pubs.acs.org/Langmuir © 2009 American Chemical Society
)
Miscibility and Hydrolytic Behavior of Poly(trimethylene carbonate) and Poly(L-lactide) and Their Blends in Monolayers at the Air/Water Interface Hye Kyoung Moon,†, Yong Seok Choi,‡ Jin-Kook Lee,† Chang-Sik Ha,† Won-Ki Lee,§ and Joseph A. Gardella, Jr.*,‡ ‡
† Department of Polymer Science and Engineering, Pusan National University, Busan 609-735, Korea, Department of Chemistry, State University of New York at Buffalo, New York 14260-3000 and §Division of Chemical Engineering, Pukyong National University, Busan 608-739, Korea
Received October 2, 2008. Revised Manuscript Received January 22, 2009 In this study, two biodegradable polymers, poly(trimethylene carbonate) (PTMC) and poly(L-lactide) (PLLA) along with a series of PTMC/PLLA blends, were used as spreading materials to form LB monolayers at the air/ water interface to study hydrolytic reaction kinetics of the monolayers with the Langmuir film balance technique. The π-A isotherms of each homopolymer and their blends showed that blends of PTMC and PLLA were miscible on the neutral subphase (pH 7.4), whereas there was evidence of phase separation on the basic subphase (pH 10.7). The hydrolysis behavior of each homopolymer was investigated at these two different pH conditions. The PTMC monolayer showed faster hydrolysis on the neutral subphase (pH 7.4) than on the basic subphase (pH 10.7). However, in the case of the PLLA monolayer, the hydrolysis on the basic subphase is faster than that on the neutral subphase. On the basis of this result, hydrolysis mechanisms of PTMC and PLLA, considering a general hydrolysis mechanism and their stereo structures, are proposed. The hydrolysis rates of blends of PTMC and PLLA were much faster than that of each homopolymer on the basic subphase (pH 10.7). This result, which can be explained by a “dilution effect”, was supported by the structure based mechanism proposed here.
Introduction Over the last 20 years, various biodegradable polymers have been studied as potential materials for uses such as degradation-controlled implants and controlled drug delivery systems.1-3 Poly(trimethylene carbonate) (PTMC), a linear aliphatic polycarbonate, is one of the biodegradable polymers suitable for these applications.4 PTMC displays good mechanical properties, and significant flexibility due to its low glass transition temperature (Tg).5 While PTMC is degraded by simple hydrolysis and promoted in vivo by enzymatic activity,6,7 its degradation rate is very slow (over several months or years) in aqueous solution.8 Studies of blending or copolymerizing PTMC with other biodegradable polymers have been reported, with the goal to accelerate in vitro hydrolysis rates of PTMC.9-12 However, these previous
)
*Corresponding author. E-mail:
[email protected]; Tel.: +1 716 645 6800 (Ext. 2111). Current address: Department of Chemistry, State University of New York at Buffalo, New York 14260-3000. (1) Chasin, M.; Langer, R. Biodegradable Polymers as Drug Delivery Systems; Marcel Dekker: New York, 1990. (2) Gopferich, A. Biomaterials 1996, 17, 103. (3) Ha, C. S.; Gardella, J. A.Jr. Chem. Rev. 2005, 105, 4205. (4) Scott, G.; Gilead, D. Degradable Polymers; Chapman & Hill: London, 1995. (5) Edlund, U.; Albertsson, A.-C. Adv. Polym. Sci. 2002, 157, 67. (6) Zhu, K. J.; Hendren, R. W.; Jensen, K.; Pitt, C. G. Macromolecules 1991, 24, 1736. (7) Zhang, Z.; Kuijer, R.; Bulstra, S. K.; Grijpma, D. W.; Feijen, J. Biomaterials 2006, 27, 1741. (8) Albertsson, A.-C.; Eklund, M. J. Appl. Polym. Sci. 1995, 57, 87. (9) Albertsson, A.-C.; Eklund, M. J. Appl. Polym. Sci., Part A: Polym. Chem. 1994, 32, 265. (10) Buchholz, B. J. Mater. Sci.: Mater. Med. 1993, 4, 381. (11) Albertsson, A.-C.; Liu, Y. J. Macomol. Sci.: Pure Appl. Chem. A 1997, 34, 1457. (12) Edlund, U.; Albertsson, A.-C. J. Appl. Polym. Sci. 1999, 72, 227.
4478
DOI: 10.1021/la8032435
studies on the degradation properties of PTMC as well as blends with other polymers have been limited to determinations of bulk hydrolysis rates. The Langmuir film balance technique and the transfer of Langmuir-Blodgett (LB) monolayer films are suitable techniques to study polymers in monolayer at the air/water interface. Degradable polymers such as polyesters, in monolayers at the air/water interface, can degrade to water-soluble oligomers or monomers by hydrolysis. These are subsequently dissolved into the water subphase,13,14 resulting in a reduction of the occupied area of the monolayer. Therefore the hydrolysis kinetics of biodegradable polymers can be determined by recording the change of the occupied area. In addition, the surface pressure-area (π-A) isotherm observed by the Langmuir film balance technique can be used to investigate the miscibility of a binary blend in a monolayer by following change in collapse surface pressure or plotting the mean area at a constant surface pressure against the mole fraction of one component.15-20 This is because π-A isotherms of the monolayers strongly depend on the hydrophilic-hydrophobic balance, chemical structure, orientation, and packing of the molecules.15
(13) Ivanova, T.; Panaiotov, J.; Boury, F.; Benoit, J. P.; Verger, R. Colloid Surf., B 1997, 8, 217. (14) Lee, W. K.; Gardella, J. A.Jr. Langmuir 2000, 16, 3401. (15) Kawaguchi, M.; Nishida, R. Langmuir 1990, 6, 492. (16) Gaines, G. L., Jr. Insoluble Monolayers at Liquid-Gas Interfaces; Interscience Publisher: New York, 1966. (17) Gareth, R. Langmuir-Blodgett Films; Plenum Press: New York, 1990. (18) Es-Sounni, A.; Leblanc, R. M. Langmuir 1992, 8, 1578. (19) Lee, W. K.; Nowak, R. W.; Gardella, J. A.Jr. Langmuir 2002, 18, 2309. (20) Thibodeaux, A. F.; Radler, U.; Shashidhar, R.; Duran, R. S. Macromolecules 1994, 27, 784.
Published on Web 2/26/2009
Langmuir 2009, 25(8), 4478–4483
Moon et al.
Article
However, Langmuir film balance studies require the formation of a stable monolayer at the air/water interface, and, for it, an appropriate hydrophilic-hydrophobic balance in the structure of polymer material is required.16 While no Langmuir film studies of PTMC have been reported, it is expected to form a stable monolayer because its hydrophilic and hydrophobic groups are balanced. Therefore, to determine the hydrolysis mechanism and initial degradation kinetics of PTMC, we adapted Langmuir film techniques. In addition, we studied blends of PTMC with poly(L-lactide) (PLLA)14,19 at the air/water interface. In this study, we report the miscibility behavior of blends of PTMC and PLLA, and the hydrolysis kinetics and mechanism for PTMC and its blends with PLLA.
Experimental Section Materials. L-Lactide and stannous octoate were purchased from Aldrich Chemical Co. (Milwaukee, WI). 1,3-Trimethylene carbonate (TMC) was synthesized and purified following a reference method.21 PTMC and PLLA were synthesized by ring-opening polymerization in vacuum-sealed glass ampoules at given temperatures using stannous octoate (0.4 mol %) as a catalyst, respectively. Each homopolymer was purified by being dissolved in methylene chloride (for PTMC) or chloroform (for PLLA), precipitated in cold excess methanol, and filtered. The structures of polymers were confirmed by 1H NMR spectroscopy (CDCl3). The molecular characteristics of both homopolymers are listed in Table 1. NaOH was used to adjust the pH of alkaline subphases, and a 30 mM Tris-HCl buffer solution was prepared for the subphase of pH 7.4. At pH 7.4, the interaction between Tris and monolayer can be ignored because the surface pressure-area isotherms and the degradation behaviors of monolayers in the absence of using Tris have the same result. For preparing the spreading solutions, PTMC/PLLA blends (10/90, 30/70, 50/50, 70/30, 90/10 by mol %) as well as each homopolymer synthesized were dissolved in chloroform with a concentration of 2 μmol/mL. Langmuir Monolayer and Deposition of LB Film. A computer-controlled KSV 2200 film balance held at 20 °C was used for the study of monolayer properties and LB film preparation. The compression rate was kept at 20 cm2/min, with constant barrier movement. The surface pressure was measured with an accuracy of 0.1 mN/m. The purified water from a Millipore Mega-Pure system, MP-6A, was used as subphase liquid. For the study of miscibility and hydrolysis kinetics, 100 μL of the polymer solution was spread onto each subphase of the adjusted pH, and about 5 min were required for the formation of a monolayer on the subphase through the evaporation of chloroform. The transfer experiments for PTMC50 were carried out to observe hydrolysis behavior with a reflection-absorption Fourier transform infrared (RA FT-IR) measurement. After the spread solvent was evaporated and the monolayer was compressed to the given surface pressure, PTMC50 hydrolyzed at pH 7.4 was transferred in multilayers using the Y-type deposition method22 onto a cleaned gold substrate. The substrate face was set as parallel to the moving direction of the barrier, and the dipping speed was 3 mm/min both downward and upward. The RA FT-IR spectrum of a sample consisting of 55 layers transferred to the gold substrate was obtained. During the film deposition, the transfer ratio was not recorded because the occupied area changed as a result of the hydrolysis. Prior to deposition, the gold substrate was cleaned ultrasonically with (21) Zhu, K. J.; Hendren, R. W.; Jensen, K.; Pitt, C. G. Macromolecules 1991, 24, 1736. (22) Petty, M. C. Langmuir-Blodgett Films: An Introduction; Cambridge University Press: Cambridge, U.K., 1996.
Langmuir 2009, 25(8), 4478–4483
Table 1. Characteristics of PTMC and PLLA Used in This Study
PTMC PLLA
Mw
Mw/Mn
Tg (°C)
Tm (°C)
sources
27K 53K
1.53 2.36
-19 51
173
synthesized synthesized
organic solvents and washed carefully with ultrapure water. Each Langmuir film experiment was analyzed in triplicate. All figures from Langmuir experiments are shown for a single run, which is representative of all three replicates. IR Spectroscopy. The FT-IR experiments were performed on a Nicolet Magna 550 FT-IR spectrometer with a resolution of 4 cm-1, purged with nitrogen to ensure a stable and identical data collection environment. The RA FT-IR spectra were acquired with the use of a Harrick Scientific variable-angle RA accessory for analysis at an incident angle of 88°. Good signal-to-noise levels in RA FT-IR spectra acquisition were obtained with 4000 scans. Molecular Structure. Three-dimensional (3D) molecular structures of the PTMC and PLLA at minimal energy state were optimized using Spartan0 04 (Wave Function Inc., Irvine, CA) and Chem 3D Pro 7.0 (CambridgeSoft Co., Cambridge, MA). Each structure was the result of modeling 15 monomer units with a methyl group as the end group.
Results and Discussion Surface Pressure-Area Isotherm Behaviors. Figure 1 shows the pressure-area isotherms for PTMC/PLLA blends (PTMC70 (PTMC:PLLA = 70:30 by mol %), PTMC50, PTMC30) as well as both PTMC and PLLA homopolymers on subphase of pH 7.4. At low surface pressure, the area occupied by monolayer of PTMC was larger than that of PLLA, and, as the content of PTMC in blends increased, their occupied areas also increased. This observation can be explained by the fact that the area per repeating unit of each monomer in the blend was expanded by relatively strong hydrophobic property brought by the trimethylene group of PTMC. In the isotherm from PTMC, the collapse surface pressure and the corresponding area were 6.0 mN/m and 13.1 A˚2/repeating unit, respectively (Table 2). In the isotherm for PLLA, however, the plateau region was observed at the surface pressure of 8.4 mN/m and the range of corresponding area from 11 to 16 A˚2/repeating unit. This plateau region in the isotherms of PLLA can be interpreted as a phase transition and a formation of 3D helical structure.14,23 In the blends that had a higher content of PLLA (g70% by mol), the plateau region was also observed, and it was shortened as the mole content of PLLA decreased. In addition, the isotherms of blends show that their collapse surface pressures change depending on the PTMC/PLLA mole ratios. This dependence on composition in blend can be considered as evidence of the miscibility of the PTMC/PLLA blend in a monolayer at pH 7.4.15-17 Figure 2 shows the π-A isotherms for monolayers of PLLA and PTMC homopolymers and their PTMC/PLLA blends (PTMC90, PTMC70, PTMC50, PTMC30, and PTMC10) on a subphase of pH 10.7. Unlike the PLLA monolayer on pH 7.4 subphase, those on pH 10.7 did not have a plateau region. This means that the monolayer does not form a 3D structure at high pressure because the hydrolysis of the film is accompanied by the increase of its (23) Klass, J. M.; Lennox, R. B.; Brown, G. R.; Bourque, H.; Pezolet, M. Langmuir 2003, 19, 333.
DOI: 10.1021/la8032435
4479
Article
Moon et al.
Figure 1. Surface pressure-area isotherms of PTMC, PLLA, and
Figure 2. Surface pressure-area isotherms of PTMC, PLLA, and
their blends (PTMC70 (PTMC:PLLA = 70:30 by mol), PTMC50, and PTMC30) in monolayers on the subphase of pH 7.4.
their blends (PTMC90, PTMC70, PTMC50, PTMC30, and PTMC10) in monolayers on the subphase of pH 10.7.
Table 2. Inflection Pointsa on the Surface Pressure-Area Isotherms of PTMC, PLLA, and Their Blends (PTMC70, PTMC50, and PTMC30) in Monolayers at pH 7.4 surface pressure (mN/m) PTMC PTMC 70 PTMC 50
area (A˚2/repeating unit)
6.0 4.4 5.0 12.5 PTMC 30 4.4 9.4 PLLA 8.4 a The values in this table have an accuracy of ( 0.1.
13.1 18.7 15.3 6.7 16.9 11.1 16.3
hydrophilicity.14 The shift in the surface pressures of collapse points in blends from those in PTMC and PLLA homopolymers, observed on the subphase of pH 7.4, was not detected at pH 10.7. This can be explained by the interaction between ester groups in PLLA or carbonate groups in PTMC and excess sodium and hydroxide ions at the air/ water interface, if those ions exist at the interface.13,14 As a result, the monolayer seems to form an immiscible mixture in the layer due to the isotherm following the tendency of those from each homopolymer. In addition, the observation that the collapse point of each homopolymer in Figure 2 appeared at a higher position than that in Figure 1 can be supported by the same reasoning. The miscibility behavior in a binary polymer system can be predicted from the pressure-area isotherms at the air/water interface.15-20 Figure 3 shows the measured area per repeating unit versus the mole fraction of PTMC in the mixture of PTMC and PLLA under different conditions of pH and surface pressure, less than the collapse surface pressure of monolayers. The mean area per repeating unit of the blend (Ablend), shown as a line in Figure 3, is given by Ablend ¼ M PTMC APTMC þ M PLLA APLLA
ð1Þ
where M is the mole fraction, and A is the occupied areas of components of PTMC and PLLA at the particular surface 4480
DOI: 10.1021/la8032435
pressure, respectively.15 According to the additivity rule, a negative deviation (the subtraction of Ablend from the measured area of blend) is expected for a miscible binary blend, while positive and zero deviations usually indicate phase separation.15-20 In Figure 3a, negative deviation was observed on the subphase of pH 7.4. This means that PTMC and PLLA homopolymers are miscible at a surface pressure of 3 mN/m, less than the collapse surface pressure of monolayer. However, Figure 3b generally shows a positive deviation at surface pressures of 3 and 5 mN/m on a subphase of pH 10.7. This can be explained as the blends of PTMC and PLLA were separated into two different phases as a result of the interruption of attractive interactions between PLLA and PTMC, induced by extra sodium and hydroxide ions existing at the air/water interface. However, deviation from the additivity of the surface areas is, in itself, not sufficient evidence to demonstrate the complete incompatibility or the ideal miscibility of the two components.15 The miscibility should be also determined by a plot of the collapse surface pressure as a function of mole fraction of one component in the binary mixture.15-17 The collapse surface pressure of the miscible monolayer depends on the composition, whereas, for the completely immiscible monolayer, it is independent of the components.15 While the collapse surface pressure depends on PTMC/PLLA ratios at pH 7.4, in Figure 2 the collapse surface pressure at pH 10.7 is generally independent of the composition, and this supports the hypothesis of the immiscbility of PLLA and PTMC in blends at pH 10.7. Hydrolysis of Monolayers of PLLA/PTMC Homopolymers. The Langmuir film balance technique was used in previous studies of hydrolytic kinetics of polyesters.13,14,24 This method results in a quantitative analysis of the degradation rate at the molecular level of the monolayer. Thus, the study of the hydrolysis behavior of monolayers of PTMC as well as PLLA can give an understanding of the hydrolysis (24) O’Brien, K. C.; Lando, J. B. Langmuir 1985, 1, 533.
Langmuir 2009, 25(8), 4478–4483
Moon et al.
Article
Figure 3. Measured (spots) and mean area (lines) per repeating unit vs the mole fraction of PTMC in PTMC/PLLA blends on a subphase of pH 7.4 at a surface pressure of 3mN/m (a) and on a subphase of pH 10.7 at surface pressures of 3 and 5mN/m (b). mechanisms of polycarbonate and polyester structures, and that of their blends. Figure 4 shows the plot of area ratio vs time for PLLA and PTMC homopolymers in monolayers on subphases of pH 7.4 and pH 10.7. In this figure, we can compare the different hydrolysis tendencies depending on the pH of the subphase. At pH 7.4, the PTMC homopolymer was degraded more dramatically than pH 10.7 (Figure 4a), while the PLLA homopolymer was more easily hydrolyzed at pH 10.7 than pH 7.4 (Figure 4b). This can be explained if we consider each stereo structure of these polymers and general hydrolysis mechanism.25 In the case of PTMC, if carboxyl oxygens are exposed externally (Figure 5a), they could donate electrons easily to nearby protons and have a partial positive charge. Consequently, hindered sites of nucleophilic attacks might be open by the rotation of the chains (Figure 5b), and then, nucleophiles (water molecules) could attack the carboxyl carbons. This explanation can be supported by the 3D molecular modeling result, which showed wider angles (at least 5°) of the PTMC backbone surrounding nucleophilic attack sites in the presence of protons than those without protons (Figure 5a,b). However, PLLA is proposed to have a different structural situation. As the opposite side of the carboxyl oxygen (the site of nucleophilic attack) of PLLA is relatively highly exposed to the outside, compared to PTMC as shown in Figure 5, a larger portion of the hydrolysis process would easily depend on the attack of nucleophiles (hydroxyl ions) on the opposite side of the carboxyl oxygen compared (25) Solomons, T. W. Organic Chemistry; John Wiley & Sons: New York, 1976.
Langmuir 2009, 25(8), 4478–4483
Figure 4. Area ratio vs time for PLLA (a) and PTMC (b) in monolayers on subphases of pH 7.4 and 10.7 at a surface pressure of 2mN/m. to the donation of carboxyl oxygen electrons to protons (Figure 5c). Hydrolysis of Monolayers of PTMC/PLLA Blends. As was the case for the homopolymers, the blends showed hydrolysis at neutral and basic subphase conditions. Figure 6a shows the hydrolysis kinetic run of PTMC50 (PTMC:PLLA = 50:50 by mol %) blend at pH 7.4. This hydrolysis behavior of the blend can be confirmed by its RA FT-IR measurement. Figure 6b(1) is a RA FT-IR spectrum of the PTMC50 film prepared by the solution cast method on gold substrate, and Figure 6b(2) was obtained from multiple layers of the hydrolyzed PTMC50 deposited on the gold substrate by using the LB film transfer technique. In the spectra, the carbonyl stretching mode from both PLLA and PTMC is observed as the strongest peak at 1764 cm-1. In addition, the -COC- stretching from PTMC is observed at 1294 cm-1 and the -COC- asymmetric stretching from PLLA appeared at 1098 and 1135 cm-1 in RA FT-IR spectrum.23 We can compare the relative absorbance of these peaks to one another to determine if the blend is degraded or not. First, we compared the absorbance of each carbonyl peak, which had approximately equivalent absorbance in both spectra. This peak was not changed even after the degradation because carbonyl functionality is not directly affected DOI: 10.1021/la8032435
4481
Article
Moon et al.
Figure 5. Parts of 3D structures for PTMC (a), [PTMC+nH]n+ (b), and PLLA (c). Gray balls are carbon, spotted balls are oxygen, and white balls are hydrogen.
by hydrolysis. Thus, the reduction of absorbance at other peaks from the solution cast film to the multilayer can be explained by the hydrolysis of the blend. Most of all, the large reduction of absorbance was observed from -COCstretching bands of PTMC at 1294 cm-1 and PLLA at 1098 and 1135 cm-1 because the ether bond is easily broken by hydrolysis. Therefore, we conclude that the hydrolysis behavior of PTMC50 observed by Langmuir film balance technique can be verified by the comparison of RA FT-IR spectra for the films before and after hydrolysis treatment. Figure 7 shows the area ratio versus time for PLLA, PTMC, and their blended (PTMC70, PTMC50, and PTMC30) monolayer films at 3mN/m on the subphase of pH 10.7. The area ratio of PTMC monolayer at the air/water interface decreased very slowly, whereas that of the PLLA monolayer decreased steeply compared to PTMC. In Figure 7, the PTMC/PLLA blends showed a steeper decrease in the area ratio than the arithmetic averages of the area ratio for the blends. This can be expected from the degradation kinetics of each homopolymer within the time scale studied here. As already noted, PTMC requires protons to be hydrolyzed owing to its steric structure as the initial step of hydrolysis. However, in this basic condition, protons are not 4482
DOI: 10.1021/la8032435
Figure 6. Area ratio vs time for monolayers of PTMC, PLLA homopolymers, and PTMC50 blend (PTMC:PLLA = 50:50 by mol) at 2mN/m on subphases of pH 7.4 (a), and RA-FT-IR spectra of PTMC50 blend (b) in the solution cast film (1) and the multilayers deposited on the gold substrate during the hydrolysis at pH 7.4 (2). readily available, and, additionally, the site of nucleophilic attack is hindered relative to that of PLLA. Therefore, the attacks of hydroxyl ions mainly concentrate on PLLA in blends, and this contributes to its drastic decrease in the area ratio compared to their arithmetic averages. This explanation is similar to the dilution effect,14,19 a previous Langmuir 2009, 25(8), 4478–4483
Moon et al.
Article
Conclusions
Figure 7. PLLA and PTMC homopolymers and their blends (PTMC70, PTMC50, and PTMC30) on subphases of pH 10.7 at a surface pressure of 3 mN/m. explanation about hydrolysis kinetics of polyester blend. In conclusion, the dilution of the PLLA by PTMC, which degrades slowly at pH 10.7 due to its steric structure, actually increases the rate of hydrolysis by making PLLA more accessible to hydrolytic attack.
Langmuir 2009, 25(8), 4478–4483
We studied the miscibility behavior and the effects on hydrolysis kinetics of PTMC and PLLA in blends. The miscibility of PTMC and PLLA at the air/water interface depended on the subphase of pH. We found PTMC and PLLA in their blends were miscible at neutral condition (pH 7.4), whereas they had the phase separation on basic condition (pH 10.7). This phase separation behavior at pH 10.7 was related to the interruption of the attractive interaction between PTMC and PLLA, brought by sodium and hydroxide ions. The hydrolytic kinetics of PTMC and PLLA homopolymers were revealed from the occupied area changes of monolayers on the subphases. While PTMC was hydrolyzed rapidly at pH 7.4, PLLA was hydrolyzed rapidly at pH 10.7. On the basis of these results, we proposed a mechanism for hydrolytic degradation of PTMC and PLLA considering their proposed structures. The rapid hydrolysis of blends at pH 10.7, compared to their arithmetic averages, was also explained by using the hydrolytic mechanism of each homopolymer, and this supported the dilution effect. Acknowledgment. This research was supported by the Korea Science & Engineering Foundation (KOSEF) to H.K.M. and the US National Science Foundation Grants to J.A.G. (CHE-0316735 and CHE-0616916).
DOI: 10.1021/la8032435
4483
