Characterization of the Spontaneously Forming Hydrogels Composed

time was required for the hydrogel to undergo chain rearrangement. DSC thermograms of the hydrogels were different from their components, PMA and PMB...
0 downloads 0 Views 122KB Size
Biomacromolecules 2002, 3, 100-105

100

Characterization of the Spontaneously Forming Hydrogels Composed of Water-Soluble Phospholipid Polymers Kwang Woo Nam, Junji Watanabe, and Kazuhiko Ishihara* Department of Materials Science, School of Engineering, The University of Tokyo, 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-8656, Japan Received July 27, 2001; Revised Manuscript Received October 17, 2001

Spontaneously forming hydrogels composed of 2-methacryloyloxyethyl phosphorylcholine (MPC) copolymers, poly(MPC-co-methacrylic acid) (PMA), and poly(MPC-co-n-butyl methacrylate) (PMB) were examined. The MPC copolymer hydrogel was observed to have a spontaneous gelation property. To determine the properties of the hydrogels and why the gelation takes place, we have studied the properties of the hydrogels by scanning electron microscopy, X-ray photoelectron spectroscopy (XPS), and differential scanning calorimetry (DSC). The morphologies of the hydrogels were spongelike with a homogeneous structure. By XPS analysis in terms of the molecular distributions in the hydrogels, it was observed that a stabilization time was required for the hydrogel to undergo chain rearrangement. DSC thermograms of the hydrogels were different from their components, PMA and PMB. For the hydrogel, a crystallization peak around -30 °C was observed. This result indicated that some ordered structures existed in the hydrogels. To determine the role of the MPC groups, aqueous solutions of poly(methacrylic acid) (PMAc) and PMB were mixed. The mixture of PMAc-PMB turned into a sol state, and the sol state remained for a week. When the mixture was cooled, a very weak hydrogel was prepared. This result suggested that the MPC groups were the dominant unit for spontaneously forming the hydrogels. Introduction Physical gels are continuous, disordered, three-dimensional networks formed by the associative forces capable of forming noncovalent cross-links.1 Mainly, the cross-link points are formed by much weaker and reversible forms of chain-chain interactions. These interactions mainly consist of hydrophobic interactions, ionic bondings, hydrogen bondings, crystalline segements, stereocomplexation, and solvent complexation.2 These actually form junction zones instead of cross-linking points, holding the amorphous region of the polymer chain together. Since these junction zones are not covalently bonded, they show reversible behavior and can easily break by changing the environment. We have already found anomalous hydrogels composed of poly(MPC-co-methacrylic acid) (PMA) and poly(MPCco-n-butyl methacrylate) (PMB).3 The hydrogels showed spontaneous gelation by mixing them at room temperature. The PMA and PMB aqueous solution did not require any cross-linker or organic solvents for the gelation. Outside stimuli, for example, UV light or even a temperature change, were not needed for these two substances to form a hydrogel. They exist in aqueous media and show hydrophilic properties, indicating that it may be used as a biomaterial. The MPC polymer not only shows excellent blood compatibility but also exhibits diversity as a material by copolymerization of other materials.4,5 Most favorable applications of the MPC copolymers were materials for artificial * To whom all correspondence should be addressed. Email: ishihara@ bmw.t.u-tokyo.ac.jp (K. Ishihara). Telephone: +81-3-5841-7124. FAX: +81-3-5841-8647.

organs,6,7 a lubricant for hemiarthroplasty,8 and cosmetics.9 Recent studies have focused on the series of MPC polymers that showed the MPC copolymer to have excellent protein nonadsorption properties,10 for use as dialysis membranes11,12 and sensors.13,14 Most recently, the MPC copolymer was investigated in terms of nanoparticles for drug carriers.15 In this study, we have prepared the MPC copolymer-based hydrogel using PMA and PMB 5 wt % aqueous solutions and investigated their gelation mechanism. The gelation mechanism was discussed in terms of the water structure and diffusivity of the MPC copolymer. Experimental Section Preparation of the Hydrogels. Two kinds of watersoluble MPC polymers, PMA (MPC mole fraction 0.3, Mn ) 6.0 × 104, Mw ) 2.9 × 105) and PMB (MPC mole fraction 0.8, Mn ) 1.2 × 104, Mw ) 1.0 × 106), which were prepared by the previously reported radical polymerization method,16,17 were kindly supplied by the NOF Corporation (Tokyo, Japan) and were used for the preparation of the hydrogel (Figure 1). The PMA and PMB aqueous solutions (5 wt %) (volume ratio in feed ) 1:1) were mixed and then vigorously mixed using a vortex stirrer for 10-20 s (A1B1). The feeding ratio was changed for preparing the hydrogel PMA/PMB ) 1/2(A1B2) and 2/1(A2B1). Poly(methacrylic acid) (PMAc) was synthesized as a reference sample to confirm the gelation mechanism. The synthesis method was done using the conventional radical polymerization method using R,R′-azobisisobutyronitrile

10.1021/bm015589o CCC: $22.00 © 2002 American Chemical Society Published on Web 11/29/2001

Characterization of Spontaneously Forming Hydrogels

Biomacromolecules, Vol. 3, No. 1, 2002 101

Figure 1. Chemical structures of PMA (a) and PMB (b).

Figure 2. SEM images of the copolymers and the hydrogels (a) PMA, (b) PMB, (c) A1B1, (d) A1B2, and (e) A2B1.

(AIBN) in a glass ampule. PMAc and PMB aqueous solutions (5 wt %) were also mixed to prepare a reference hydrogel. Surface Characterization of the Hydrogels. Scanning electron microscopy (SEM) (JSM-5400, JEOL, Tokyo, Japan) and X-ray photoelectron spectroscopy (XPS) (ESCA850, Shimadzu, Kyoto, Japan) were used as surface characterization techniques. The hydrogels were lyophilized and then broken into pieces. For the SEM observations, the samples were kept in a vacuum and removed just before the gold sputtering treatment. For the XPS measurements, the lyophilized hydrogels were also prepared. The carbon, oxygen, nitrogen, and phosphorus were analyzed at a depth of 50 Å. The results were used to compare the element ratios between the freshly made and 1-week-stored hydrogels. Thermal Property of the Hydrogels. The hydrogels were lyophilized and characterized using a differential scanning calorimeter (DSC) (DSC6000, Seiko, Chiba, Japan) in the range of -100 to 200 °C at the scanning rate of 3 °C/min. The melting temperatures of the PMA, PMB, A1B1, A1B2, and A2B1 were measured. To define the interaction between the water and the chains of the hydrogels, the hydrogels that contain water were also investigated. First, the total amount of free water was calculated using the endothermic peak at 0 °C. To measure the amount of free water inside the hydrogel, water was then added to each lyophilized gel from a 10% to 95% weight fraction. The gels were left at room temperature to stabilize before the measurement. Small portions were taken out to measure by DSC. The measurement was repeated five times for each sample, and the method to yield the amount of free water was the same as that used for pure water. Results and Discussion Preparation of the Hydrogels. The hydrogels were spontaneously formed when the two MPC polymer solutions containing PMA and PMB were put together and mixed for 20 s. For A1B1 and A1B2, the gelation occurred in less than

10 s. The shape and the size were controllable depending on the container. They were translucent right after the gelation, but they required at least 1 day to show full transparency. The water content was changed from 20% to 95%. This result indicated that the mechanical strength was changed by the composition of the copolymers in the hydrogels. A2B1 showed the weakest mechanical strength, while A1B2 showed highest mechanical strength. The formation of the hydrogels was reversible with temperature. The hydrogel turned liquid when it was heated to 60 °C and returned to the hydrogel state again when it was cooled. The gelation occurred only under aqueous conditions. The hydrogels dissolved in alcohol or any other polar solvent. The hydrogels were completely dissolved by the addition of urea even they were in an aqueous medium at room temperature. This means that hydrogen bonding is an important factor to maintain the gel state. Morphology of the Hydrogel. The morphology of the hydrogels was observed using SEM. The hydrogels showed spongelike structures after lyophilization, as shown in Figure 2. The average pores size (≈7 µm) was almost the same among the samples, but A2B1 had the smallest pore size and a very sparse structure. A spongelike structure could be seen only for PMB, while PMA did not have any pores at all. The existence of the spongelike structure in the hydrogels was a consequence of the complete blending between PMA and PMB. This result indicated that these two components (PMA and PMB) were homogeneous. This does not happen to all the MPC-based polymers, for many do not show these spongelike structures (data not shown). Characterization of the Hydrogels with XPS. At first, it was thought that the phospholipid polar groups would be mainly located on the surface of the hydrogel and the butyl methacrylate and methacrylic acid would be mainly located inside. Furthermore, the difference between the translucency right after mixing and the transparency after 1 day had to be clarified. To define the elemental distributions on the hydrogel surfaces, XPS was used. As shown in Table 1, the elements were evenly distributed after 1 week. The phospholipid groups were not concentrated at certain points,

102

Biomacromolecules, Vol. 3, No. 1, 2002

Nam et al.

Table 1. XPS Results of the Hydrogel Taken after 30 s and 1 Week

C 1s O 1s N 1s P 2p N/C P/C

experimental results

theoretical values

a weeka

fresh gel (a)b

fresh gel (b)b

14.2 7.6 1.0 1.0 0.1 0.1

16.9 6.6 1.0 1.2 0.1 0.1

15.6 2.0 1.0 1.1 0.1 0.1

16.0 6.2 1.0 1.2 0.1 0.1

a Average. b A small portion was taken out from different parts of the same freshly made hydrogel.

indicating that PMA and PMB would be totally entangled inside. Also, the hydrogels that had been freshly made and stored for 1 week showed a difference in the elemental distribution. The 1-week-stored hydrogel showed almost the same distribution compared with the theoretical data, while the freshly made hydrogel showed a different distribution of oxygen atoms. The elemental distribution in the hydrogel was changed according to the chain rearrangement, suggesting that it takes some time for the chains to be homogeneously and completely entangled. However, the chain rearrangement would not include the total displacement of the whole backbone but just rearrangement of the side chains. Thermal Properties of the Hydrogels. The thermal properties of the hydrogels were observed by the DSC measurement. As shown in Figure 3a, the melting temperature of the polymer gel in the dry state was 115 °C. A2B1 and A1B1 showed almost the same melting temperature. However, A2B1 showed a much lower melting temperature that was slightly higher than PMA. A crystalline peak was observed at about -30 °C. This crystalline peak was observed among all the dried samples; however, for PMA, PMB, and the PMA-PMB mixture, the crystalline peaks around -30 °C were not observed (Figure 3b). Specifically, the hydrogel showed three different peaks, indicating that there are three different factors that are causing the crystallization. However, what actually is causing this is not yet clear and will be discussed in our next article. There are two exothermic peaks observed around 0 °C in this case. These peaks are thought to be the water that had been adsorbed from the atmosphere. Actually, the MPC copolymers show very high humidity sensitivity. However, the hydrogels did not show any peaks in this temperature range,

meaning that the water sensitivity would be decreased. It can be said that the ordered structure caused by the crystallization of the BMA unit was thought to be impossible, for the BMA unit has side chains that are too short to form the crystalline structure. However, it is possible to suggest that the PMB forms its own hydrophobic region, enhanced by the addition of water.18 Most of the DSC thermograms in the MPC polymer hydrogel have a similar tendency as shown in Figure 4. For the hydrated samples, the crystalline peak disappeared and an endothermic peak near 0 °C appeared. The rapid endotherm took place at about 40 °C, and the lowest melting point was seen at about 100 °C. The DSC thermograms were different between the freshly made hydrogel and the other hydrogels. The freshly made hydrogel showed the middle of the fully hydrated and dehydrated state. This result indicated that the hydrogel was fully stabilized within an hour. Transparency, which can be observed only after full stabilization, was too low, and the mechanical properties did not reach the maximum point, meaning that after the chain rearrangement, the surrounding environment would affect the properties of the hydrogel. When the water content changed, as seen in Figure 5, the melting peaks significantly varied. When the water content is below 50%, the melting peak would be over 100 °C. However, when the water content is over 60%, the melting peak suddenly shifts to a lower temperature indicating that the interaction between the water molecules would be changed. Therefore, effect of the water molecules on the polymer chains would be small until a 50% water content, but over a 60% water content, the polymer chains would be affected a lot by the water molecules. Furthermore, the amount of water inside the hydrogel may be altering the hydrogel properties. The interaction between the carboxylic groups and the water seems to be independent of the hydrogel formation. According to Eisenberg et al.,19 the water could be divided into three parts in the unfrozen state: free water, the disordered water, and the unfrozen water. However, we just divided the water into two large parts, the free and bound water, since it was difficult to distinguish the disordered water and unfrozen water by DSC measurements. When the water content is low inside the hydrogel, the bound water and the free water inside the hydrogel do not have a big difference between the different types of hydrogels indicating

Figure 3. DSC graphs of the copolymers and the hydrogels in the dry state: (a) melting peak; (b) crystalline peak.

Characterization of Spontaneously Forming Hydrogels

Biomacromolecules, Vol. 3, No. 1, 2002 103

Figure 4. DSC graphs of the A1B1 hydrogel versus time: (a) 95 wt % water; (b) 70 wt % water.

Figure 7. Free water percentage of the respective hydrogel and the water percentage. Figure 5. DSC graphs of the A1B1 based on the water content.

Figure 6. Enthalpy change of the hydrogels according to the water percentage within the hydrogels.

that the water molecules would not be in the same state as before (Figure 6). However, this does not directly indicate that the carboxylic groups would interact with the water molecules. It would rather react with the other carboxylic groups via water. If there was no water or very little water, they would not be reacting with each other to form a hydrogel. This is because the carboxylic groups would exist in the ionic form, and when they encounter the PMB water solution, they form hydrogen bonding between the carboxylic group due to the low permittivity and polarity. If there were over 80% water in the PMA-PMB blend, about 80% of the total water would exist as free water. As shown in Figure 7, the bound water rapidly decreases when the water content in the hydrogel is over 80%. Also in 90%,

the free water takes over 85% of the total water. Also, when the water is 95%, the free water takes over 93% of the total water. The lowest free water was seen for A2B1. It was considered that the relative ratio between the carboxylic groups in PMA and butyl group in PMB regulated the water molecules. The hydrophobic region formed by the BMA units and the polymer main chains has a low permittivity and low polarity. This brings the carboxylate anions nearby to form complete neutral carboxylic acid groups, which may induce hydrogen bonding between the carboxylic acid groups. These are located where the hydrophobic region is less effective, and carboxylate anions would react with water making the bound water region. The large amount of the carboxylate anions may lead to a reaction between the carboxylate anions and the water molecules. This would eventually induce the weak mechanical strength of the hydrogel and the sparse morphology. The Effect of the Phospholipid Polar Groups on Gel Formation. The role of the phospholipid polar groups was not clarified at this point. The polar groups in the MPC unit of the polymer are assumed to be due a function such as interacting with each other using water as the medium. Berkowitz et al.20-22 and Rand et al.23,24 proposed a model for the interaction of the phospholipid polar groups stably existing in water. However, we decided to rule out this possibility, for the MPC hydrogel was not layered but entangles in the polymer chains. As we described earlier, the XPS result showed that the phospholipid polar groups were not located in specific parts or on the surface of the hydrogels.

104

Biomacromolecules, Vol. 3, No. 1, 2002

Nam et al.

Figure 8. Surface morphology of the PMAc-PMB blend: (a) the pure poly(methacrylic acid) homopolymer surface; (b) the surface of the PMAc-PMB 30 s after mixing; (c) the surface of the PMAc-PMB blend gel after 2 weeks.

However, Zhang et al.25 mentioned that in PMB, the phospholipid polar groups would be facing out, while chains extend in. Therefore, it can be assumed that the phospholipid polar groups of different copolymers are facing each other once they are mixed. To determine how important the MPC unit in the hydrogel is, PMAc and PMB were mixed under the same conditions used for the PMA-PMB hydrogel preparation. The mixture of PMAc-PMB turned into a sol state. The transparency rapidly decreased, and aggregates could be seen. The sol state remained for a week, but turned into the hydrogel with a very weak mechanical strength under the lower temperature conditions. Lowering the temperature promotes formation of the hydrogen bonding and would lower the mobility of the PMAc chain and eventually form a weak hydrogel. In short, the PMAc-PMB water mixture is undergoing a sol-gel transition. When the surfaces of the samples were observed by SEM, PMAc showed a very different surface morphology from that of PMA. Unlike the uniform and nonporous surface of PMA, PMAc had the network structure shown in Figure 8. This network structure was certainly different from the sponge structure. Therefore, we could easily detect that the phospholipid polar groups are changing the morphology of the PMAc. When the PMAc was mixed with PMB, the gelation does not take place, as previously mentioned. The morphology of the sample 30 s after the mixing showed that there are two different states existing, as shown in Figure 8b. It showed that some had been mixed and formed the porous structure, but some parts showed no mixing at all. This is because the PMAc chains and PMB chains are not compatible. However, after 2 weeks in the refrigerator (4 °C), a very uniform, and homo-PMA-like morphology was observed. Figure 8c shows the SEM images after 2 weeks. The sample showed that the surface was completely mixed and formed a porous structure resembling the PMA-PMB hydrogel. Therefore, it could be said that the MPC polymer in the copolymers are bringing the polar PMAc and nonpolar BMA together by increasing their miscibility which is lowering the diffusivity of the PMAc chains. The hydrogel is thought to be formed by the following procedure. First, the PMB would be aggregated in the aqueous conditions. It forms a region where a hydrophobic interaction is available. These kinds of regions would be formed where the PMB moieties exist throughout the hydrogel. Here, the polarity and the permittivity would be much less compared to the other nonhydrophobic regions. The carboxylic groups that exist as carboxylic anions in PMA

Chart 1. A Schematic View of How Gelation Occurs: PMB Aqueous Solution Forms Hydrophobic Regions (a) and PMA Chains in Aqueous Solution Chains Would Be Full of Carboxylic Anions Groups (b), When They Are Mixed Together, Carboxylic Anions near the Hydrophobic Regions Become Carboxylic Groups and Form Strong Hydrogen Bondings

that are located near the n-butyl methacrylate would be deionized as the polarity and permittivity decrease when these two components are blended in water. However, one problem would be that the carboxylic groups and n-butyl methacrylate groups would not have much compatibility. The role of the phospholipid polar group is to enhance the miscibility of these two completely different compounds. Second, the phospholipid polar groups would lower the diffusivity of the PMAc. This would increase the cluster of water inside the polymer, so as to render part of the water comparatively immobile.26 In this way, they would form a hydrogel that has a relatively low mechanical strength but still has enough strength to hold the water inside as shown in Chart 1. The hydration force was not considered here. Once they are formed, the hydrogels will undergo a chain rearrangement process. In this way, they form a physically cross-linked hydrogel that contains 95% water. Conclusion Spontaneously forming hydrogels were investigated in terms of the gelation factors and hydrogel properties. The hydrogel was prepared by mixing PMA and PMB aqueous solutions at room temperature. The time to form a hydrogel was suppressed by the following conditions: (i) heating the PMA and PMB mixtures, (ii) using the PMAc and PMB mixtures. These results indicated that the hydrogen bonding in the PMA polymer composed of MPC and MA needed to form a hydrogel. Furthermore, the properties of the hydrogel were examined by SEM observations, and the hydrogel was found to be homogeneous and to have a spongelike structure. On the basis of the XPS and DSC observations, a chain rearrangement occurred in the hydrogel. Taking these results into account, the PMA and PMB aqueous solution rapidly turned into the gel state by forming hydrogen bonds. These findings are of great importance for designing new soft biomaterials.

Characterization of Spontaneously Forming Hydrogels

Acknowledgment. We are grateful to Dr. Yasuhiko Iwasaki of Tokyo Medical and Dental University and Dr. Ken Suzuki of the NOF Corporation for their helpful discussions and SEM measurements. References and Notes (1) Park, K.; Shalaby, W. S. W.; Park, H. Biodegradable Hydrogels for Drug DeliVery; Technomic Publishing Co.: Lancaster, PA, 1993. (2) Tanaka, F.; Edwards, S. F. Macromolecules 1992, 25, 1516. (3) Ishihara, K.; Watanabe, J.; Sakamoto, N.; Shuto, K.; Suzuki, K. Trans. Soc. Biomater. 2001, 24, 367. (4) Ishihara, K.; Aragaki, R.; Ueda, T.; Watanabe, A.; Nakabayashi, N. J. Biomed. Mater. Res. 1990, 24, 1069. (5) Ishihara, K.; Ziats, N. P.; Tierney, B. P.; Nakabayashi, N.; Anderson, J. M. J. Biomed. Mater. Res. 1991, 25, 1397. (6) Ishihara, K.; Tsuji, T.; Kurosaki, T.; Nakabayashi, N. J. Biomed. Mater. Res. 1994, 28, 225. (7) Yoneyama, T.; Ishihara, K.; Nakabayashi, N.; Ito, M.; Mishima, Y. J. Biomed. Mater. Res. 1998, 43, 15. (8) Foy, J. R.; William, P. F., III; Powell, G. L.; Ishihara, K.; Nakabayashi, N.; LaBerge, M. Proc. Inst. Mech. Eng. 1999, 213, 5. (9) Shaku, M.; Kuroda, H.; Oba, A.; Ohkura, S.; Ishihara, K.; Nakabayashi, N. Cosmet. Toiletries Mag. 1997, 112, 65. (10) Ishihara, K.; Nomura, H.; Mihara, T.; Kurita, K.; Iwasaki, Y.; Nakabayashi, N. J. Biomed. Mater. Res. 1998, 39, 323. (11) Ishihara, K.; Shinozuka, T.; Hanazaki, Y.; Iwasaki, Y.; Nakabayashi, N. J. Biomater. Sci., Polym. Ed. 1999, 10 (3), 271.

Biomacromolecules, Vol. 3, No. 1, 2002 105 (12) Hasegawa, T.; Iwasaki, Y.; Ishihara, K. Biomaterials 2001, 22, 243. (13) Oki, A.; Adachi, S.; Takamura, Y.; Ishihara, K.; Ogawa, H.; Ogawa, Y.; Ichiki, T.; Horiike, Y. Electrophoresis 2001, 22, 341. (14) Nowak, T.; Nishida, K.; Shimoda, S.; Konno, Y.; Ichinose, K.; Sakakida, M.; Shichiri, M.; Nakabayashi, N.; Ishihara, K. J. Artif. Organs 2000, 3, 39. (15) Konno, T.; Kurita, K.; Iwasaki, Y.; Nakabayashi, N.; Ishihara, K. Biomaterials 2001, 22 (13), 1883. (16) Ishihara, K.; Ueda, T.; Nakabayashi, N. Polym. J. 1990, 22, 355. (17) Ueda, T.; Oshida, H.; Kurita, K.; Ishihara, K.; Nakabayashi, N. Polym. J. 1992, 24, 1259. (18) Matsuda, A.; Gong, J. P.; Osada, Y. Polym. Gels Network 1998, 6, 307. (19) Eisenberg, D.; Kauzmann, W. The Structure of and Properties of Water; Oxford University Press: 1969. (20) Marrink, S. J.; Berkowitz, M.; Berendsen, H. J. C. Langmuir 1993, 9, 3122. (21) Essmann, U.; Perera, L.; Berkowitz, M. Langmuir 1995, 11, 4519. (22) Perera, L.; Essmann, U.; Berkowitz, M. Langmuir 1996, 12, 2625. (23) Lis, L. J.; McAlister, M.; Fuller, N.; Rand, R. P.; Parsegian, V. A. Biophys. J. 1982, 37, 657. (24) Rand, R. P.; Parsegian, V. A. Biochim. Biophys. Acta 1998, 988, 351. (25) Zhang, S. F.; Rolfe, P.; Wright, G.; Lian, W.; Milling, A. J.; Tanaka, S.; Ishihara, K. Biomaterials 1998, 19, 691. (26) Van Krevelen, D. W. Properties of Polymers; Elsevier Science: New York, 1990; Vol. 3.

BM015589O