid at room temperature [15]. PET foams exhibit
good shape-memory properties also by performing the memorizing step in cold conditions. This is
quite a rare occurrence in the world of SMPs as typically these polymers are strongly damaged when
deformed in a rigid state. Small cubic samples were subjected to a memory-recovery cycle which
consisted of a cold compression to reduce the foam sample thickness up to 50%, and a subsequent
recovery of the shape by heating the samples in a muffle. Final recovery from 97% to 100% was found
as a function of the compression direction.
In this study, the recovery properties of PET foams are discussed as a function of the foam
orientation. Because of the shape anisotropy induced by the extrusion process of PET foam panels,
different behaviors are expected by testing shape memory along the extrusion direction and the
two transverse ones. Memory-recovery cycles were performed by means of compression at room
temperature (cold compression) and subsequent hot recovery. This thermo-mechanical cycle was
repeated three times for each principal direction of the initial foam panel.

2. Experimental Methods
Uniaxial compression tests and thermo-mechanical cycles were carried out along three mutually
orthogonal directions on PET foams to evaluate the anisotropy of shape recovery and residual stiffness
after thermo-mechanical cycling. Samples were extracted from commercial extruded panels, and their
properties were measured by a combination of several tests, including differential scanning calorimetry
(DSC), dynamic mechanical analysis (DMA), and optical microscopy. Results from these tests were
correlated to measured SM behavior under multiple thermo-mechanical cycles.

2.1. Materials
Two commercial PET foam panels (by Point Plastic srl, Colleferro, Italy) were used for
experimental tests. The PET foam panels were made of virgin PET resin (Polyclear Refresh PET
1101 by Invista, Charlotte, NC, USA). These types of commercial foam PET panels are commonly
used as the core of wind blades and sold by densities. Generally, these PET foams are produced by
extrusion with blowing agents (e.g., cyclopentane) and other additives. Blowing agents expand when
pressure is released at the extrusion die to form the foam structure. Because of the low cooling rate,
high crystallinity is reached. Therefore, PET foams have high structural strength which is combined
with low weight and high recyclability. The acquired panels are generally used as core materials for
sandwich structures in technical applications such as composite wind blades.
Two extruded foam panels (1000 × 600 mm2 ) were used with thicknesses of about 17 and 39 mm,
and nominal densities of 65 and 110 kg/m3 , respectively. In the following, the former foam panel
with lower density and thickness is named “low-density foam”, while the latter foam panel is called
“high-density foam”. Three material directions were defined on the basis of the panel orientation
in the extrusion process as shown in Figure 1. “Direction 1” represents the extrusion direction,
whereas “Direction 2” and “Direction 3” are taken in the transverse plane, i.e., the panel cross section.
In particular, Direction 2 is aligned with the panel width and Direction 3 with the panel thickness.
Polymers 2018, 10, 115 3 of 16

Cubic samples
Polymers 2018, 10, xwere
FOR extracted from the panels by cutting for mechanical testing and thermo-mechanical
PEER REVIEW 3 of 15
cycling. The edge length of the cubic samples was fixed equal to the thickness of the foam panel
according
according toto Figure
Figure 1,
1, therefore
therefore 17
17 ×
× 17
17 ××17
17mm
mm33samples
sampleswere
werecut
cutfrom
from the
the low-density
low-density panel
panel and
and
39
39 ××39
39××3939mm
mmsamples
3 3 samplesfrom
fromthe high
the density
high panel.
density panel.

Definition of
Figure 1. Definition of material
material directions
directions (and
(and foam
foam orientations)
orientations) used
used for
for PET
PET (polyethylene-
(polyethylene-
terephthalate) foam samples.

2.2. Foam
Foam Morphology
Foam cell
cell morphology
morphologywas wasinvestigated
investigatedby byobserving
observing the
thesurfaces
surfaces of of
thethe
cubic foam
cubic foamsamples
samplesby
means
by meansof aofpolarized
a polarizedoptical
optical microscope
microscope (Nikon
(NikonEclipse
Eclipse80i,
80i,Nikon
NikonCorporation,
Corporation,Minato,
Minato, Japan).
Japan).
According
According to tothe
thedirections
directionsofofFigure
Figure1,1, three
three different
different planes
planes cancan be found
be found for observation
for observation (1-2,(1-2, 2-3,
2-3, 1-3);
1-3); only1-2
only the the 1-2 corresponds
plane plane corresponds to an as-received
to an as-received surface
surface of the foamof the foam
panel panel2-3,
whereas whereas 2-3, 1-3
1-3 originated
originated fromcutting.
from the panel the panelIn cutting.
order toIn orderthe
assess to shape
assess of
the shape
the cells,oftheir
the average
cells, their
sizeaverage size and
and anisotropy,
anisotropy,
images wereimages
taken inwere takenfor
parallel in each
parallel for each
direction. Celldirection.
dimensionsCell indimensions in cross-sectional
cross-sectional images were
images wereindetermined
determined in all
all three main three main
directions. Thedirections. Theratio
geometrical geometrical ratio was
was calculated as:calculated as:

1 n ℎ
=1 ∑ h
R= (1)
(1)
n i =1 b
where
where hh and
and bb denote
denote cell
cell dimensions
dimensions inin the
the two
two directions
directions transversal
transversal toto each
each other
other (e.g.,
(e.g., 11 and
and 2)
2)
with
with respect
respect to
to the
the observation
observation plane
plane (1-2),
(1-2), and
and nn is
is the
the number
number ofof cells
cells measured
measured for
for aa given
givensample.
sample.
Observations
Observations were made on
were made on the
the samples
samples before
before and
and after
after multiple
multiple memory-recovery
memory-recovery cycles.
cycles.

2.3.
2.3. Calorimetric
Calorimetric and
and Thermal
Thermal Mechanical
Mechanical Analysis
Analysis
Thermal analysishas
Thermal analysis hasbeen
beencarried
carried outout
on on a differential
a differential scanning
scanning calorimeter
calorimeter (Perkin(Perkin
Elmer Elmer
DSC6,
DSC6, Waltham,
Waltham, MA, USA) MA,toUSA) to evaluate
evaluate the glassthe glass transition
transition temperature
temperature (Tg ) of the ) of the
(TgPET PETThis
foams. foams.
valueThisis
value is important to fix the process temperature in the shape recovery step
important to fix the process temperature in the shape recovery step of the thermo-mechanical cycle. of the thermo-mechanical
cycle. According
According to the
to the ISO ISO 11357:2
11357:2 standard,standard,
Tg can T beg can be calculated
calculated as theofpoint
as the point of inflection
inflection in the
in the step step
change
change observed in the heat flow curve. DSC tests were performed in the
observed in the heat flow curve. DSC tests were performed in the temperature range between 25 and temperature range between
25
300and
◦ C,300 °C, nitrogen
under under nitrogen flux,
flux, with with a heating
a heating rate of 10rate of 10 °C/min.
◦ C/min. Calorimetric
Calorimetric tests on tests on low-density
low-density foams
foams are critical
are critical becausebecause
of theirof lowtheir low thermal
thermal conductivity
conductivity and the and the consequent
consequent low sample low mass.
sample Formass.
this
For this reason, only results for the higher density foam are reported below.
reason, only results for the higher density foam are reported below. In this case, about 3 mg of mass In this case, about 3 mg
of mass was put into the aluminum crucible. The alternative is pulverizing
was put into the aluminum crucible. The alternative is pulverizing the foam and using powders for the the foam and using
powders for thedegradation
test but polymer test but polymerwould occur.degradation wouldtheoccur.
Nevertheless, Nevertheless,
PET polymer the PET
formulation was thepolymer
same
formulation was the same for both foam panels and differences were
for both foam panels and differences were not expected for the Tg . Crystallinity should be different not expected for the Tg.
Crystallinity should be different and this datum cannot be extended
and this datum cannot be extended from the high-density foam to the low-density foam. from the high-density foam to
the low-density
A comparison foam.between the two foams can be made by using dynamic mechanical analysis (DMA).
The glass transitionbetween
A comparison temperature the two
Tg . foams
can becan alsobe made byby
measured using dynamic mechanical
the inflection point of theanalysis
storage
(DMA).
modulusThe glass
or the transitionpeak
maximum temperature
of the loss Tg.factor.
can beBeingalso measured
stiff samples, by the
DMA inflection
tests canpoint of the
be carried
storage modulus or the maximum peak of the loss factor. Being stiff
out easily even if foam collapse under loading can affect results. In this study, the compression samples, DMA tests can be
carried out easily even if foam collapse under loading can affect results.
mode was used (by Netzsch DMA 242C, Netzsch Instrument Inc., Selb, Germany). Samples with In this study, the compression
mode was used (by Netzsch DMA 242C, Netzsch Instrument Inc., Selb, Germany). Samples with 10
× 10 × 5 mm3 approximate size were cut from the panels and tested along the thickness direction
(Direction 3). Tests were performed at 5 °C/min from 25 to 170 °C, at the frequency of 1 Hz.
Polymers 2018, 10, 115 4 of 16

Polymers
10 × 102018,
× 510,
mm x FOR PEER REVIEWsize
3 approximate 4 of 15
were cut from the panels and tested along the thickness direction
◦ ◦
(Direction 3). Tests were performed at 5 C/min from 25 to 170 C, at the frequency of 1 Hz.
2.4. Thermo-Mechanical Cycling
2.4. Thermo-Mechanical
The shape memoryCycling properties of PET foams were evaluated by using thermo-mechanical cycles
whereThe theshape
shapememorywas fixed under compression.
properties of PET foams A universal material
were evaluated by testing machine (MTS Insight
using thermo-mechanical 5,
cycles
MTS
where Systems
the shape Corporation,
was fixed Eden
underPrairie, MN, USA)
compression. was usedmaterial
A universal to maketesting
tests along the three
machine (MTSdirections
Insight 5,
shown in Figure
MTS Systems 1. Figure 2Eden
Corporation, shows a typical
Prairie, MN,memory-recovery
USA) was used tothermo-mechanical
make tests along the cycle.
threeIndirections
the first
step,
shown a maximum
in Figure 1.compressive strain
Figure 2 shows εm wasmemory-recovery
a typical applied at room temperature (25 °C) cycle.
thermo-mechanical at 10 mm/min.
In the firstAfter
step,
load removal, in the second step, the sample was left at room temperature
a maximum compressive strain εm was applied at room temperature (25 C) at 10 mm/min. After load◦ for 1 h in an unconstrained
condition
removal, so as tosecond
in the stabilize thethe
step, final strain was
sample εu. Theleftdifference between the for
at room temperature resulting
1 h in final deformation
an unconstrained
and the programmed
condition so as to stabilizedeformation mainly
the final strain εu . depends on thebetween
The difference molecular and shape
the resulting parameters
final deformation of and
the
foam [16]. In the third step, samples were heated in the absence of load
the programmed deformation mainly depends on the molecular and shape parameters of the foam [16].at the temperature T h = T g +

20
In °C
thefor 80 min
third step,insamples
a mufflewere
to recover
heatedthe in strain. A finalofresidual
the absence load at strain (εp) can result.
the temperature Th =Three ◦
samples
Tg + 20 C for
were
80 min tested
in a for eachtofoam
muffle and the
recover direction,
strain. and
A finalthe residual
thermo-mechanical cycle
strain (εp ) can was repeated
result. three times
Three samples were
on each sample. The maximum strain ε was fixed to 50%. Elastic
tested for each foam and direction, and the thermo-mechanical cycle was repeated three times on each
m modulus and maximum
compressive
sample. The stress maximum werestrain
extracted
εm was from each
fixed totest.
50%.InElastic
order modulus
to quantify
andthe shape memory
maximum properties,
compressive stress
shape fixity (R ) and shape recovery (R ) were estimated according to
were extracted from each test. In order to quantify the shape memory properties, shape fixity (Rf ) and
f r Equations (2) and (3),
respectively:
shape recovery (Rr ) were estimated according to Equations (2) and (3), respectively:
ε
Rf = ε ×
= εu 100 (2)(2)
× 100
εm
ε
= εr × 100
Rr = ε × 100 (3)(3)
ε m
where εεrr==εεmm− −
where εp εisp the recovered
is the recoveredstrain. These
strain. parameters
These parametersrefer
refertotothe
theability
abilitytotopreserve
preservethe
theapplied
applied
deformation
deformation and to recover the initial shape. Shape fixity and shape recovery were evaluated foreach
and to recover the initial shape. Shape fixity and shape recovery were evaluated for each
thermo-mechanical
thermo-mechanicalcycle, cycle,direction,
direction,and
andfoam
foamsosoas
asto
to evaluate
evaluate the
the influence
influence of of occurring
occurring damages
damages
and
andmaterial
materialanisotropy.
anisotropy.

Figure Experimentalprocedure
Figure2.2.Experimental procedurefor
forshape
shapememory-recovery
memory-recoverytest.
test.

2.5.
2.5.Conditioning
ConditioningTests
Tests
SMPs
SMPs maymay have
have memory
memory of of the
the processing
processing conditions
conditions during
during manufacturing.
manufacturing. PET PET foams
foams areare
subjected strongthermo-mechanical
subjected to strong thermo-mechanical stresses
stresses during
during the extrusion
the extrusion process,
process, and of
and some some
theseofstresses
these
stresses
could becould be released
released in the
in the first first memory-recovery
memory-recovery cycle. Incycle.
orderInto
order to deepen
deepen our understanding
our understanding of this
of this aspect,
aspect, some thermal
some thermal conditioning
conditioning tests weretests were performed
performed on the low-density
on the low-density foams wherefoamshigherwhere
frozen
higher
stressesfrozen stresses were
were expected becauseexpected because
of the lower of the
panel lower panel
thickness. thickness.
Accordingly, Accordingly,
uniaxial compressionuniaxial
tests
compression
were performed testson
were
17 × performed
17 × 17 mm 3 cubic
on 17 × 17 ×samples
17 mm3after
cubicaging
samples
in aafter aging
muffle for in a muffle
2 h, for 2 h,
and compared
and
withcompared
results from with results from
as-received as-received
not-aged samples.not-aged samples.
Three aging Three aging
temperatures were temperatures
chosen (50, 100wereand
chosen◦
150 C).(50, 100 foam
Three and 150 °C). Three
samples foam for
were used samples were used
each aging for each
condition. aging condition.
Conditioned Conditioned
and un-conditioned
and un-conditioned
samples were tested at samples were tested
room temperature at room
along temperature
Direction 2 of Figurealong Direction
1, according 2 of
to the Figurestep
memory 1,
according to the memory step
of the thermo-mechanical of the thermo-mechanical cycles.
cycles.
Polymers 2018, 10, 115 5 of 16
Polymers 2018, 10, x FOR PEER REVIEW 5 of 15

3. Results
3. Results andand Discussion
Discussion
Understanding
Understanding the shape
the shape memorymemory
behavior behavior
of foamsof is afoams
complex is task
a complex
because oftask
thebecause
combinationof the
combination
of material of material
and structural and structural
parameters. parameters.
Moreover, Moreover,
manufacturing manufacturing
processes processes strongly
strongly influences foam
influences
properties, foam properties,
homogeneity homogeneity
and anisotropy. and anisotropy.
Nevertheless, Nevertheless,
PET foams have shown PET foams
very highhave shown very
SM properties
alsohigh SM properties
by performing coldalso by performing
memory cold memory
steps. Typically, steps. brittle
SMPs exhibit Typically, SMPsinexhibit
behavior brittle
the rigid statebehavior
with
veryinlow
the formability.
rigid state with very
In PET low formability.
foams, this peculiarInbehavior
PET foams, thisdepend
should peculiar onbehavior should depend
the combination of high on
the combination
material crystallinityofandhigh material low
moderately crystallinity and moderately
glass transition temperature.lowCrystals
glass transition temperature.
always behave like
Crystals
a rigid phase,always
whereas behave like a PET
amorphous rigid phase, whereas
is partially amorphous
soft at room PET and
temperature is partially
providessoft at room
structural
temperatureOn
deformability. andtheprovides structural
other hand, deformability.
molecular mobility isOn notthe othertohand,
enough allowmolecular
full shapemobility
recoveryisatnot
roomenough to allowAs
temperature. fulla shape
result, recovery
amorphous at room temperature.
PET also allows foam Ascell
a result, amorphous
collapsing in cold PET also allows
compression
butfoam cell higher
it needs collapsing in cold compression
temperatures but it needs higher temperatures for shape recovery.
for shape recovery.

3.1.3.1.
Material Transitions
Material Transitions
By By
assuming a bulk density ofof1.38 3
assuming a bulk density 1.38g/cm
g/cm3,, aa relative
relative density of 5% 5% and
and8% 8%results
resultsfor
forthethelow
lowand
andhigh-density
high-densityfoam. foam.Indeed,
Indeed, most of the foam volume is composed of voids.
most of the foam volume is composed of voids. This is an issue for This is an issue
formeasuring
measuringmaterial
materialtransitions
transitionswith with thermal
thermal analyses.
analyses. Figure
Figure 3 shows
3 shows the the
DSCDSC curve
curve of a high
of a high density
density sample.
sample. The TThe Tg evaluation
g evaluation is affected
is affected by the by the occurrence
occurrence of relaxation
of relaxation enthalpy enthalpy but aclose
but a value value to 73
close to 73 ◦ C may be inferred. The occurrence of the relaxation enthalpy peak is in agreement with the
°C may be inferred. The occurrence of the relaxation enthalpy peak is in agreement with the expected
expected SM behavior
SM behavior as the polymer
as the polymer is able istoable
freeze to stresses
freeze stresses
into theinto the structure.
structure. Moreover,
Moreover, that is that
also is also
evidence
evidence of the strong influence of the manufacturing process on the material
of the strong influence of the manufacturing process on the material status. The estimated value of status. The estimated
value
73 °C for◦ C
of 73 Tgfor Tgagreement
is in is in agreementwith with the expected
the expected datumdatum for kind
for this this kind of resin.
of resin. A melting
A melting peakpeakis also
is also visible
visible in theinDSCthe DSC
scan scan
even even if its shape
if its shape and size andissize is probably
probably affectedaffected
by lowby low sample
sample mass, and mass,foam
andcollapse
foam collapse into the crucible during melting. Nevertheless, a recrystallization
into the crucible during melting. Nevertheless, a recrystallization peak, before the melting peak, before the
melting
peak,peak,
seemsseems to be to be absent
absent even even
though though it is typical
it is typical for products.
for PET PET products. It is possible
It is possible to deduce
to deduce that the
thatfoam
the foam
itselfitself is close
is close to itstomaximum
its maximum crystallinity.
crystallinity. After
After foamfoam extrusion,
extrusion, thethe panel
panel coolingrate
cooling rateisislow
lowbecause
becauseofofthe thevery
verylimited
limitedthermal
thermaldiffusivity,
diffusivity,and andamorphous
amorphousPET PETisisminimized.
minimized.

Figure
Figure 3. Differential
3. Differential scanning
scanning calorimetry
calorimetry (DSC)
(DSC) curve
curve for for high-density
high-density foam.
foam.

DMADMA teststests
confirmconfirm thefrom
the data dataDSC,
from DSC, as
as shown shown4.inByFigure
in Figure 4. Bythe
considering considering the peak
peak temperature
of the loss factor (tan δ) in correspondence with the inflection point of the storage modulus (E’), storage
temperature of the loss factor (tan δ) in correspondence with the inflection point of the a Tg
modulus
value of 84 ◦ C(E’), a Tfor
results g value of 84 °C results for the ◦low-density foam, and 89 °C for the high-density
the low-density foam, and 89 C for the high-density foam. The glass transition
foam.
is in The glass
the same rangetransition
of the DSC is in the same
curve but the range of thevalue
extracted DSC is
curve butbecause
higher the extracted value
of a small is higher
thermal
because of a small thermal lag. The peak of the loss modulus (E’’) is closer
lag. The peak of the loss modulus (E”) is closer to the DSC value. Unfortunately, results are partially to the DSC value.
Unfortunately, results are partially affected by sample collapse under load
affected by sample collapse under load during heating, as shown by the storage modulus increase during heating, as shown
by the
before thestorage modulus increase
glass transition. before the
This occurrence glassbetransition.
could This
responsible foroccurrence couldinbemeasured
the differences responsible Tg for
the differences in measured Tg of the low and high-density foam as well as between the DSC and
DMA results, even if a perfect superposition is very rare. Nevertheless, the effect of molecular
Polymers 2018, 10, 115 6 of 16

Polymers 2018, 10, x FOR PEER REVIEW 6 of 15

of the low and high-density foam as well as between the DSC and DMA results, even if a perfect
mobility during
superposition is heating is evident
very rare. despitethe
Nevertheless, theeffect
highofcrystallinity level. Theduring
molecular mobility storage modulus
heating of the
is evident
foam shows
despite a sudden
the high decrease
crystallinity after
level. Tg storage
The even if the melting
modulus of temperature
the foam shows is far away (250
a sudden °C, Figure
decrease after 3).
Tg
Also, differences in foam stiffness for the two densities
◦ are clear. At 30 °C the storage
even if the melting temperature is far away (250 C, Figure 3). Also, differences in foam stiffness for modulus is the
3.7
MPa for the low-density
two densities are clear. Atfoam,
30 ◦ C and 6.5 MPa
the storage the for is
modulus high-density
3.7 MPa for foam. However,foam,
the low-density this difference
and 6.5 MPa in
the
the glassy state is completely
for high-density overcome
foam. However, indifference
this the rubbery in state (over state
the glassy 150 °C) where foams
is completely behave in
overcome in the
the
same way. ◦
rubbery state (over 150 C) where foams behave in the same way.

Figure 4. Dynamic mechanical analysis (DMA) curves for low-density (a) and high-density (b) foams.
Figure 4. Dynamic mechanical analysis (DMA) curves for low-density (a) and high-density (b) foams.

3.2. Morphology and Mechanical Properties

3.2. Morphology and Mechanical Properties
Micrographs of the two different foams along the three principal directions are shown in Figure
Micrographs of the two different foams along the three principal directions are shown in Figure 5.
5. The foam structure is closed-cell with an average cell size of 0.54 mm for the high-density foam,
The foam structure is closed-cell with an average cell size of 0.54 mm for the high-density foam,
and 0.73 mm for the low-density foam. However, the measured cell size is not uniform along all the
and 0.73 mm for the low-density foam. However, the measured cell size is not uniform along all the
foam directions. In the 1-2 plane, the average cell size is 0.7 and 0.98 mm, respectively, in the 2-3 plane
foam directions. In the 1-2 plane, the average cell size is 0.7 and 0.98 mm, respectively, in the 2-3 plane
the average cell size is 0.42 and 0.55 mm, respectively, and in the 1-3 plane, the average cell size is
the average cell size is 0.42 and 0.55 mm, respectively, and in the 1-3 plane, the average cell size is
0.45 and 0.59 mm, respectively. According to these observations, foam cells were mainly elongated
0.45 and 0.59 mm, respectively. According to these observations, foam cells were mainly elongated
along Direction 1 (the extrusion direction) and that is apparent from the comparison between plane
along Direction 1 (the extrusion direction) and that is apparent from the comparison between plane
1-2 and 2-3. Cell sizes in plane 2-3 and 1-3 are comparable even if a small increase is present in the
1-2 and 2-3. Cell sizes in plane 2-3 and 1-3 are comparable even if a small increase is present in the
latter. A further comparison can be made in terms of the geometrical ratio of the cells along the three
latter. A further comparison can be made in terms of the geometrical ratio of the cells along the
principal planes as reported in Table 1. For both foams, the geometrical ratio maximum is in the 1-2
three principal planes as reported in Table 1. For both foams, the geometrical ratio maximum is in the
plane, as qualitatively confirmed by Figure 5a,b. Between foams, the cell ratio is higher for the low
1-2 plane, as qualitatively confirmed by Figure 5a,b. Between foams, the cell ratio is higher for the low
density sample, where a value of 2.25 is reached. A high geometrical ratio is also present in the 1-3
density sample, where a value of 2.25 is reached. A high geometrical ratio is also present in the 1-3
plane but in this case the maximum value (1.52) is achieved by the high-density foam. However, no
plane but in this case the maximum value (1.52) is achieved by the high-density foam. However, no
anisotropy is observable in the 2-3 plane for both foams. The presence of anisotropy in both the 1-2
anisotropy is observable in the 2-3 plane for both foams. The presence of anisotropy in both the 1-2
and 1-3 planes suggests that cell elongation is not perfectly aligned with the extrusion direction.
and 1-3 planes suggests that cell elongation is not perfectly aligned with the extrusion direction.
 Polymers 2018, 10, 115 7 of 16
Polymers 2018, 10, x FOR PEER REVIEW 7 of 15

Figure 5. Foam structure for all foam directions and densities.

Figure 5. Foam structure for all foam directions and densities.

Table 1. Geometrical ratio of the foam cells for the different planes of observation.
Table 1. Geometrical ratio of the foam cells for the different planes of observation.
Geometrical Ratio
Foam Type
1-2 Geometrical
1-3 Ratio 2-3
Foam Type
Low-Density Foam 2.25 ±
1-2 0.45 1.46 ±1-3
0.39 1.03 ± 0.17
2-3
High-Density Foam 2.01 ± 0.37 1.52 ± 0.21 1.06 ± 0.16
Low-Density Foam 2.25 ± 0.45 1.46 ± 0.39 1.03 ± 0.17
High-Density Foam 2.01 ± 0.37 1.52 ± 0.21 1.06 ± 0.16
Microstructural observations reveal the strong anisotropy of PET foams. It is expected that
mechanical and functional performances are affected by this anisotropy. The foam response to
Microstructural
external loads dependsobservations reveal
on the direction theapplied
of the strong anisotropy of PET6a,b
forces, and Figure foams.
showIttheis expected
compressive that
mechanical and functional performances are affected by this anisotropy. The foam
stress-strain curves along the three principal directions. These tests correspond to the first loading response to external
loadsofdepends
stage on the directioncycle.
the memory-recovery of theThree
applied forces,are
samples and Figurefor
shown 6a,b show
each the compressive
direction. Tests along stress-strain
the same
curves along the three principal directions. These tests correspond to the
direction are almost perfectly superimposed so as to show the good homogeneity of the initial foam first loading stage of the
memory-recovery cycle. Three samples are shown for each direction. Tests along the same direction
panels.
are almost
The effectperfectly superimposed
of the sample so asistoevident.
orientation show the goodrigidity
Higher homogeneity
and higherof thestrength
initial foam panels.
are measured
The effect of the sample orientation is evident. Higher rigidity and
along Direction 1 (the extrusion direction) whereas similar performances are measured in the higher strength are measured
other
two directions (along the panel width and thickness). Along Direction 1, a compressive strengthother
along Direction 1 (the extrusion direction) whereas similar performances are measured in the can
twobe
also directions
extracted(along
for boththe foams
panel width
as the and
plateauthickness).
value ofAlong Direction
the stress. Along 1, the
a compressive strength can
other two directions, a
also be is
plateau extracted for both
not visible. foams
Instead, twoaslinear
the plateau
stages value of the with
are present stress.different
Along the other
slope. In two
thesedirections,
cases, a
a plateau is strength
compressive not visible.can Instead, two linear
be extracted as the stages
maximum are present withthe
stress (i.e., different
stress atslope. In theseFoam
50% strain). cases,
a compressive strength can be extracted as the maximum stress (i.e., the
densification is not visible along Direction 1 whereas it seems to start after the second linear stage instress at 50% strain).
Foam
the otherdensification
two directions.is notHigher
visiblecell
along Direction
sizes 1 whereas
lead to higher it seems
material to start after
deformability the second
whereas linear
molecular
stage in the other two directions. Higher cell sizes lead to higher material
orientation is responsible for stiffness and strength. A difference is appreciable between the width deformability whereas
and thickness direction only for the low-density foam because of the high cell size.
Polymers 2018, 10, 115 8 of 16

molecular orientation is responsible for stiffness and strength. A difference is appreciable between the
Polymersand
width 2018,thickness
10, x FOR PEER REVIEW
direction only for the low-density foam because of the high cell size. 8 of 15

Figure 6.
Figure 6. Cold
Cold compression tests for
compression tests for low
low density
density (a)
(a) and
and high
high density
density samples
samples (b).
(b).

In order
In order to to quantify
quantify this
this difference
difference in in mechanical
mechanical properties,
properties, the the Young’s
Young’s modulus
modulus (E (Ec)) was
was
c
estimated from the slope of the initial linear stages, even if non-elastic
estimated from the slope of the initial linear stages, even if non-elastic contributes could be contributes could be
superimposed as well. The maximum stress reached at 50% of the
superimposed as well. The maximum stress reached at 50% of the strain was taken as the foam strain was taken as the foam
strength (σ
strength (σS).). These
Thesevalues
valuesare arereported
reportedinin Table
Table 2 in2 terms
in terms of average
of average and and dispersion
dispersion of measured
of measured data.
S
data. The compressive modulus (E c) in the extrusion direction is approximately 2 times that in the
The compressive modulus (Ec ) in the extrusion direction is approximately 2 times that in the transverse
transverse
direction fordirection
low densityfor samples.
low density The foamsamples. Theinfoam
strength strengthdirection
the extrusion in the is extrusion direction
approximately twiceis
approximately twice that of the other two directions as well. Data for
that of the other two directions as well. Data for the high-density foam are higher but mechanical the high-density foam are
higher but mechanical
anisotropy seems to beanisotropy seems
similar, a part to be
from similar,
a lower a part from
uniformity inathe
lower uniformity
transverse in the
plane. Fortransverse
example,
plane. For example, in the low-density foam, the mean compressive modulus
in the low-density foam, the mean compressive modulus in Direction 3 is 12% higher than in Direction in Direction 3 is 12%
higher than in Direction 2. This difference increases up to 42% for the high-density
2. This difference increases up to 42% for the high-density foam. Data for strength are not perfectly foam. Data for
strength are not perfectly aligned due to the occurrence of foam densification
aligned due to the occurrence of foam densification during testing. By using the compressive modulus during testing. By using
the the
for compressive
evaluationmodulus for the evaluation
of an anisotropy of anofanisotropy
ratio, a value 2.1 results ratio,
alongathe value of 2.1 results
extrusion along
direction the
for the
extrusion direction for the low-density foam, with a result of 2.4 for the high-density
low-density foam, with a result of 2.4 for the high-density foam. These anisotropy ratios are very close foam. These
anisotropy
to ratios are
the geometrical very
ratios close
of the to of
cells the geometrical
Table ratios of the
1. By considering the compressive
cells of Tablestrength,
1. By considering
the anisotropythe
compressive strength, the anisotropy
ratio decreases down to 1.6 and 2.0, respectively.ratio decreases down to 1.6 and 2.0, respectively.

Table 2. Average
Table 2. compressive properties
Average compressive properties of
of PET
PET foams
foams during
during first
first cold
coldcompression.
compression.

1° Compression Low-Density Foam High-Density Foam

1◦ Compression
Direction Low-Density
Ec (MPa) σSFoam
(MPa) High-DensityσFoam
Ec (MPa) S (MPa)

Direction
1 Ec (MPa)
13.66 ± 0.92 σS±(MPa)
0.71 0.01 Ec (MPa)
55.26 ± 4.9 σS (MPa)
2.14 ± 0.02
12 6.03 ± 0.19
13.66 ± 0.92 0.41 ±
0.71 ±0.01
0.01 18.84 ± 2.1
55.26 ± 4.9 2.14 ±±0.02
1.08 0
23 6.78 ± 1.19
6.03 ± 0.19 0.47 ±
0.41 ±0.01
0.01 27.36 ± 1.43
18.84 ± 2.1 1.05 ± 0.01
1.08 ± 0
3 6.78 ± 1.19 0.47 ± 0.01 27.36 ± 1.43 1.05 ± 0.01
3.3. Sample Conditioning
3.3. Sample Conditioning
In order to investigate how the stresses during the extrusion process could affect PET foams in
the first thermo-mechanical
In order to investigate how cycle,
thesome thermal
stresses duringconditioning
the extrusiontests were could
process performed. In particular,
affect PET foams in
three
the aging
first temperatures were
thermo-mechanical chosen
cycle, some(50, 100 and
thermal 150 °C) and
conditioning testsuniaxial compression
were performed. In tests were
particular,
performed
three agingon the low-density
temperatures were foams.
chosenThe (50,effect of the
100 and ◦ C) and uniaxial
150conditioning treatment on the mechanical
compression tests were
performance
performed on istheshown in Figure
low-density foams.7 for
Thethe low-density
effect foam in treatment
of the conditioning terms of onroom-temperature
the mechanical
compression iscurves.
performance shownThe non-aging
in Figure 7 for condition is alsofoam
the low-density reported for comparison.
in terms Direction
of room-temperature 2 has been
compression
chosen for
curves. Thetesting
non-aging because of theislow
condition alsoanisotropy
reported forratio. The compression
comparison. Directioncurves
2 has are
beenqualitatively
chosen for
similar because
testing and theyofdiffer only
the low in some quantitative
anisotropy data.
ratio. The compression curves are qualitatively similar and they
differ only in some quantitative data.
 Polymers 2018, 10, x FOR PEER REVIEW 9 of 15
Polymers 2018, 10, 115 9 of 16
Polymers 2018, 10, x FOR PEER REVIEW 9 of 15

Figure 7. Compression curves after aging for low-density foams along Direction 2.
Figure 7. Compression
Figure 7. Compression curves
curves after
after aging
aging for
for low-density
low-densityfoams
foamsalong
alongDirection
Direction2.2.
Compression strength and elastic modulus were extracted from these curves and are reported
in Figure 8a,b, respectively.
Compression strength and By elastic
and increasing
modulusthe conditioning
were extracted
extractedtemperature,
thesethe
from these compression
curves and are strength
are reported
reported
Compression strength elastic modulus were from curves and
decreases
in but effect
Figure 8a,b,
8a,b, is small.
respectively. ByBetween
increasing thethemaximum (in the
conditioning absence of the
temperature, aging) and the minimum
compression strength
in Figure respectively. By increasing the conditioning temperature, the compression strength
(aging temperature
decreasesbutbuteffect of
effectis is 150 °C),
small. an 8%
Between reduction is observed. The same effect seems to disappear in
decreases small. Between thethe maximum
maximum (inabsence
(in the the absence of aging)
of aging) and the and the minimum
minimum (aging
the elastic modulus which is always comparable for all the samples before and after conditioning.
temperature of 150 ◦ C),
(aging temperature of 150 °C),reduction
an 8% an 8% reduction
is observed.is observed.
The sameThe effectsame effect
seems seems to in
to disappear disappear in
the elastic
Results are comparable
the elasticwhich
modulus with a small
whichcomparable material
is always comparable degradation
for all the due to
samples thermal
beforeaging. This degradation
and after conditioning. is
modulus is always for all the samples before and after conditioning. Results are
stronger above
Results are comparablethe glass transition temperature. Nevertheless, other secondary effects could be
comparable with a smallwith a small
material material degradation
degradation due to thermaldue to thermal
aging. Thisaging. This degradation
degradation is stronger is
superimposed,
stronger above such
the as recrystallization,
glass transition loss of molecular
temperature. orientation,
Nevertheless, other residual
secondary stress recovery
effects could and
be
above the glass transition temperature. Nevertheless, other secondary effects could be superimposed,
thermal shrinkage.
superimposed, such as These results suggest that
recrystallization, strong material changes during multiple thermo-
such as recrystallization, loss of molecular loss of molecular
orientation, residualorientation, residual
stress recovery andstress
thermalrecovery and
shrinkage.
mechanical
thermal cycles
shrinkage. should
These not depend
results on
suggest temperature
that strong effects
material alone but
changes on their
during combination
multiple with
thermo-
These results suggest that strong material changes during multiple thermo-mechanical cycles should
applied stresses.
mechanical cycles should not depend
not depend on temperature effects aloneon buttemperature effects alone
on their combination withbut on their
applied combination with
stresses.
applied stresses.

Figure8.8. Compressive
Figure Compressive strength
strength (a)
(a)and
andelastic
elasticmodulus
modulus(b)(b)
forfor
different aging
different conditions
aging of the
conditions of low-
the
density
low-densityfoam.
foam.
Figure 8. Compressive strength (a) and elastic modulus (b) for different aging conditions of the low-
density foam.
3.4.Shape
3.4. ShapeMemory
MemoryProperties
Properties
3.4. Shape
The Memory
Theshape
shape Properties
memory
memory properties
properties of foams
of PET PET foams werebytested
were tested meansby means ofthermo-mechanical
of multiple multiple thermo-
mechanical
cycles. Sample
The cycles.
shape Sampleproperties
memory pre-conditioning
pre-conditioning of PETwas
was omitted omitted
because
foams of
were because
the testedofby
negligiblethe negligible
effect
means of the effect conditions.
aging
of multiple of the aging
thermo-
conditions.
Figure 9 shows
mechanical Figure 9 shows
a typical
cycles. Sample foam asample
typicalafter
pre-conditioning foamthe sample
wasmemory after
omitted step the memory
which
because ofcreates step which creates
a non-equilibrium
the negligible effect aaging
non-
of theshape,
equilibrium
Figure 9a, andshape,
after Figure
the 9a,
recovery and after
step, the
Figure recovery
9b. The step,
recoveryFigure
step 9b.
was The recovery
performed instep
a was
muffle performed
at 100 ◦C
conditions. Figure 9 shows a typical foam sample after the memory step which creates a non-
in
for a muffle at
a total period
equilibrium 100
shape, °C for
ofFigure a total
80 min;9a,
this
and period
time
afterwasof 80 min;
thenecessary this time
to complete
recovery step, was necessary
theThe
Figure 9b. recovery to
recoverycomplete
phase
stepdue the
wasto recovery
the high
performed
in a muffle at 100 °C for a total period of 80 min; this time was necessary to complete the recovery
Polymers 2018, 10, 115 10 of 16
Polymers 2018, 10, x FOR PEER REVIEW 10 of 15

porosity
phase dueof to
thethe
PET foams.
high Threeof
porosity samples
the PET forfoams.
each foam and
Three direction
samples forwere
eachtested
foam in three
and consecutive
direction were
memory-recovery cycles for a total of 54 compression tests. Compressive strength and elastic
tested in three consecutive memory-recovery cycles for a total of 54 compression tests. Compressive modulus
were extracted
strength from modulus
and elastic each compression curve whereas
were extracted shape
from each fixity andcurve
compression shapewhereas
recoveryshape
were calculated
fixity and
before each thermo-mechanical cycle according to Equations (2) and (3) respectively.
shape recovery were calculated before each thermo-mechanical cycle according to Equations Because(2) ofand
the
huge amount of data, it is important to discuss one aspect at a time.
(3) respectively. Because of the huge amount of data, it is important to discuss one aspect at a time.

Figure 9. Low density sample after memory (a) and recovery (b) step of a single thermo-mechanical
Figure 9. Low density sample after memory (a) and recovery (b) step of a single thermo-mechanical cycle.
cycle.

Table 33 reports
Table reports shape
shape fixity
fixity and shape recovery
and shape recovery at at the
the end
end ofof the
the first
first thermo-mechanical
thermo-mechanical cycle. cycle.
Shape fixity is always far from 100% because of large strain recovery at
Shape fixity is always far from 100% because of large strain recovery at the end of the compression the end of the compression step.
Material
step. rigidity
Material plays an
rigidity important
plays role in role
an important this mismatch as the highest
in this mismatch as the shape
highest fixity wasfixity
shape achieved
was
by the high-density foam which was stiffer. Moreover, for the same
achieved by the high-density foam which was stiffer. Moreover, for the same foam, higher values foam, higher values were found
along found
were the extrusion
along the direction (Direction
extrusion 1) which
direction was stiffer
(Direction 1) whichthan both the transverse
was stiffer than both directions. In fact,
the transverse
the high-density
directions. In fact,foam in the extrusion
the high-density foam direction shows shape
in the extrusion fixity shows
direction of 64%.shapeThis value decreases
fixity of 64%. This to
52% in the thickness direction (Direction 3), and to 53% in the transverse
value decreases to 52% in the thickness direction (Direction 3), and to 53% in the transverse directionsdirections of the low-density
foam.
of Nevertheless,
the low-density for both
foam. foams, low
Nevertheless, forvalues of shape
both foams, low fixity depend
values of shapeon the elastic
fixity depend strains recovered
on the elastic
at the end of the memory step, whereas plastic strains seem to be negligible.
strains recovered at the end of the memory step, whereas plastic strains seem to be negligible. This This fact is confirmed
by the
fact high values
is confirmed by oftheshape recovery,
high values whichrecovery,
of shape is alwayswhich higher is than
always 97% for the
higher than low-density
97% for thefoam,low-
and 92%foam,
density for theand high-density
92% for thefoam. Higher foam.
high-density shapeHigher
recovery wasrecovery
shape achievedwas in the low-density
achieved in the foam
low-
becausefoam
density of thebecause
initial lower
of theshape
initialfixity.
lowerAshape
smallfixity.
contribution
A smallof plastic strain
contribution of is present,
plastic mainly
strain in the
is present,
extrusion direction of the high-density foam. Being softer, the low-density
mainly in the extrusion direction of the high-density foam. Being softer, the low-density foam foam immediately relaxes
a part of the applied
immediately relaxes strain
a partand canapplied
of the also achieve
strain100%and shape
can alsorecovery
achievealong100%Direction 3. The data
shape recovery along in
Table 3 confirms that material anisotropy leads to an anisotropic shape
Direction 3. The data in Table 3 confirms that material anisotropy leads to an anisotropic shape memory behavior. Certainly,
this anisotropic
memory behavior. behavior
Certainly,is more evident forbehavior
this anisotropic the high-density foam than
is more evident for theforhigh-density
the low-density foamfoamthan
for which the shape recovery variability along the three directions is very
for the low-density foam for which the shape recovery variability along the three directions is very low. Cell orientation plays
a positive
low. role in terms
Cell orientation of shape
plays fixity role
a positive but ainnegative
terms ofrole in terms
shape fixityofbutrecovery
a negative strain. Nevertheless,
role in terms of
in comparison
recovery strain.with other SMPs, PET
Nevertheless, foams exhibitwith
in comparison excellent
otherproperties.
SMPs, PET In fact, theseexhibit
foams values for shape
excellent
properties. In fact, these values for shape fixity and shape recovery were obtained with aSMPs
fixity and shape recovery were obtained with a cold memory step, while typical testing of other cold
apply thestep,
memory memorywhilestep aftertesting
typical heating. of other SMPs apply the memory step after heating.

Table 3.3. Shape

Table Shapefixity and
fixity shape
and recovery
shape afterafter
recovery the first
thethermo-mechanical cycle forcycle
first thermo-mechanical all foams andfoams
for all directions.
and
directions.
Low-Density Foam High-Density Foam
Direction Low-Density Foam High-Density Foam
Direction Shape Fixity (%) Shape Recovery (%) Shape Fixity (%) Shape Recovery (%)
Shape Fixity (%) Shape Recovery (%) Shape Fixity (%) Shape Recovery (%)
1 1 58.04 ±
58.04 0.68
± 0.68 97.65± ±
97.65 1.18
1.18 63.77
63.77 ± 1.73
± 1.73 ± 0.71± 0.71
92.7592.75
2 52.91 ± 5.60 99.22 ± 1.34 59.72 ± 2.68 97.35 ± 0.80
2 52.91 ± 5.60 99.22 ± 1.34 59.72 ± 2.68 97.35 ± 0.80
3 53.33 ± 0.68 100 ± 0 51.80 ± 0.89 98.46 ± 0.51
3 53.33 ± 0.68 100 ± 0 51.80 ± 0.89 98.46 ± 0.51

In applying multiple thermo-mechanical cycles, material

thermo-mechanical cycles, material and
and microstructural
microstructural changes
changes can be
expected in the foam because of the combination
combination of heat and stresses.
heat and stresses. Pre-conditioning tests have
shown that material effects only related to thermal aging should be negligible. However, a lot of
Polymers 2018, 10, 115 11 of 16

Polymers 2018, 10, x FOR PEER REVIEW 11 of 15

Polymers
shown 2018,
that 10, x FOR PEER
material REVIEW
effects only related to thermal aging should be negligible. However, a11lot of of
15
energy is dissipated into the foam during mechanical loading and this energy can lead to positive or
energy
energy is dissipated
dissipated into
is effects. into the
the foam
foam during
during mechanical
mechanical loading
loading and
and this
this energy
energy can
can lead
lead to
to positive
positive or
or
negative
negative
negative effects.
effects.
Figures 10 and 11 show the compression step of the second and third thermo-mechanical cycle,
Figures
Figures 10 10 and 11 show the
the compression step of
of the second and
and third thermo-mechanical cycle,
respectively, forand
the 11
lowshow
density compression
(a) and high step
densitythe
(b)second
foam. These third thermo-mechanical
curves cycle,
are similar to those of
respectively,
respectively, for
for the
the low
low density
density (a)
(a) and
and high
high density
density (b)
(b) foam.
foam. These
These curves
curves are
are similar
similar to
to those
those of
of
the first compression test (Figure 6) apart from the occurrence of a more pronounced densification
the
the first
first compression test test (Figure 6)
6) apart from the
the occurrence of aa more
more pronounced densification
stage at compression
high strain. Also, (Figure
the stress apart from
plateau occurrence
disappears along of
Direction pronounced
1. densification
In particular, the main
stage
stage at
at high
high strain.
strain. Also,
Also, the
the stress
stress plateau
plateau disappears
disappears along
along Direction
Direction 1.
1. In
In particular,
particular, the
the main
main
differences in compression in Direction 1 are strongly related to the mechanical damage during the
differences
differences in
in compression
compression in
in Direction
Direction 1
1 are
are strongly
strongly related
related to
to the
the mechanical
mechanical damage
damage during
during the
the
first cold compression step.
first
first cold
cold compression
compression step.step.

Figure 10. Cold compression tests for low density (a) and high density samples (b) in the second cycle.
Figure 10. Cold
Cold compression
compression tests for low density
density (a) and high
high density
density samples (b) in the second
second cycle.

Figure 11. Cold compression tests for low density (a) and high density samples (b) in the third cycle.
Figure
Figure 11.
11. Cold
Cold compression
compression tests
tests for
for low
low density
density (a)
(a) and
and high
high density
density samples
samples (b)
(b) in
in the third cycle.
the third cycle.

The
The compressive
compressive strength
strength and and elastic
elastic modulus
modulus cancan also
also be
be extracted
extracted from from the the following
following twotwo
The compressive
compression strength and elastic modulus canthe
also be extracted from the following
compression stages to make a comparison with values from the first stage. Data are reported in
stages to make a comparison with values from first stage. Data are reported in Figure
Figure
two compression
12, stages to properties
make a comparison during
with values from the firstdensity,
stage. Data biggest
are reported in
12, which
which shows
shows thatthat both
both properties decrease
decrease during cycling.
cycling. For
For each
each density, the the biggest drop
drop isis
Figure 12, which shows
measured that boththe properties decrease during cycling. For each density,The the biggest
measured between
between the the first
first and
and the second
second cycle
cycle and
and along
along the
the extrusion
extrusion direction.
direction. The loss loss of
of
drop is measured
properties between the first and the second cycle and along first
the extrusion direction. The loss
properties isis also
also strong
strong inin the
the transverse
transverse directions
directions between
between thethe first twotwo cycles.
cycles. Comparing
Comparing the the
of properties
second and is
third also strong
cycle, a in
small the transverse
drop is directions
always present, between
but not the
as first two
significant cycles.
as theComparing
previous the
one.
second and third cycle, a small drop is always present, but not as significant as the previous one.
second and third
However, cycle, aeffect
small drop is always present, but not as thesignificant as the previous one.
However, in in all
all cases,
cases, the
the effect of of panel
panel orientation
orientation is
is stronger
stronger than
than the effect
effect ofof thermo-mechanical
thermo-mechanical
However,
cycling. in allthis
cases, theofeffect ofthe
panel orientation is stronger than the effectcycle”
of thermo-mechanical
cycling. From this point of view, the first cycle could be seen as a “training cycle” which
From point view, first cycle could be seen as a “training which partially
partially
cycling. From this
reduces point of view, the first cycle could be seen as a “training cycle” which partially
reduces material
material anisotropies.
anisotropies. Nevertheless,
Nevertheless, mechanical
mechanical performances
performances at at the
the end
end ofof this
this training
training cycle
cycle
reduces material
are anisotropies. Nevertheless, mechanical performances at the end of this training cycle
are reduced,
reduced, and
and continue
continue to to minimally
minimally diminish
diminish in
in following
following cycles.
cycles.
are reduced, and continue to minimally diminish in following cycles.
Polymers 2018, 10, 115 12 of 16
Polymers 2018, 10, x FOR PEER REVIEW 12 of 15

Figure 12.
Figure 12. Compressive
Compressivestrength
strength(a,b)
(a,b) and
and elastic
elastic modulus
modulus (c,d)
(c,d) for low
for low density
density (a,c) (a,c) and high
and high density
density (b,d)
(b,d) foam samples in all directions after each thermo-mechanical
foam samples in all directions after each thermo-mechanical cycle. cycle.

From a quantitative point of view, between the first two cycles, compressive strength along
From a quantitative point of view, between the first two cycles, compressive strength along
Direction 1 underwent a reduction of about 42% and 48% for low- and high-density foams
Direction 1 underwent a reduction of about 42% and 48% for low- and high-density foams respectively.
respectively. Subsequently, between the second and the third cycle, a further reduction of 12% and
Subsequently, between the second and the third cycle, a further reduction of 12% and 14% occurred.
14% occurred. Along the same direction, the elastic modulus underwent an even more severe
Along the same direction, the elastic modulus underwent an even more severe reduction between the
reduction between the first two cycles (64% for low-density foam and 67% for high-density foam)
first two cycles (64% for low-density foam and 67% for high-density foam) whereas the effect of the
whereas the effect of the third cycle was almost negligible.
third cycle was almost negligible.
The effect of foam anisotropy on mechanical performances is very strong and is not erased by
The effect of foam anisotropy on mechanical performances is very strong and is not erased by
thermo-mechanical cycling. It is questionable if a comparable effect is also present for shape memory
thermo-mechanical cycling. It is questionable if a comparable effect is also present for shape memory
properties in terms of shape fixity (Figure 13) and shape recovery (Figure 14). According to the static
properties in terms of shape fixity (Figure 13) and shape recovery (Figure 14). According to the
properties (Figure 12), shape fixity and shape recovery are reduced due to thermo-mechanical cycles
static properties (Figure 12), shape fixity and shape recovery are reduced due to thermo-mechanical
but the changes are smaller. Dramatic drops, as seen with the compressive strength and elastic
cycles but the changes are smaller. Dramatic drops, as seen with the compressive strength and elastic
modulus, were never observed for Rf and Rr. Foams always preserve good shape memory behavior,
modulus, were never observed for Rf and Rr . Foams always preserve good shape memory behavior,
and the maximum reduction of Rf after all the cycles, is never higher than 10% (3% for Rr).
and the maximum reduction of Rf after all the cycles, is never higher than 10% (3% for Rr ).
The important role of the panel orientation is confirmed and is never lost due to cycling,
The important role of the panel orientation is confirmed and is never lost due to cycling,
although it is not comparable to the static data. Shape fixity in Direction 1 of both foams after the
although it is not comparable to the static data. Shape fixity in Direction 1 of both foams after
third thermo-mechanical cycle is comparable with Direction 2. Multiple cycles affect mainly material
the third thermo-mechanical cycle is comparable with Direction 2. Multiple cycles affect mainly
stiffness and strength rather than shape memory properties, and the static properties depend mainly
material stiffness and strength rather than shape memory properties, and the static properties depend
on these characteristics. This can be explained by the fact that static performances depend mainly on
mainly on these characteristics. This can be explained by the fact that static performances depend
the rigid part of the foam material, the PET crystals, which are altered by cyclic loading. In principle,
mainly on the rigid part of the foam material, the PET crystals, which are altered by cyclic loading.
fragmentation of crystallites, alteration of the orientation or a fragmentation of chains in initially pre-
In principle, fragmentation of crystallites, alteration of the orientation or a fragmentation of chains in
oriented asymmetric pore structure (in Direction 1), or the onset/alteration of pore wall buckling/pore
initially pre-oriented asymmetric pore structure (in Direction 1), or the onset/alteration of pore wall
wall breaking may be possible contributors to the shifting of the mechanical properties during the
buckling/pore wall breaking may be possible contributors to the shifting of the mechanical properties
first and subsequent cold deformation cycles. Further experiments in this direction may contribute
to an improved overall understanding of the observed data. However, shape memory behavior
Polymers 2018, 10, 115 13 of 16

Polymers
during 2018,
Polymersthe 10,
10,xxand
2018,first FOR PEER
PEERREVIEW
REVIEWcold deformation cycles. Further experiments in this direction13
FORsubsequent of
of15
13may 15
contribute to an improved overall understanding of the observed data. However, shape memory
depends on
on the
depends depends the soft
soft part of
of the
partsoft the foam
foam material, amorphous PET,
PET, which is
is almost unaltered after
behavior on the part of thematerial, amorphous
foam material, amorphous which
PET, whichalmost unaltered
is almost after
unaltered
thermo-mechanical
thermo-mechanical cycling.
cycling. Therefore,
Therefore, cell
cell structure
structure and
and morphology
morphology can
can be
be affected
affected by
by cycles
cycles but
but
after thermo-mechanical cycling. Therefore, cell structure and morphology can be affected by cycles
not their
notnot
their ability
ability to store
to store a new
a new shape
shape and to
andand recover
to recover the previous
thethe
previous one.
one.
but their ability to store a new shape to recover previous one.

Figure 13.
Figure13. Shape
13.Shape fixity
Shapefixity for
fixityfor low
forlow density
lowdensity (a)
density(a) and
(a)and high
andhigh density
highdensity (b)
density(b) foam
(b)foam samples
foamsamples in
samplesin all
inall directions
alldirections after
directionsafter
after
Figure
each
each thermo-mechanical
thermo-mechanical cycle.
cycle.
each thermo-mechanical cycle.

Figure 14.
Figure14.
Figure Shape
14.Shape recovery
Shaperecovery percentage
recoverypercentage after
percentageafter each
aftereach memory-recovery
eachmemory-recovery cycle
memory-recoverycycle for
cyclefor low-density
forlow-density foam
low-densityfoam (a)
foam(a) and
(a)and
and
high-density foam
high-densityfoam
high-density (b).
foam(b).
(b).

In
In order
order to
to confirm
confirm thethe correlation
correlation between
between foam
foam structure
structure and and measured
measured performances,
performances,
In order to confirm the correlation between foam structure and measured performances,
microscopic
microscopic observations were repeated at the end of each thermo-mechanical cycle. Results
observations were repeated at the end of each thermo-mechanical cycle. Results areare
microscopic observations were repeated at the end of each thermo-mechanical cycle. Results are
shown
shownin inFigures
Figures1515and
and16 16for
forlow-
low-andandhigh-density
high-densityfoam,
foam,respectively.
respectively.The TheFigures
Figuresshowshowthat
thatmany
many
shown in Figures 15 and 16 for low- and high-density foam, respectively. The Figures show that many
broken
broken cells
cellsand
and warped
warped cellcell walls
wallsappear
appear during
during cycling.
cycling. Because
Because of of this
thisdamage,
damage, itit was
was not
not possible
possible
broken cells and warped cell walls appear during cycling. Because of this damage, it was not possible
to
to estimate the geometrical ratio of cells, as was possible for un-cycled samples. The presence of
estimate the geometrical ratio of cells, as was possible for un-cycled samples. The presence of these
these
to estimate the geometrical ratio of cells, as was possible for un-cycled samples. The presence of these
ruptures
ruptures highlights
highlights that
that the
the sharp
sharp reduction
reduction of of static
static mechanical
mechanical properties
properties ofof low
low and
and high
high density
density
ruptures highlights that the sharp reduction of static mechanical properties of low and high density PET
PET
PET foams
foams could
could be
be affected
affected by by such
such events.
events. Applied
Applied memory-recovery
memory-recovery cycles cycles are
are very
very severe
severe due
due toto
foams could be affected by such events. Applied memory-recovery cycles are very severe due to the
the cold compression step. For this reason, material damage is associated with
the cold compression step. For this reason, material damage is associated with thermo-mechanical thermo-mechanical
cold compression step. For this reason, material damage is associated with thermo-mechanical cycles.
cycles.
cycles. This
This damage
damage is is not
not able
able to
to erase
erase foam
foam anisotropy,
anisotropy, but but stiffness
stiffness andand strength
strength are are reduced.
reduced.
This damage is not able to erase foam anisotropy, but stiffness and strength are reduced. Nevertheless,
Nevertheless, despite the cellular materials being partially damaged, their shape
Nevertheless, despite the cellular materials being partially damaged, their shape memory properties memory properties
despite the cellular materials being partially damaged, their shape memory properties are preserved
are
are preserved
preserved andand sufficient
sufficient toto guarantee
guarantee aa shape
shape recovery
recovery of of about
about 90%90% after
after three
three thermo-
thermo-
and sufficient to guarantee a shape recovery of about 90% after three thermo-mechanical cycles.
mechanical cycles.
mechanical cycles.
Polymers 2018, 10, 115 14 of 16
Polymers 2018, 10, x FOR PEER REVIEW 14 of 15
Polymers 2018, 10, x FOR PEER REVIEW 14 of 15

Figure 15. Low-density foam microstructure after each thermo-mechanical cycle: plane 1-2 (a); 1-3 (b)
Figure 15. Low-density foam microstructure after each thermo-mechanical cycle: plane 1-2 (a); 1-3 (b)
Figure
and 2-315.
(c).Low-density foam microstructure after each thermo-mechanical cycle: plane 1-2 (a); 1-3 (b)
and 2-3 (c).
and 2-3 (c).

Figure 16. High-density foam microstructure after each thermo-mechanical cycle: plane 1-2 (a); 1-3
Figure
(b) and 16.
2-3 High-density
(c). foam microstructure after each thermo-mechanical cycle: plane 1-2 (a); 1-3
Figure 16. High-density foam microstructure after each thermo-mechanical cycle: plane 1-2 (a); 1-3 (b)
(b) and 2-3 (c).
and 2-3 (c).
4. Conclusions
4. Conclusions
PET foams are not considered shape memory materials but experimental results show that they
4. Conclusions
have PET foams to
the ability arerecover
not considered
their shape shape memory
after materialsstep.
a cold memory but experimental
The combination results show that
of shape memorythey
PET foams are not considered shape memory materials but experimental results show that they
have the ability
behavior to recover
with other theirknown
already shape after a cold memory
properties of PET step.foams, Thecould
combination
open a of shape
wide memory
variety of
have the ability to recover their shape after a cold memory step. The combination of shape memory
behavior with
applications. PET other
foams already known
are light, stiff and properties
thermally ofinsulating,
PET foams, andcould
already open
haveaa wide variety of
great number
behavior with other already known properties of PET foams, could open a wide variety of applications.
applications.asPET
possibilities foams
a core are light,
material stiff and thermally
in transport, wind turbine insulating, and already
construction have a great
or shipbuilding. number of
Self-repairing
PET foams are light, stiff and thermally insulating, and already have a great number of possibilities as
possibilitiescould
properties as a corebe material in transport,
easily added becausewind turbine
of their SMconstruction or shipbuilding.
behavior. Moreover, PET foams Self-repairing
can be
a core material in transport, wind turbine construction or shipbuilding. Self-repairing properties could
propertiesfrom
produced could be easily
recycled PETadded
and can because
be 100% ofrecyclable.
their SM Nevertheless,
behavior. Moreover,the effectPET foams can be
of manufacturing
be easily added because of their SM behavior. Moreover, PET foams can be produced from recycled
produced on
processes from foamrecycled PET and needs
SM properties can beto 100% recyclable.
be correctly Nevertheless,
evaluated. the effect
For example, ofSM
the manufacturing
behavior of
PET and can be 100% recyclable. Nevertheless, the effect of manufacturing processes on foam SM
processes
foams on foamtoSM
is expected properties needs
be dependent on the to be correctly
production evaluated.
method and theFor example,
process the SM Results
conditions. behavior of
also
properties needs to be correctly evaluated. For example, the SM behavior of foams is expected to be
foams is that
indicate expected
the SM to be dependent
properties of onPET the production
foams method
are strongly and the process
anisotropic. However, conditions. Results PET
after heating, also
dependent on the production method and the process conditions. Results also indicate that the SM
indicate
foams that the SM
recovered properties
the initial shapeofup PET foams
to 98% inare
the strongly anisotropic. However,
third thermo-mechanical cycle. after heating, PET
properties of PET foams are strongly anisotropic. However, after heating, PET foams recovered the
foams Inrecovered
this study,the initial shape
commercial PET upfoams
to 98%were in the third
used thermo-mechanical
with two densities (65cycle. and 110 kg/m3). These
initial shape up to 98% in the third thermo-mechanical cycle.
foams Inwere
this study,
strongly commercial
anisotropicPET withfoams were usedalong
cell orientation with two densities direction
the extrusion (65 and 110 kg/m3). These
as underlined by
In this study, commercial PET foams were used with two densities (65 and 110 kg/m3 ).
foams were strongly
morphological anisotropic
observations. with cell orientation
Consequently, the largest along themodulus
elastic extrusionand direction as underlined
compressive strengths by
These foams were strongly anisotropic with cell orientation along the extrusion direction as underlined
morphological
were measured observations. Consequently,
along this extrusion direction.the The largest elastic
behavior modulus
of the foams and in thecompressive strengths
transverse directions
by morphological observations. Consequently, the largest elastic modulus and compressive strengths
were
is not measured along this
exactly isotropic but extrusion
the relateddirection.
anisotropy Theisbehavior
lower. Inofquantitative
the foams in the transverse
terms, the elasticdirections
modulus
were measured along this extrusion direction. The behavior of the foams in the transverse directions
is notthe
and exactly isotropicstrength
compressive but the related anisotropydirection
in the extrusion is lower.wereIn quantitative
2 times higher terms,thanthe elastic
that inmodulus
the two
isandnotthe
exactly isotropic but theinrelated
compressive anisotropy is lower. 2Intimes quantitativethan terms,thatthe elastic
orthogonal directionsstrength the extrusion
for the low-density foam, direction
and 3 were
times higher higher for the high-density in the two
foam.
modulus
orthogonal and the compressive
directions for the strength
low-density in the
foam, extrusion
and 3 direction
times were
higher 2
for times
the higher
high-density than that
foam.
Mechanical anisotropy partially decreases by thermo-mechanical cycling, but the effect of material
inMechanical
the two orthogonal
anisotropydirections
partially for the low-density
decreases foam, and 3 times
by thermo-mechanical cycling,higher
but for
the the high-density
effect of material
damage is superimposed. After the first cycle, the mechanical properties are reduced by about 50%.
foam.
damage Mechanical anisotropy
is superimposed. partially decreases by thermo-mechanical cycling, but theabout
effect50%.
of
Shape memory properties After
are lesstheaffected
first cycle,by the mechanical
anisotropy and properties are reduced by
by thermo-mechanical cycling. After
material damage
Shape thermo-mechanicalis superimposed.
memory properties cycles, are lessthe After
affected the first cycle,
by anisotropy the mechanical properties
and by thermo-mechanical are reduced by
three minimum shape recovery value is about 90%cycling. whichAfterwas
about
three 50%. Shape memory properties
thermo-mechanical are less affectedshape by anisotropy and is by thermo-mechanical
measured in the extrusion cycles,direction theforminimum
the high-density recovery
foam. Thisvalueis a good about
result90% which was
considering the
cycling.
measured After
in thethree thermo-mechanical
extrusion direction for cycles,
the the minimum
high-density foam. shape
This is recovery
a good valueconsidering
result is about 90% the
great number of broken and deformed foam cells after multiple thermo-mechanical cycles. At room
which was measured
great numberloads of broken in the extrusion
and deformed direction
foam PET for
cellsfoams the high-density
after multiple foam.
thermo-mechanical This is a good result
temperature, or impacts may damage with a consequent reductioncycles. At room
of mechanical
considering
temperature, theloads
greatornumber
impacts of broken and deformed foam
withcells after multiple thermo-mechanical
properties, but foams preservemay damage
excellent PET foams
shape recovery. a consequent
Anyhow, reduction
a larger number of mechanical
of thermo-
properties, cycles
mechanical but foams couldpreserve
be usefulexcellent
to better shape
understand recovery.
the SM Anyhow,
propertiesa larger
of PETnumber
foams and of thermo-
further
mechanical
studies maycycles could this
investigate be useful
aspect.toHowever,
better understand
the ability thetoSM properties
recover of PETshape
their initial foamsafterand afurther
room
studies may investigate this aspect. However, the ability to recover their initial shape after a room
Polymers 2018, 10, 115 15 of 16

cycles. At room temperature, loads or impacts may damage PET foams with a consequent reduction
of mechanical properties, but foams preserve excellent shape recovery. Anyhow, a larger number of
thermo-mechanical cycles could be useful to better understand the SM properties of PET foams and
further studies may investigate this aspect. However, the ability to recover their initial shape after
a room temperature deformation extends the possibilities of using such foams. In fact, these properties
are suitable for defining a new class of self-repairing materials where shape recovery is associated with
self-healing agents. From this perspective, a potential application of SM PET foams could be as the
core of self-repairing sandwich structures with composite skins for marine applications, as previously
evaluated by the authors [17].

Author Contributions: Loredana Santo, Denise Bellisario and Fabrizio Quadrini sheared, conceived and designed
the experiments, the analysis of the data and they wrote the paper.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Xie, T. Recent advances in polymer shape memory. Polymer 2011, 52, 4985–5000. [CrossRef]
2. Squeo, E.A.; Quadrini, F. Shape memory epoxy foams by solid-state foaming. Smart Mater. Struct. 2010, 19,
105002. [CrossRef]
3. Santo, L. Shape memory polymer foams. Prog. Aerosp. Sci. 2016, 81, 60–65. [CrossRef]
4. Santo, L.; Quadrini, F.; Mascetti, G.; Dolce, F.; Zolesi, V. Results of Shape Memory Foam Experiment.
Acta Astronaut. 2013, 91, 333–340. [CrossRef]
5. Sokolowski, W.; Tan, S.; Pryor, M. Light weight shape memory self-deployable structures for gossamer
applications. In Proceedings of the 45th AIAA/ASME Conference on Structures, Structural Dynamic and
Materials, Palm Springs, CA, USA, 19–22 April 2004; Volume 1660, pp. 1–10.
6. Sokolowski, W.; Tan, S. Advanced self-deployable structures for space applications. J. Spacecr. Rocket. 2007,
44, 750–754. [CrossRef]
7. Andersons, J.; Kirpluks, M.; Stiebra, L.; Cabulis, U. Anisotropy of the stiffness and strength of rigid
low-density closed- cell polyisocyanurate foams. Mater. Des. 2016, 92, 836–845. [CrossRef]
8. Dawson, J.R.; Shortall, J.B. The microstructure of rigid polyurethane foams. J. Mater. Sci. 1982, 17, 220–224.
[CrossRef]
9. Ridha, M.; Shim, V.P.W. Microstructure and tensile mechanical properties of anisotropic rigid polyurethane
foam. Exp. Mech. 2008, 48, 763–776. [CrossRef]
10. Hamilton, A.R.; Thomsen, O.T.; Madaleno, L.A.O.; Jensen, L.R.; Rauhe, J.C.M.; Pyrz, R. Evaluation of the
anisotropic mechanical properties of reinforced polyurethane foams. Compos. Sci. Technol. 2013, 87, 210–217.
[CrossRef]
11. Tey, S.J.; Huang, W.M.; Sokolowski, W.M. Influence of long-term storage in cold hibernation on strain
recovery and recovery stress of polyurethane shape memory polymer foam. Smart Mater. Struct. 2001, 10,
321–325. [CrossRef]
12. Tobushi, H.; Okumura, K.; Endo, M.; Hayashi, S. Thermomechanical properties of polyurethane-shape
memory polymer foam. J. Intell. Mater. Syst. Struct. 2001, 12, 283–287. [CrossRef]
13. Huang, W.M.; Lee, C.W.; Teo, H.P. Thermomechanical behavior of a polyurethane shape memory polymer
foam. J. Intell. Mater. Syst. Struct. 2006, 17, 753–760. [CrossRef]
14. Quadrini, F.; Bellisario, D.; Santo, L.; Del Gaudio, C.; Bianco, A. Shape Memory Foams of Microbial Polyester
for Biomedical Applications. Polym. Plast. Technol. Eng. 2013, 52, 599–602. [CrossRef]
15. Santo, L.; Bellisario, D.; Quadrini, F. Shape recovery of PET foams after cold compression. AIP Conf. Proc.
2016, 1769, 050005.
Polymers 2018, 10, 115 16 of 16

16. Lendlein, A.; Kelch, S. Shape memory polymers. Angew. Chem. Int. Ed. 2002, 41, 2034–2057. [CrossRef]
17. Santo, L.; Quadrini, F. Shape memory composite sandwich with self-healing properties for marine applications.
In Proceedings of the 20th International Conference on Composite Materials, Copenhagen, Denmark, 19–24 July 2015.

© 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).