Crystallization of Polymers Under 1D Confinement
235
where is a fitting parameter, convoluted to the value ofξ, taking into account
nucleation density. The results obtained for PET imply that ξ 1. While the physics
behind Eq. (5) is intuitive, the lack of experimental points below 20 nm, that is the
region where a finite value of ξ could be determined, cannot grant validation of the
framework.
To overcome this issue, we will consider the work by Martinez-Tong et al. [24] on
thin films of poly(L-lactide acid) PLLA, a system forming less dense adsorbed layers,
which should correspond to measurable ξ values. Before discussing on experiments
on this biodegradable polymer, we remark that it is possible to reduce the adsorbed
amount of thin films of PET by capping with a layer of another polymer [59]. The
physics behind the reduction in adsorbed amount was discussed by Simavilla and
coworkers [58].
PLLA nanocapacitors for dielectric measurements were prepared via a similar
procedure to that described for PET and PHB. Here will consider the materialspecific details. Thin films of thickness ranging between 300 and 8 nm of poly(Llactide) (PLLA, M w = 67,000 g/mol, PDI < 1.4, Sigma-Aldrich) were spin-coated
(3000 rpm, chloroform 99.9%) on top of the lower electrode. In order to guarantee
the evaporation of the solvent, the spin-coated films were left on a hot plate at 333 K
for one hour. Finally, a second aluminum layer (≈50 nm) was evaporated on top
of the polymer film. The annealing procedure resulted in semicrystalline samples.
Amorphous films were obtained by heating the nanocapacitors at 453 K (liquidus
temperature) for 30 s followed by cooling on a cold plate (278 K). Samples were
stored at room temperature before measurements. Dielectric relaxation spectroscopy
experiments were conducted isothermally at 343 K. Sweeping frequencies from
10 MHz to 0.1 Hz requires about 150 s, a measuring time that is much lower than
the crystallization time of any sample studied in this work. Such a condition permits
the investigation of the kinetics of crystallization in real time.
Same as in the case of PET, a preliminary study permitted to identify the most
convenient crystallization temperature to follow within a reasonable time the whole
thickness range. The data in Fig. 8 reveal a huge increase in glass stability upon
reduction of the thickness. Figure shows the frequency dependence of the dielectric
loss for two PLLA thin films of thicknesses h = 150–8 nm, respectively, representative of a bulk-like and an interfacial behavior. The top panels show the first dielectric
sweep (t = 0 s), where the polymers are in the amorphous state, while the lower
panels show the response after 25 min for the thick film and after 1600 min for the
thinner film.
While the dielectric signal of the interfacial layer is almost unaltered by annealing
for almost 3 h, the strength of the segmental peak of the thick film decreased by far.
This system shows a lot of similarities to PET. The crystallization time increases
upon reduction of the thickness, the segmental time is constant down to 30 nm and
then sharply increases, the kinetics of reduction in ε; see Fig. 9, can be described
by Eq. (5). In the case of thicker films, the drop in dielectric strength is mostly
related to crystallization, while for the thinner samples the role of adsorption becomes
predominant. Differently than PET, the crystallization time of thin films of PLLA
does not reach values outside of the experimental time window in the case of the
235
where is a fitting parameter, convoluted to the value ofξ, taking into account
nucleation density. The results obtained for PET imply that ξ 1. While the physics
behind Eq. (5) is intuitive, the lack of experimental points below 20 nm, that is the
region where a finite value of ξ could be determined, cannot grant validation of the
framework.
To overcome this issue, we will consider the work by Martinez-Tong et al. [24] on
thin films of poly(L-lactide acid) PLLA, a system forming less dense adsorbed layers,
which should correspond to measurable ξ values. Before discussing on experiments
on this biodegradable polymer, we remark that it is possible to reduce the adsorbed
amount of thin films of PET by capping with a layer of another polymer [59]. The
physics behind the reduction in adsorbed amount was discussed by Simavilla and
coworkers [58].
PLLA nanocapacitors for dielectric measurements were prepared via a similar
procedure to that described for PET and PHB. Here will consider the materialspecific details. Thin films of thickness ranging between 300 and 8 nm of poly(Llactide) (PLLA, M w = 67,000 g/mol, PDI < 1.4, Sigma-Aldrich) were spin-coated
(3000 rpm, chloroform 99.9%) on top of the lower electrode. In order to guarantee
the evaporation of the solvent, the spin-coated films were left on a hot plate at 333 K
for one hour. Finally, a second aluminum layer (≈50 nm) was evaporated on top
of the polymer film. The annealing procedure resulted in semicrystalline samples.
Amorphous films were obtained by heating the nanocapacitors at 453 K (liquidus
temperature) for 30 s followed by cooling on a cold plate (278 K). Samples were
stored at room temperature before measurements. Dielectric relaxation spectroscopy
experiments were conducted isothermally at 343 K. Sweeping frequencies from
10 MHz to 0.1 Hz requires about 150 s, a measuring time that is much lower than
the crystallization time of any sample studied in this work. Such a condition permits
the investigation of the kinetics of crystallization in real time.
Same as in the case of PET, a preliminary study permitted to identify the most
convenient crystallization temperature to follow within a reasonable time the whole
thickness range. The data in Fig. 8 reveal a huge increase in glass stability upon
reduction of the thickness. Figure shows the frequency dependence of the dielectric
loss for two PLLA thin films of thicknesses h = 150–8 nm, respectively, representative of a bulk-like and an interfacial behavior. The top panels show the first dielectric
sweep (t = 0 s), where the polymers are in the amorphous state, while the lower
panels show the response after 25 min for the thick film and after 1600 min for the
thinner film.
While the dielectric signal of the interfacial layer is almost unaltered by annealing
for almost 3 h, the strength of the segmental peak of the thick film decreased by far.
This system shows a lot of similarities to PET. The crystallization time increases
upon reduction of the thickness, the segmental time is constant down to 30 nm and
then sharply increases, the kinetics of reduction in ε; see Fig. 9, can be described
by Eq. (5). In the case of thicker films, the drop in dielectric strength is mostly
related to crystallization, while for the thinner samples the role of adsorption becomes
predominant. Differently than PET, the crystallization time of thin films of PLLA
does not reach values outside of the experimental time window in the case of the