11180
Langmuir 2005, 21, 11180-11184
Hydrolytic Behavior of Enantiomeric Poly(lactide) Mixed Monolayer Films at the Air/Water Interface: Stereocomplexation Effects Won-Ki Lee Division of Chemical Engineering, Pukyong National University, Busan 608-739, Korea
Tadahisa Iwata Polymer Chemistry Laboratory, RIKEN Institute, 2-1 Hirosawa, Wako-shi, Saitama 351-0198, Japan
Joseph A. Gardella, Jr.* Department of Chemistry, State University of New York at Buffalo, New York 14260-3000 Received April 27, 2005. In Final Form: August 24, 2005 The Langmuir film balance technique was used to determine the hydrolytic kinetics of monolayers of the stereocomplex formed from mixtures of enantiomeric polylactides, poly(L-lactide) (L-PLA) and poly(D-lactide) (D-PLA), spread at the air-water interface. The present study investigated parameters such as degradation medium, mixture composition, and time on the relative degradation rate. The π-A isotherms of monolayers of the mixtures provide clear evidence for the presence of a stereocomplex; the isotherms of monolayers of individual polyenantiomer show a transition at about 8.5 mN/m, whereas the transition of monolayers containing a stereocomplex formed from the equimolar mixture shifted to higher surface pressure, about 11 mN/ m. The rate of hydrolysis was recorded by a change in occupied area when the monolayer is maintained at a constant surface pressure. The hydrolysis of the mixture monolayers under basic conditions was slower than that of individual polyenantiomer monolayers, depending on the composition or the degree of complexation. In the presence of proteinase K, the enzymatic hydrolysis rate of mixture monolayers with >50 mol % L-PLA was much slower than that of the single-component L-PLA monolayer. The monolayers formed from mixtures with e50 mol % L-PLA did not show any change of occupied areas. This result is explained by the inactivity of D-PLA and stereocomplexed chains to the enzyme. From both results, it can be concluded that the retardation of the hydrolysis of mixture monolayers is mainly due to a strong interaction between D- and L-lactide unit sequences, which prevents the penetration of water or enzyme into the bulk.
Introduction Although it has been widely recognized that long-term stability of synthetic polymers is an attractive property, there are increasing demands for the use of degradable polymers that do not require long-term strength retention, in particular to minimize polymer waste management caused by synthetic nondegradable polymers and for various medical applications.1-3 The most widely researched biodegradable polymers are aliphatic polyesters, e.g., poly(glycolide) (PGA), poly(lactide) (PLA), poly(3hydroxybutyrate), and poly(-caprolactone), which have nontoxic and biocompatible properties both as polymers and as their degradation products. Among them, PLAs have attracted much attention because lactic acid, which is converted to the cyclic dimer lactide, can be produced from renewable resources such as corn starch and sugar beets and high-molecular-weight PLAs can be produced by the ring-opening polymerization of lactides.3 Despite their structural similarity to PGA, the PLAs are quite different in chemical, physical, and mechanical properties * To whom correspondence should be addressed. E-mail:
[email protected]. (1) Doi, Y. Microbial Polyesters; VCH Publishing: New York, 1990. (2) Scott, G.; Gilead, D. Degradable Polymers; Chapman & Hill: London, 1995. (3) Piskin, E. In Degradable Polymers; Scott, G., Ed.; Kluwer Academic Publishers: Dordrecht, The Netherlands, 2002.
because of the presence of a pendent methyl group on the a carbon. This structure causes chirality at the a carbon, and thus, L, D, and D,L isomers are possible. Modulation of the stereochemical composition leads to PLAs with quite different properties: isotactic crystalline L-PLA, D-PLA, and amorphous D,L-PLA.1-7 Crystalline L-PLA and D-PLA form left-handed 103 helices8 and right-handed 103 helices,9 respectively, known as the R form. Since Ikada et al.9 reported the formation of stereocomplexes between enantiomeric L-PLA and D-PLA, numerous studies have been performed on the crystallization, physical properties, and crystalline structure of the stereocomplex.7,9-13 In the stereocomplex crystallites formed as a result of stereocomplexes, equimolar L- and D-lactide unit sequences are packed side-by-side and form 31 helices in an orthorhombic (4) Chabot, F.; Vert, M.; Chapelle, S.; Granger, P. Polymer 1983, 24, 536. (5) Li, S.; Vert, M. Macromolecules 1994, 27, 3107. (6) Reeve, M. S.; McCarthy, S. P.; Downey, M. J.; Gross, R. A. Macromolecules 1994, 27, 825. (7) Tsuji, H.; Miyauchi, S. Biomacromolecules 2001, 2, 597. (8) DeSantis, P.; Kovacs, A. J. Biopolymers 1968, 6, 209. (9) Ikada, Y.; Jamshidi, K.; Tsuji, H.; Hyon, S. H. Macromolecules 1987, 20, 904. (10) Brizzolara, D.; Cantow, H. J.; Diederichs, K.; Keller, E.; Domb, A. J. Macromolecules 1996, 29, 191. (11) Cartier, L.; Okihara, T.; Lotz, B. Macromolecules 1997, 30, 6313. (12) Li, S. Macromol. Biosci. 2003, 3, 657. (13) Urayama, H.; Kanamori, T.; Fukushima, K.; Kimura, Y. Polymer 2003, 44, 5635.
10.1021/la051137b CCC: $30.25 © 2005 American Chemical Society Published on Web 10/04/2005
Hydrolytic Behavior of PLA Films at the Air/Water Interface
cell (β form).10,11 The stereocomplex exhibits a melting temperature higher by 50 °C than that of each homopolymer. Only a few studies, however, deal with the degradation behavior of stereocomplexation crystallites even though morphology plays a critical role in degradation phenomena.7,14 Tsuji studied the hydrolytic degradation rate of PLA stereocomplex films, reporting that the stereocomplexes were more hydrolytically stable in a phosphate buffer solution than each PLA crystal.14 In general, it has been known that hydrolysis is initially restricted to the amorphous phase and to the fringes of the crystallites. However, the hydrolysis kinetics of homocrystallized enantiomeric PLA blend films is faster than that of amorphous films.14 Recently, Serizawa et al. reported the alkaline hydrolysis of enantiomeric PLA stereocomplexes by using layer-by-layer assembly and quartz crystal microbalance techniques.15 However, they concluded the opposite of Tsuji:14 that the stereocomplex is more easily hydrolyzed than each homocrystal and that its hydrolytic rate is similar to that of amorphous PLA. The difference in the reported hydrolytic behaviors in these two studies is probably due to different experimental methods. The Langmuir film balance technique is particularly useful in studying molecular packing status in two dimensions. Bourque et al. reported that the Langmuir film of an equimolar mixture of L-PLA and D-PLA spread at the air-water interface forms helical stereocomplex structures.16 A similar behavior was observed in monolayers of isotactic and syndiotactic poly(methyl methacrylate) mixtures.17 In our previous studies,18 the Langmuir film balance method was successfully applied to determine the degradation behavior of various polyester monolayers. In the present study, we have investigated the alkaline hydrolysis of enantiomeric mixtures of L-PLA and D-PLA at the air/water interface to obtain a fundamental understanding of degradation kinetics of the stereocomplexes. Experimental Section Materials. The L-PLA and D,L-PLA used in this study were synthesized by ring-opening polymerization of L-lactide and D,L-lactide, respectively, in vacuum-sealed glass ampules using stannous octoate as a catalyst. All materials were purified by precipitation in excess methanol from a chloroform solution and dried in a vacuum. D-PLA was supplied by Mitsubishi Chemical Co. The characteristics of the homopolymers and the mixture used in this study are listed in Table 1. Each solution of D-PLA and L-PLA in chloroform was separately prepared to have two concentrations of 2 µmol in monomer units per milliliter and 2 µmol in monmomer units per milliliter for preparations of monolayers and spin-cast thin films, respectively. The mixture solutions of D-PLA and L-PLA were prepared from each homopolymer solution. To obtain well-stereocomplexed films, thin films (1 × 1 cm2) were heated to 250 °C and then annealed at 190 °C for 3 h before use. Degradations. The subphase pH for alkaline hydrolysis of Langmuir monolayers was controlled by adding NaOH to the subphase. The enzymatic hydrolysis of monolayers and spincast thin films was carried out at 20 °C in 25 mM Tris-HCl buffer (pH 8.5) with proteinase K from the mold T. album (Sigma (14) (a) Tsuji, H. Polymer 2000, 41, 3621. (b) Tsuji, H.; Carpio, C. A. D. Biomacromolecules 2003, 1, 7. (c) Tsuji, H. Biomaterials 2003, 24, 537. (15) Serizawa, T.; Arikawa, Y.; Hameda, K.; Yamashita, H.; Fujiwara, T.; Kimura, Y.; Akashi, M. Macromolecules 2003, 26, 1762. (16) Bourque, H.; Laurin, I.; Pezolet, M.; Klass, J. M.; Lennox, R. B.; Brown, G. R. Langmuir 2001, 17, 5842. (17) Brinkhuis, R. H. G.; Schouten, A. J. Macromolecules 1992, 25, 2725. (18) (a) Lee, W. K.; Gardella, J. A., Jr. Langmuir 2000, 16, 3401. (b) Lee, W. K.; Nowak, R. W.; Gardella, J. A., Jr. Langmuir 2002, 18, 2309.
Langmuir, Vol. 21, No. 24, 2005 11181 Table 1. Characteristics of the Poly(lactide)s Used in This Study code L-PLA D-PLA D,L-PLA D-PLA20a
Mn
Mw/Mn
Tg (°C)
72 000 65 000
1.7 2.8
58 58
176 167
64 000 -
2.2 -
52 57
173, 213
Tm (°C)
remarks synthesized Mitsubishi Chem. Co. synthesized D-PLA/L-PLA 20/80 by wt
a Digital gives the D-PLA weight percentage in D-PLA/L-PLA mixtures.
Chemical Co.), which is a well-known enzyme for catalyzing the hydrolysis of long L-lactide unit sequences. The thin films were placed in a sterilized plastic cuvette containing 1 mL of buffer solution. The degradation was started by the injection of 0.1 mL of 2 mg/mL of a proteinase K into the sterilized plastic cuvette. The degraded thin films on a substrate were carefully washed with an excess of distilled water and dried in a vacuum. Measurements. The thin films were subjected to measurements of a wide-angle X-ray diffraction spectrometer (Rigaku RINT-2500) and a differential scanning calorimeter (PerkinElmer Pyris 1). Monolayer properties were studied by using a computer-controlled KSV 2200 film balance held at 20 °C. A compression rate of 30 cm2/min was used throughout. The surface pressure could be measured with an accuracy of 0.1 mN/m. The water was purified with a Millipore Mega-Pure system (MP-6A). Approximately 1600 mL of purified water was used as a subphase liquid (pH 7.3). After spreading, the solvent was allowed to evaporate for 1 min to minimize the hydrolysis during solvent evaporation and compression. The time for solvent evaporation was determined on the basis of the reproducibility of the L-PLA monolayer on pure water taking advantage of very slow hydrolysis under measured conditions. AFM measurements of thin films cast on Si wafers were conducted using a scanning force microscope (SPA 300 instrument with SPI 3700 controller; Seiko Instrument Co.) at room temperature. The cantilever used in this study is triangular with a microfabricated Si3N4 microtip (Olympus Co.) and a spring constant of 0.022 N/m. The scanning direction was perpendicular to the long axis of the cantilever.
Results and Discussion Thin Films. Figure 1A,B shows the DSC curves and X-ray diffraction patterns, respectively, for L-PLA, D-PLA. and their mixture films after thermal treatment. It is clearly seen that both homopolymers give a single endothermic peak around 170 °C, in good agreement with other reports, whereas a new peak appears near 220 °C for the mixture films.3,9 When the mixture ratio is 50/50 by weight, the peak at 170 °C disappears, and the peak at 220 °C becomes sharper. This is consistent with earlier reports of the formation of stereocomplexes in which equimolar L- and D-lactide unit sequences are packed side-byside.7,9-13 The new crystalline structure was supported by X-ray diffraction pattern: 11.8°, 16°, and 18.5° at 2θ for L-PLA and D-PLA, in contrast to 21° and 24° for mixture films (indicated by arrows in Figure 1B). To investigate the enzymatic degradation of enantiomeric PLA mixture thin films, morphological measurements were conducted. Figure 2 shows the AFM topographic images of L-PLA and D-PLA50 thin films before and after enzymatic hydrolysis of 1 h at 20 °C in 25 mM Tris-HCl buffer with proteinase K. As-prepared L-PLA (A) and D-PLA50 (C) films showed volume-filled fibrillike and small spherulite-like crystalline morphologies, respectively, throughout the film surface. However, enzymatic attack leads to different morphologies at the surfaces of L-PLA and D-PLA50 films. The surface of the L-PLA film shows partial erosion due to the enzymatic attacks (B). The remaining fibrils indicate that the
11182
Langmuir, Vol. 21, No. 24, 2005
Lee et al.
Figure 3. AFM line profile data of various mixture films (A) before and (B) after enzymatic degradation of 1 h. Figure 1. (A) DSC thermograms and(B) XRD patterns of mixture films of L-PLA and D-PLA. The dotted oval indicates the new melting peaks from stereocomplexes, and additional peaks due to the stereocomplexes are shown as arrows.
Figure 4. Area ratio vs time for PLA monolayer films maintained at 7 mN/m on a subphase of pH 10.5.
enzymatic degradation. The surface roughness of the L-PLA film after 1 h of degradation was much higher than that of the as-prepared sample, whereas D-PLA50 did not
Figure 2. AFM topographic images of as-prepared (A) L-PLA and (C) D-PLA50 films; B and D correspond to the same samples after enzymatic degradation of 1 h.
preferential enzymatic attack starts in disordered chains inserted between the crystalline fibrils. However, no major difference was found on the surface morphologies of the D-PLA50 films after enzymatic attack of 1 h (D). This result can be attributed to the presence of enzymatically inactive stereocomplexes at the film surface. It is known that enzymatic hydrolysis of biodegradable semicrystalline polymers preferentially occurs in the less-ordered or amorphous region. Therefore, the change of film surface roughness induced by enzyme attacks can provide information on the erosion of surface amorphous regions if the erosion of surface crystal regions is negligible during an initial erosion process. Figure 3A,B shows the line profile data from AFM topographic images for L-PLA and D-PLA50 films, respectively, before and after 1 h of
show any difference in the surface roughness before and after degradation. Thus, it seems reasonable that the surface enzymatic degradation of mixture films is retarded by the formation of stereocomplexes. Alkaline Degradation of Monolayers. In our previous studies,18 we showed that the Langmuir trough is a useful tool for measuring the conformational changes and hydrolytic kinetics of hydrolyzable monolayers on a molecular scale. Because the low-molecular-weight monomers, dimers, trimers, and some oligomers generated by the hydrolysis of polyester monolayers are soluble and dissolve into the subphase water, the change in area occupied by monolayers on an alkaline subphase reflects the dissolution of degradation products of the monolayers into the subphase and, thus, the hydrolytic stability. We first consider the effects of pH on the hydrolytic behavior of single-component monolayers at pH’s of 7.3 and 10.5. Figure 4 shows the change of area plotted as the area ratio (A/A0, where A0 and A represent the areas occupied by the film at times 0 and t, respectively) occupied by monolayers of D-PLA, L-PLA, and D,L-PLA as a function of time at a constant surface pressure of 7 mN/m on the subphase of pH 10.5. The initial time, t ) 0, was established when the surface pressure reached a desired stable value, meaning that the effect of dissolving low-molecular-weight hydrolysis products (monomers, dimers, oligomers) was neglected during the compression. Under the conditions studied here, the hydrolytic degradation of D,L-PLA is faster than those of L-PLA and D-PLA, even though the
Hydrolytic Behavior of PLA Films at the Air/Water Interface
Figure 5. Pressure-area isotherms of (A) L-PLA, (B) D-PLA50, and (C) D-PLA20 monolayer films on subphases of pH’s 7.3 and 10.5.
occupied areas of these monolayers at a constant surface pressure of 7 mN/m are nearly equivalent. This trend results from the different configurational structures of the D- and L-PLA monolayers formed at the air/water interface. Some of the differences in hydrolytic rate between L-PLA and D-PLA can be attributed to the difference in their Mw/Mn ratio.14 Figure 5 shows pressure-area (π-A) isotherms of L-PLA, D-PLA50, and D-PLA20 monolayers on subphases of pH 7.3 and 10.5. At low surface pressure, the areas occupied by the monolayers of L-PLA on the subphase of pH 10.5 are larger than those on the subphase of pH 7.3. This can be explained by inter- and intramolecular repulsion due to the ionization of the monolayers.19 The plateau region at ca. 8.5 mN/m appears at pH 7.3 but is not detectable at pH 10.5. The plateau region in the isotherm of the L-PLA monolayer on pure water was interpreted as a phase transition and the formation of a three-dimensional structure by Ivanova et al.19 The areas occupied by the monolayers at pH 10.5, however, were larger than those on pure water above the surface pressure of the plateau region. This means that the monolayer does not form a three-dimensional structure at high surface pressure because the hydrolysis of the film is accompanied by an increase of its hydrophilicity (by the anchoring effect,18 which means that the functional end groups generated by the hydrolysis of ester groups increase their hydrophilicity). Similar behaviors were observed in the surface pressure-area isotherms of D-PLA monolayers. However, the isotherms of D-PLA50 monolayers on subphases of pH 7.3 and 10.5 exhibit a distinctly different behavior in the plateau region as compared to those of the homopolymers, as shown in Figure 5B. A plateau region observed for the D-PLA50 monolayers on pure water has shifted to higher surface pressure, about 11 mN/m, suggesting a conversion to a newly formed stable structure, i.e., complexes between D-PLA and L-PLA. The isotherms of D-PLA20 monolayers in Figure 5C, however, show two plateau regions at surface pressures of 8.5 and 11 mN/m, corresponding to transitions of L-PLA and stereocomplexed monolayers, respectively. A very similar behavior was observed for D-PLA80 monolayers. Thus, equimolar amounts of D- and L-PLA form stereocomplexes in the monolayers and leave excess D- or L-PLA crystals. To investigate the alkaline hydrolysis of stereocomplexes of enantiomeric PLAs, the kinetic curves of mixture monolayers of D-PLA and L-PLA at a constant surface pressure of 7 mN/m on a subphase of pH 10.5 were recorded, as shown in Figure 6. It is clear that the rate of hydrolytic degradation of the D-PLA50 monolayer is much slower than the arithmetic average calculated from Figure 4. This can be explained by the hydrolytic stability (19) . Ivanova, T.; Panaiotov, I.; Boury, F.; Benoitm J. P.; Verger, R. Colloids Surf. B: Biointerfaces 1997, 8, 217.
Langmuir, Vol. 21, No. 24, 2005 11183
Figure 6. Area ratio vs time for mixture monolayer films maintained at 7 mN/m on a subphase of pH 10.5.
Figure 7. Effect of mixing methods on area ratio for mixture monolayer films maintained at 7 mN/m on a subphase of pH 10.5.
or the resistance to hydrolysis of the stereocomplex. The has two structures: one is from that is not in the stereocomplex form, and the other is from the stereocomplex. Thus, the hydrolytic behaviors of the D-PLA20 and D-PLA80 monolayers were partially similar to those of pure L-PLA and D-PLA monolayers, respectively. This is likely due to preferential hydrolysis of each homopolymer monolayer in the initial stage compared to that of the stereocomplexed monolayers. To further understand the hydrolytic behavior of PLA stereocomplexes at the air/water interface, experiments to investigate the hydrolytic rate of heterogeneous monolayers were performed by separately spreading each solution on the subphase [the D-PLA solution was spread after the L-PLA solution (referred to as Hetero-Mix)], instead of spreading a mixed solution (Homo-Mix, corresponding to D-PLA50). The π-A isotherm of the HeteroMix monolayer showed a plateau region at ca. 8.5 mN/m, which is the same as that of the L-PLA and D-PLA homopolymer monolayers (not shown here). This indicates that the Hetero-Mix monolayers are in an initially phaseseparated status. Figure 7 shows a plot of A/A0 vs time for both Homo-Mix and Hetero-Mix monolayers maintained at 7 mN/m on the subphase of pH 10.5. The hydrolytic behavior of the Hetero-Mix monolayers showed two-step kinetics: the area ratio rapidly decreased during an initial stage (region I) and then the change in the area ratio was similar to that of the Homo-Mix monolayers (region II). The hydrolytic behavior in region I is attributed to a phaseseparated structure (each excess homopolymer phase), whereas in region II, the hydrolytic degradation of the area where L-PLA and D-PLA exist side-by-side, that is, the formation of stereocomplexes, occurs. This is strongly supported by Li and Vert’s result.5 They reported that stereocomplexes were formed during the hydrolysis of D,LPLA, resulting in greater hydrolysis of random unit sequences that did not join the stereocomplexes. Enzymatic Degradation of Monolayers. As mentioned earlier, proteinase K preferentially degrades the D-PLA20 mixture film excess L-PLA (D-PLA)
11184
Langmuir, Vol. 21, No. 24, 2005
Figure 8. Area ratio vs time for mixture monolayer films maintained at 7 mN/m on a subphase with proteinase K content of 0.048 mg. L-LA block as opposed to the D-LA block. On this basis, the effect of the stereocomplexes on the enzymatic hydrolysis of enantiomeric PLA mixture monolayers was investigated. Figure 8 shows a plot of the area ratio, A/A0, vs time for various mixture monolayers with pressure maintained at 7 mN/m on a subphase with dissolved proteinase K. The area ratio of mixture monolayers with g50 mol % D-PLA is nearly 1, regardless of the degradation time. This behavior is not surprising if one considers the preferential degradation of L-PLA because the monolayers of mixtures with >50% D-PLA are composed of stereocomplex and D-PLA phases, which leaves no reactive components. In these compositions, the A/A0 ratio increases with time (over 1). This behavior can be explained by enzyme molecules that are adsorbed or bonded to surface monolayers. This is also consistent with experimental observations in the enzymatic degradation of D,LPLA, where there was very slow enzymatic degradability during an initial degradation. In this experiment, the area ratio decreased as the degradation started. However, the area ratio of mixture monolayers with
