Biodegradable Thermoplastic Starch/Polycaprolactone Blends with Co-Continuous Morphology Suitable for Local Release of Antibiotics

Abstract

We report a reproducible preparation and characterization of highly homogeneous thermoplastic starch/pol(ε-caprolactone) blends (TPS/PCL) with a minimal thermomechanical degradation and co-continuous morphology. These materials would be suitable for biomedical applications, specifically for the local release of antibiotics (ATB) from the TPS phase. The TPS/PCL blends were prepared in the whole concentration range. In agreement with theoretical predictions based on component viscosities, the co-continuous morphology was found for TPS/PCL blends with a composition of 70/30 wt.%. The minimal thermomechanical degradation of the blends was achieved by an optimization of the processing conditions and by keeping processing temperatures as low as possible, because higher temperatures might damage ATB in the final application. The blends' homogeneity was verified by scanning electron microscopy. The co-continuous morphology was confirmed by submicron-computed tomography. The mechanical performance of the blends was characterized in both microscale (by an instrumented microindentation hardness testing; MHI) and macroscale (by dynamic thermomechanical analysis; DMTA). The elastic moduli of TPS increased ca four times in the TPS/PCL (70/30) blend. The correlations between elastic moduli measured by MHI and DMTA were very strong, which implied that, in the future studies, it would be possible to use just micromechanical testing that does not require large specimens.

Keywords: microindentation; micromechanical properties; poly(ε-caprolactone); polymer blends; structure–properties relations; thermoplastic starch.

Figure A1

The results of standard microbiological tube dilution tests, in which a defined piece of TPS/PCL/ATB sample (ATB dispersed in TPS phase) was put into a tube containing 5 mL of Mueller-Hinton broth (Oxoid, Brno, Czech Republic); each tube is then inoculated with a standardized bacterial suspension of the CCM 4223 Staphylococcus aureus and observed after 24 h. The turbid solution observed in positive control (A) and tubes containing TPS/PCL/ATB systems with lower concentrations of ATB (C–E) indicated the growth of bacteria, while the clear solution observed in negative control (B) and tubes containing TPS/PCL/ATB systems releasing higher concentrations of ATB (F–L) proved the bacteriostatic effect, i.e., inhibition of the bacterial growth.

Figure A2

SEM micrographs showing smoothed and etched surfaces of selected TPS/PCL blends prepared by two-step preparation procedure (upper row) and three-step preparation procedure (lower row).

Figure A3

Selected examples of the frequency-dependence of storage shear modulus G′ and of the loss shear modulus G″, as observed in oscillatory rheological investigations carried out at 120 °C with TPS/PCL blends; the examples demonstrate the different degree of liquid-like and solid-like character for the blends: (A) TPS/PCL (20/80); (B) TPS/PCL (30/70); (C) TPS/PCL (40/60); and (D) TPS/PCL (80/20).

Figure A4

SEM micrographs of fracture surfaces of TPS/PCL blends. The materials were fractured while submerged in liquid nitrogen. The micrographs demonstrate that the fracture propagated mostly through both phases (and not along the interface), which indicates good interfacial adhesion between TPS and PCL components.

Figure A5

Selected micromechanical properties of the TPS and PCL and TPS/PCL blends in the form of bar charts with standard deviations: (A) indentation hardness, (B) Martens hardness, (C) indentation modulus and (D) indentation creep.

Figure A6

Complete results of DMTA measurements for all TPS/PCL blends in temperature range −80–160 °C: (A) storage modulus, (B) loss modulus, (C) absolute value of complex modulus, and (D) damping factor.

Figure A7

Correlation matrix table showing Pearson’s coefficients r for all pairs of experimentally determined properties in macro- and microscale. The table is presented as a heatmap (darker color means stronger correlation). The properties in the table are: storage modulus, loss modulus and damping factor from dynamic mechanical thermal analysis (DMTA/G′, DMTA/G″ and DMA/tan(δ)) and elastic modulus, indentation hardness, Martens hardness, indentation creep, and elastic part of indentation work from MHI measurements (MHI/E_IT, MHI/H_IT, MHI/H_M, MHI/C_IT and MHI/η_IT). We note that the correlation matrix is symmetric with respect to the main diagonal, which contains values equal to 1 (autocorrelations).

Figure 1

Principle of MHI measurements, which were employed in this work: (a) Micromechanical properties were deduced from the experimental F–h curves (where F is the indenter loading force and h is the indenter penetration depth) by means of (b) formulas and (c) relations containing experimental parameters, such as maximum loading force (F_max), slope at the beginning of the unloading curve (S), penetration depths at the beginning and end of the maximal load (h₁ and h₂), and areas under loading and unloading curve (W_elast and W_total). Additional parameter, contact depth (h_c), was calculated in terms of the Oliver and Pharr theory and employed in the calculation of E_IT and H_IT, as described in our previous work [41,42]. The figure shows real, representative F–h curves of TPS/PCL systems, which the illustrate substantial changes of all studied properties as a function of composition.

Figure 2

Torque moments as a function of melt-mixing time and maximum temperatures during melt-mixing of neat TPS (black line) and TPS/PCL (70/30) blend (blue line). The initial increase of the torque moments corresponds to the filling of the mixing chamber, which was heated to 120 °C.

Figure 3

Absolute value of the complex viscosity (|η*|) as a function of oscillatory shear angular frequency (ω) during the oscillatory shear measurements for all TPS/PCL blends.

Figure 4

SEM micrographs showing smoothed surfaces of the TPS/PCL blends; the TPS phase was etched off by hydrochloric acid.

Figure 5

Phase co-continuity in TPS/PCL blends: (a) a theoretical prediction of the TPS/PCL composition exhibiting co-continuous morphology and (b) an experimental proof of co-continuity in the TPS/PCL (70/30) blend by means of subμ-CT measurement. The theoretical prediction (a) suggested that co-continuous morphology should appear at higher concentrations of TPS (as explained in Section 3.1.4). The experimental verification of the co-continuity of the TPS/PCL (70/30) blend (b) is displayed in the form of a 3D reconstruction of 100 × 100 × 100 μm³ volume of the sample measured by subμ-CT. The model illustrates co-continuity in both phases. The etched-off phase (TPS; shown in blue) forms interconnected pores. The connections among the pores can be observed in the thin slices on the right.

Figure 6

Micromechanical properties of TPS, PCL, and of the TPS/PCL blends (red points) vs. theoretical predictions (black lines). The plots show (A) indentation hardness, (B) indentation modulus, (C) indentation creep and (D) elastic work of indentation. The theoretical predictions comprise linear model (LIN; the final blend property is a linear combination of the blend component properties) and equivalent box model (EBM; a simple predictive scheme for isotropic polymer blends, which considers phase continuity and interfacial adhesion). The models are described in Section 3.2.1.

Figure 7

Results of DMTA: storage modulus (G′) is a function of temperature (T) for all TPS/PCL blends. The inset shows detail of the region around room temperature.

Figure 8

Scatterplot matrix graph showing correlation between selected micromechanical (G′) and micromechanical properties (E_IT, H_IT and H_M). Diagonal elements of the scatterplot matrix graph show distribution of the measured quantities, whereas off-diagonal elements show correlations between each pair of quantities. The translucent bands around the regression lines represent 95% confidence interval of the regression estimate. Moreover, all off-diagonal plots show the values of Pearson’s correlation coefficient r in the upper right corner.