Preparation of macroporous biodegradable poly(L-lactide-co-epsilon-caprolactone) foams and characterization by mercury intrusion porosimetry, image analysis, and impedance spectroscopy

Two poly(L-lactide-co-epsilon-caprolactone) random copolymers containing 5 and 40 mol % of epsilon-CL, namely P(LA-co-CL(5)) and P(LA-co-CL(40)), respectively, have been made macroporous by freeze-drying solutions in dimethylcarbonate. Most of the freeze-dried foams, prepared by varying polymer concentration and cooling rate, exhibited two main pore populations: (1). longitudinally oriented tube-like macropores with diametersor=100 microm, and (2). interconnected micropores (10-100 microm). Pore characteristics, including macropore density, mean diameter, and interdistance, as well as micropore density, area, and shape, were determined by image analysis of scanning electron micrographs in order to study the influence of processing and formulation parameters on foam structure and properties. The pore orientation and the 3-D texture also were studied by image analysis and impedance spectroscopy. In the case of the P(LA-co-CL(5)), the macropore diameter increased with the cooling rate while the micropore diameter decreased. The micropores also became more circular when the cooling rate was increased. The pore size and morphology of the P(LA-co-CL(40)) were quite unchanged by varying the cooling rate. All the other conditions being the same, the P(LA-co-CL(5)) foams were better organized than the P(LA-co-CL(40)) foams, and pore orientation was improved at the higher cooling rate. Pore size and morphology also can be controlled by changing the polymer concentration (Cp), as we showed by studying P(LA-co-CL(5)) foams prepared by freeze-drying solutions in the 1-10 w/v % Cp range. Macropore density, average diameter, and interdistance of P(LA-co-CL(5)) foams increased with Cp, but the micropore characteristics remained almost unchanged no matter the Cp. The reliability of the characterization methods has been discussed, with special attention to mercury intrusion porosimetry, which is used primarily for measurement of pore volume and pore size distribution. However, this technique is reported here as a destructive and unreliable method for the characterization of fragile P(LA-co-CL(40)) foams. This study shows that image analysis and impedance spectroscopy can give reliable information relative to the pore morphology and anisotropy of freeze-dried foams.