Nanostructured morphology of a random P(DLLA-co-CL) copolymer
© Peponi et al; licensee Springer. 2012
Received: 30 September 2011
Accepted: 5 February 2012
Published: 5 February 2012
Skip to main content
© Peponi et al; licensee Springer. 2012
Received: 30 September 2011
Accepted: 5 February 2012
Published: 5 February 2012
The random architecture of a commercial copolymer of poly(DL-lactic acid) and poly(ε-caprolactone), poly(DL-lactide-co-caprolactone), has been characterized by chemical structure analysis from hydrogen-1 nuclear magnetic resonance results. Moreover, spherical nanodomains have been detected in the thin films of this copolymer obtained after solvent evaporation. These nanodomains studied by atomic force microscopy and transmission elecron microscopy grow progressively under annealing until they collapse and form a homogenous disordered structure. This is the first time that the nanostructure of random poly(DL- lactic acid)/poly-(ε-caprolactone) copolymers is revealed, representing one of few experimental evidences on the possible nanostructuration of random copolymers.
In the past years, the request of polymers for applications in the biomedical sector has grown drastically. Among others, poly-(lactic acid) [PLA], derived from renewable resources, is currently being used in a number of biomedical applications, such as in sutures, stents, drug delivery devices, and tissue engineering . Besides, the biodegradable petroleum-based polyester poly-(ε-caprolactone) [PCL] has also been widely studied [2–4]. However, these polymers are inappropriate for numerous uses where highly flexible biodegradable materials are required . Therefore, different strategies have been reported to properly modify the intrinsic properties of both polymers, including the use of additives and nanoparticles [6–8]. Another possible strategy is constituted by blending or copolymerizing them together, allowing the fabrication of a variety of biodegradable materials with improved properties in comparison with those of the parent homopolymers [3, 9]. Biodegradable PLA-blend-PCL materials can offer a wide variety of physical properties; the glassy PLA with a relatively high degradation rate shows better tensile strength, while the rubbery PCL with a much slower degradation rate shows better toughness . As reported in the scientific literature, the PCL/PLA blend can form typical immiscible morphologies (of few micrometer scales) such as spherical droplets, fibrous and co-continuous structures by varying the homopolymer composition [11, 12].
In the general case of copolymers, their final properties depend not only on their composition but also on their architecture (i.e., random, alternate, or block). Random and alternate copolymers are reported to be typically one-phase disordered materials with concentration fluctuations of a relatively short range . On the other hand, block copolymers present phase separation in the nanometer range, taking advantage of the covalent bonding between the immiscible constituting blocks which are able to self-assemble into well-defined ordered nanostructures with domain dimensions of 5 to 100 nm [14–17]. It is, therefore, not surprising that block copolymers have attracted worldwide attention of physicists, chemists, and engineers, developing numerous applications ranging from thermoplastic elastomers, adhesives, sealants, polymer blend compatibilizers, emulsifiers, and other recent advances in their medical applications [17–21]. The main features of these nanostructures, such as their composition, morphology, dimensions, spacing, and order are of primary significance for the chemical, mechanical, optical, and electromagnetic properties exhibited [22, 23].
Few studies have been reported on PLA/PCL copolymers with particular focus on their crystallization behavior [3, 5]. In this work, a commercial random copolymer based on poly(DL-lactic acid) [PDLLA] and PCL is studied, focusing the attention on its chemical architecture and nanostructured morphology.
The copolymer based on poly(DL-lactic acid) and poly(ε-caprolactone), poly(DL-lactide-co-caprolactone) [P(DLLA-co-CL)], was supplied by Sigma Aldrich (St. Louis, MO, USA) with a nominal 86 mol% of PDLLA. Their solubility parameters calculated based on the Hoftyzer-van Krevelen theory  are 27 and 25, respectively. Pure chloroform from Sigma Aldrich was used as solvent.
A solution of 0.01 g of P(DLLA-co-CL) in 5 mL of chloroform has been obtained by stirring the sample for 12 h at room temperature in a closed vessel.
The copolymer was characterized by hydrogen-1 nuclear magnetic resonance [1H-NMR] in a Varian Mercury 400 apparatus (Varian Inc., Palo Alto, CA, USA) at 400 MHz, using CDCl3 as solvent, and by a relaxation time between pulses of 5 s. The residual signal of the deuterated solvent was used as the internal reference (7.26 ppm).
Raman spectra were obtained using a Renishaw in via Reflex Raman System (Renishaw plc, Wotton-under-Edge, UK) employing a laser wavelength of 785 nm (laser power at sample = 10 mW; microscope objective = × 100). Spectra were recorded at room temperature after the exposure time of 10 s, which is necessary to decay the fluorescence.
The morphological features of the copolymer films were investigated using atomic force microscopy [AFM] and transmission electron microscopy [TEM]. The AFM is operating in a tapping mode [TM] with a scanning probe microscope (Nanoscope IV, Multimode TM from Veeco-Digital Instruments, Plainview, NY, USA). Height and phase images were obtained under ambient conditions with a typical scan speed of 0.5 to 1 line/s, using a scan head with a maximum range of 100 μm × 100 μm. The TEM measurements were performed on a JEOL JEM-2100 TEM instrument (JEOL Ltd., Akishima, Tokyo, Japan), with a LaB6 filament, with an operating voltage of 200 kV. For the morphological analysis by atomic force microscopy, a transparent thin film (ca. 300 nm) was obtained using a spin coater SCS P-6700 (Special Coating Systems, Inc., Indianapolis, IN, USA) at 4,000 rpm for 140 s followed by solvent evaporation at ambient conditions for 24 h, while for the TEM analysis, the solution, which was twice diluted, has been cast directly on the grid and evaporated at the same conditions.
Differential scanning calorimetry [DSC] measurements were performed with a Mettler-Toledo DSC-822 calorimeter (Mettler-Toledo, Inc., Columbus, OH, USA) calibrated with high-purity indium. All experiments were conducted under a nitrogen flow of 20 mL·min-1, using 7- to 10-mg samples in closed aluminum pans, in a temperature range from -90°C to 200°C with a rate of 10°C min-1, using a heating-cooling-heating cycle. The second heating scan was used to calculate the glass transition temperature [Tg] of the matrix.
Small-angle X-ray scattering [SAXS] measurements were taken at beamline BM16 at the European Synchrotron Radiation Facility (Grenoble, France). Samples were placed in between aluminum foils within a Linkam hot stage (Linkam Scientific Instruments, Tadworth, UK) and heated at 10°C min-1 while the SAXS spectra were recorded. Calibration of temperature gave a difference of approximately 7°C between the temperature reading at the hot stage display and the real temperature at the sample.
The main results on the physicochemical and thermal behaviors of the analyzed random copolymer P(DLLA-co-CL) have been the number average molecular weight [Mn] calculated by 1H-NMR and Tg obtained by DSC. In particular, the Mn was calculated through two different approaches: the first one, taking into account the terminal groups, produced a value of 21,000 g/mol, while the ratio between the caprolactone [CL] units and the initiator produced a value of 28,000 g/mol. From thermal analysis, we obtained a Tg of about 24°C.
Moreover, the measured PDLLA content obtained from 1H-NMR was 89.8 mol% in contrast with the value of PDLLA which was 86 mol% given by the supplier. This difference can be considered a small one and in the range of the possible deviation in different batches. In fact, the ratio of the lactic acid [LA] and CL signals allows a quite high accuracy for this calculation with a dispersion of 0.3% in three repetitions. We think that the supplier has provided an average value that could change from batch to batch without reporting the exact value for each batch.
If the copolymer is a di-block copolymer, the ratio of the signal of polymerized CL linked to LA molecules to the signal of terminal CL should be 1, and in our case, it is approximately 8.9. Furthermore, the ratio of the signal due to CL linked to LA to the signal of CL linked to CL is approximately 3.15, indicating the preponderance of isolated CL units in the polymer backbone. From these results, it is clear that the chemical structure of the copolymers approaches more likely the structure of a random copolymer. As the molar content of CL in the copolymer is low, 10.2%, it is reasonable to presume that CL units are isolated in between PLA units (-LA-CL-LA-) or form blocks of double CL units (-LA-CL-CL-LA-), with the existence of longer CL blocks being negligible. Then, from the signals due to CL linked to LA and to CL linked to CL, a 68 mol% of isolated CL units and 32 mol% of double CL units are calculated. Once the total CL and CL-CL units are determined, it is possible to calculate the mean length of the LA blocks which results to 12.
Summarizing the 1H-NMR results, the P(DLLA-co-CL) copolymer used in this study has the structure of a predominantly random copolymer with most of the CL units isolated in the copolymer backbone, therefore causing its inability to crystallize, and with blocks of polymerized LA units that are also unable to crystallize, producing an amorphous copolymer. Neither melting nor crystallization was found in the DSC thermogram (not shown), indicating the amorphous nature of the copolymer. The amorphous structure was also confirmed by SAXS (data not shown).
The nanostructuration observed is coherent with the computer simulation of the phase diagram of random copolymers carried out by Houdayer and Muller . Based on our knowledge, this is the first time that the nanostructuration of a P(DLLA-co-CL) copolymer is reported. Moreover, very few experimental demonstrations of the nanostructuration of random copolymers have been reported in the scientific literature [28, 29]. Taking into account the small amount of PCL, less than 15 mol%, it is assumed that spherical CL-enriched domains have been obtained. In this case, we consider that, because of the chemical nature of the copolymer, the higher affinity of chloroform for CL than for LA (as obtained by the solubility parameters calculated by the Hoftyzer and van Krevelen theory ) has favored the phase separation of CL-enriched domains in a matrix of pure LA or of LA with a lower CL content.
For longer annealing times, the nanostructured domains collapse and a disordered homogeneous structure is formed (not shown). It turns out that the phase segregation is characterized by a non-equilibrium geometrical rearrangement of the interfaces which tends to aggregate, minimizing the surface energy, and evolve to a dissolution of the nanostructured domains in the PDLLA-rich phase.
A P(DLLA-co-CL) copolymer has been studied in terms of chemical structure and morphological behavior. In particular, we demonstrated the random architecture of the copolymer with a LA/CL mole ratio of 89.8:10.2 with a number average molecular weight of 28,200 g/mol. From the morphological point of view, interesting nanostructured spherical domains have been obtained representing CL-enriched spheres with an average diameter of 48 nm. The annealing treatment enlarged progressively the CL-enriched domains, maintaining their spherical shape until they collapse and a homogeneous disordered structure is obtained. This is the first time that the nanostructure of random PDLLA/PCL copolymers is revealed, representing one of few experimental evidences on the possible nanostructuration of random copolymers.
This project has been supported by Projects MAT2010-21494-C03-03 and MAT2008-00619/MAT of the Spanish Ministry for Science and Innovation (MICINN). LP acknowledges the support of the Juan de la Cierva grant from MICINN (MICINN-JDC). The authors thank GEMPPO at the IEM-CSIC for the use of the TEM.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.