10348
Langmuir 2007, 23, 10348-10352
Double Plasma Treatment-Induced Graft Polymerization of Carbohydrated Monomers on Poly(ethylene terephthalate) Fibers Loı¨c Bech,† Be´ne´dicte Lepoittevin,† Abdelhakim El Achhab,† Emmanuel Lepleux,‡ Lionel Teule´-Gay,§ Caroline Boisse-Laporte,*,§ and Philippe Roger*,† Laboratoire de Chimie Organique Multifonctionnelle, Baˆ timent 420, Equipe Glycochimie Mole´ culaire et Macromole´ culaire, UMR 8182 CNRS, Institut de Chimie Mole´ culaire et des Mate´ riaux d’Orsay, UniVersite´ Paris-sud 11, 91405 Orsay Cedex, France, ScienTec, ZI de Courtaboeuf, 17 aVenue des Andes, Baˆ timent le Ce` dre, 91952 Les Ulis, France, and Laboratoire de Physique des Gaz et des Plasmas, UMR 8578, Baˆ timent 210, UniVersite´ Paris-sud 11, 91405 Orsay Cedex, France ReceiVed May 15, 2007. In Final Form: July 3, 2007 This study deals with the grafting of carbohydrate monomers on poly(ethylene terephthalate) fibers by double argon plasma treatment. Two monomers were used: allyl R-D-galactopyranoside and 2-methacryloxyethyl glucoside. The quantity of grafted carbohydrates was determined by phenol/sulfuric acid colorimetric titration. The graft density was observed to vary according to the monomer used. Allyl R-D-galactopyranoside yields to smaller graft densities compared to 2-methacryloxyethyl glucoside, suggesting transfer reactions occurring at the surface with allyl R-D-galactopyranoside. Fibers with the highest graft levels were obtained with the higher monomer concentration and the lower quantity of fiber treated in a plasma reactor. The grafting density can be modulated by the monomer concentration and mass of fiber exposed in the plasma reactor. For 0.5 mg of fibers, the graft densities for 23 and 68 mM allyl R-D-galactopyranoside are, respectively, 18 and 35 nmol/cm2. For 0.5 mg of fibers, the graft densities for 19 and 38 mM 2-methacryloxyethyl glucoside are, respectively, 150 and 250 nmol/cm2. Comparative study without the preactivation treatment shows the efficiency of the preactivation: for a mass of fiber of 0.5 mg and a 2-methacryloxyethyl glucoside concentration of 38 mM, the grafting density without plasma pretreatment is 38 nmol/cm2. Attenuated total reflectance Fourier transform infrared spectra confirmed the anchoring of the glycopolymer onto the poly(ethylene terephthalate) surfaces. Atomic force microscopy and scanning electronic microscopy pictures indicated their morphological changes.
Introduction In past years, there has been much interest in developing methods to modify or improve the surface properties of polymer materials. Among them, surface graft polymerization induced by plasma treatment is often used.1,2 Surface modification finds numerous applications in the preparation of biomaterials, for example, for the control of interfacial interactions between materials and adsorbing biological molecules.3-5 PET is often used in biomedical material as cardiovascular implants, artificial blood vessels, and others because of its excellent mechanical properties and moderate biocompatibility.6-8 However, for biomedical applications, the polymer surfaces are desired to be highly hydrophilic and able to prevent protein * To whom all correspondence should be addressed. (P.R.) E-mail:
[email protected]. Tel: +33-169154716. Fax: +33-169154715. (C.B.-L.) E-mail:
[email protected]. Tel: +33169158173. Fax: +33-169157844. † UMR 8182 CNRS, Universite ´ Paris-sud 11. ‡ ScienTec. § UMR 8578, Universite ´ Paris-sud 11. (1) Kato, K.; Uchida, E.; Uyama, Y.; Ikada, Y. Prog. Polym. Sci. 2003, 28, 209-259. (2) Zou, X. P.; Kang, E. T.; Neoh, K. G. Surf. Coat. Technol. 2002, 149, 119-128. (3) Siow, K. S.; Britcher, L.; Kumar, S.; Griesser, H. J. Plasma Process. Polym. 2006, 3, 392-418. (4) Poncin-Epaillard, F.; Legeay, G. J. Biomater. Sci., Polym. Ed. 2003, 14, 1005-1028. (5) Wang, J.; Pan, C. J.; Huang, N.; Sun, H.; Yang, P.; Leng, Y. X.; Chen, J. Y.; Wan, G. J.; Chu, P. K. Surf. Coat. Technol. 2004, 196, 307-311. (6) Legeay, G.; Poncin-Epaillard, F.; Arciola, C. R. Int. J. Artif. Organs 2006, 29, 453-461. (7) Sugiyama, K.; Kato, K.; Kido, M.; Shiraishi, K.; Ohga, K.; Okada, K.; Matsuo, O. Macromol. Chem. Phys. 1998, 199, 1201-1208. (8) Huh, M. W.; Kang, I. K.; Lee, D. H.; Kim, W. S.; Lee, D. H.; Park, L. S.; Min, K. E.; Seo, K. H. J. Appl. Polym. Sci. 2001, 81, 2769-2778.
adsorption.9 The plasma treatment alone gives surfaces that are relatively unstable (rapid loss of surface hydrophilicity), so one method is to treat the surface of PET with appropriate monomers to maintain the surface with a high level of hydrophilicity and functionality and to achieve a compatible interface with biological materials. Glycomonomers, giving polymer-containing carbohydrates moieties, are excellent candidates because of their biocompatibility and specific protein recognition.10 These materials are very stable for a long time.11 Kou et al. observed that contact angle measurements on polypropylene membranes modified by the grafting of R-allyl glucoside revealed that the hydrophilicity is permanent and no hydrophobic recovery is observed. In our previous studies,12,13 carbohydrate molecules were grafted onto PET fibers or film surfaces by UV irradiation or reductive amination. We have decided to investigate plasma treatment induced by carbohydrate monomers, so in this study, double plasma treatment has been used to modify the surface properties of PET fibers. The first step is the activation of the PET surface by argon plasma treatment leading to the formation of radicals by the scission of chemical bonds. Then, the fibers are exposed to air immediately after the plasma treatment in order to form peroxides and hydroperoxide functions as previously described in the literature.7,14,15 Then, the activated fibers are (9) Deng, H. T.; Xu, Z. K.; Dai, Z. W.; Wu, J.; Seta, P. Enzyme Microbiol. Technol. 2005, 36, 996-1002. (10) Deng, H. T.; Xu, Z. K.; Wu, J.; Ye, P.; Liu, Z. M.; Seta, P. J. Mol. Catal. B: Enzym. 2004, 28, 95-100. (11) Kou, R. Q.; Xu, Z. K.; Deng, H. T.; Liu, Z. M.; Seta, P.; Xu, Y. Langmuir 2003, 19, 6869-6875. (12) Renaudie, L.; Le Narvor, C.; Lepleux, E.; Roger, P. Biomacromolecules 2007, 8, 679-685. (13) Bech, L.; Meylheuc, T.; Lepoittevin, B.; Roger, P. J. Polym. Sci., Part A: Polym. Chem. 2007, 45, 2172-2183.
10.1021/la701400b CCC: $37.00 © 2007 American Chemical Society Published on Web 08/21/2007
Graft Polymerization on PET
Langmuir, Vol. 23, No. 20, 2007 10349
Scheme 1. Reaction Scheme for the Grafting of Glycosylated Monomers on PET Fibers by Double Plasma Treatmenta
a
Restricted to linear grafted chains, but ramifications cannot be excluded. Scheme 2. Acidic Hydrolysis of Carbohydrates from the Surface and the Formation of Colored Compoundsa
a
Analyzable at 490 nm.
dipped in monomer solution and dried. The second plasma treatment is applied to polymerize the monomers adsorbed on fibers. This strategy is schematically depicted in Scheme 1. Two monomers are studied: allyl R-D-galactopyranoside (AG) and 2-methacryloxyethyl glucoside (GEMA). To our knowledge, few data have been reported using twostep plasma treatment for polymer surface modification,16 and most studies described in the literature preferably use one plasmainduced polymerization after monomer adsorption10,17 or combined surface activation by plasma treatment followed by polymerization initiated by thermal activation7,18 or UV activation.19-21 In this study, we used two glycomonomers with different reactivities in a two-step plasma treatment for PET surface modification. Experimental Part Materials. Poly(ethylene terephthalate) (PET) fibers with a 12 µm diameter were kindly supplied by Tergal Fibres (Gauchy, France). A 50% solution of 2-methacryloxyethyl glucoside mixed anomers in water (GEMA, 2/1 R/β) was purchased from Polysciences, Inc. and was used without purification. Sulfuric acid (95%), trifluoroacetic (14) Inagaki, N.; Tasaka, S.; Narushima, K.; Kobayashi, H. J. Appl. Polym. Sci. 2001, 85, 2845-2852. (15) Shin, Y.; Son, Y.; Yoo, D. I. J. Appl. Polym. Sci. 2007, 103, 3655-3659. (16) Hochart, F.; De Jaeger, R.; Levallois-Gru¨tzmacher, J. Surf. Coat. Technol. 2003, 165, 201-210. (17) Tsafack, M. J.; Hochart, F.; Levallois-Gru¨tzmacher, J. Eur. Phys. J. Appl. Phys. 2004, 26, 215-219. (18) Andreozzi, L.; Castelvetro, V.; Ciardelli, G.; Corsi, L.; Faetti, M.; Fatarella, E.; Zulli, F. J. Colloid Interface Sci. 2005, 289, 455-465. (19) Ying, L.; Yin, C.; Zhuo, X. R.; Leong, K. W.; Mao, H. Q.; Kang, E. T.; Neoh, K. G. Biomacromolecules 2003, 4, 157-165. (20) Zhang, J.; Cui, C. Q.; Lim, T. B.; Kang, E. T. Macromol. Chem. Phys. 2000, 201, 1653-1661. (21) Chen, Y.; Liu, P. J. Appl. Polym. Sci. 2004, 93, 2014-2018.
acid (99%), and phenol (99%) were purchased from Aldrich and were used without purification. Water was purified with a Milli-Q system from Millipore. Allyl R-D-galactopyranoside (AG) was synthesized and purified according to conditions previously described in the literature.22,23 Plasma Setup. The experimental apparatus, as shown in Figure 1, consists of a 12-cm-diameter and 40-cm-high quartz cylinder fixed on top of a post-discharge chamber (stainless steel cylinder, 50 cm diameter, 50 cm height). This reactor has been fully described.24,25 In the post-discharge chamber, the substrate holder was fixed 6 cm from the end of the quartz cylinder. The plasma was generated using a Surfaguide reactor excitator26 fed from opposite sides by two microwave (MW) sources at 2.45 GHz. The argon flux was controlled by mass flow meters. Experimental conditions were the following: pressure fixed at 60 mTorr, argon flux regulated at 150 sccm, both generators set at 500 W power. Fibers were exposed to plasma for 1 min during pretreatment and for 2 min during treatment. PET Samples. Each PET fiber sample was precisely weighted using a Sartorius MC210S balance with a precision of (0.05 mg. Plasma Pretreatment of PET Surfaces. The PET fibers were cleaned by Soxhlet extraction with dichloromethane for 24 h. They were dried by vacuum overnight. The fibers (precisely weighed) were introduced into the plasma reactor. The pressure was reduced (22) Talley, E. A.; Vale, M. D.; Yanovsky, E. J. Am. Chem. Soc. 1945, 67, 2037-2039. (23) Goebel, M.; Nothofer, H. G.; Ross, G.; Ugi, I. Tetrahedron 1997, 53, 3123-3134. (24) Bechu, S.; Boisse-Laporte, C.; Leprince, P.; Marec, J. J. Vac. Sci. Technol., A 1997, 15, 668-672. (25) Bennissad, N.; Boisse-Laporte, C.; Valle´e, C.; Granier, A.; Goullet, A. Surf. Coat. Technol. 1999, 116-119, 868-873. (26) Moisan, M.; Zakrzewski, Z. J. Phys. D: Appl. Phys. 1991, 24, 10251048.
10350 Langmuir, Vol. 23, No. 20, 2007
Bech et al.
Figure 2. Grafting density (nmol/cm2) of AG on the PET surface determined by colorimetric titration as a function of the surface area (cm2) of PET fibers exposed to plasma discharge. Figure 1. Plasma reactor used with a quartz cylinder (a), post discharge chamber (b), substrate holder (c), and Surfaguide reactor excitator (d). to about 15 mTorr in 15 min, and then the fibers were treated with plasma for 1 min. Plasma Treatment of PET Surfaces. The fibers, which were pretreated with plasma, were immersed into 2 mL of an aqueous solution of monomer in a hemolysis tube. After 30 min, the fibers were dried overnight to constant weight. Then the fibers were introduced into the plasma reactor. The pressure was reduced to 20-30 mTorr, and then the fibers were treated with plasma for 2 min. The fibers were washed with water and dried to constant weight. The conversion rate (G) was calculated with the following equation G)
w 2 - w0 w1 - w0
(1)
where w0 is the mass of the native fibers, w1 is the mass of the fibers with the dried adsorbed monomer, and w2 is the mass of the fibers after polymerization, washing, and drying. The values are the result of the mean of at least five measurements, and the standard deviations were calculated. Colorimetric Titration. PET fibers (precisely weighed) were placed in a tube with a screw. A solution of trifluoroacetic acid (2 M in water) was added. The mixture was held at 120 °C for 2 h, and then the solution was transferred to a balloon and evaporated. The fibers were washed three times with methanol, and the solution was added to the balloon. The solvents were evaporated, and 500 µL of water was added to solubilize the carbohydrates. Then 200 µL of this solution was placed in a hemolysis tube, where 200 µL of a phenol solution (6% in water) and 1 mL of sulfuric acid were added. The mixture was stirred and analyzed by UV-visible spectroscopy at 490 nm. The concentration was determined by reference to a calibration curve with glucose as the standard for 2-methacryloxyethyl glucoside and galactose as the standard for allyl R-D-galactopyranoside. Characterization. Samples were visualized by scanning electron microscopy. Before analysis, each sample was gold sputtered and viewed as secondary electron images (8 kV) with a Hitachi S4500 SEM. The diameter is the average obtained by the statistical treatment of at least 50 measurements. AFM observations were carried out in air at atmospheric pressure with a PICO LE (Molecular Imaging, Tempe, AZ) microscope. AFM images were acquired exclusively in tapping mode using a silicon cantilever. For the analysis, one fiber was fixed on a mica support by gluing each extremity of the fiber with double-coated tape. For a line containing N data points, the mean roughness (Ra) was calculated from eq 2
x
N
∑ (Z - Zh )
Ra )
Results and Discussion The influence of two parameters is studied: (1) the quantity of fibers exposed to plasma discharge (0.50, 1.00, 2.00, and 5.00 mg) and (2) the type of monomer (allylic or methacrylic) and its water concentration (4.0, 10.0, 19.0, and 38.0 mM for GEMA and 23.0, 45.0, and 68.0 mM for AG). After a thorough washing to eliminate non-grafted byproducts, the quantity of grafted carbohydrate is quantitatively determined by a phenol/sulfuric acid colorimetric method. This titration method is simple, rapid, sensitive, and specific to carbohydrates.27,28 Sugar-coated fibers were treated with trifluoroacetic acid to hydrolyze the glycosidic bond. Then, the sugar solubilized in water was reacted with phenol/sulfuric acid to form a colored compound according to Scheme 2. The sugar concentration was obtained spectrophotometrically after the determination of a calibration curve. The results are plotted in Figures 2 and 3. With the specific surface of one fiber of 12 µm diameter being equal to 2.51 cm2/ mg, the grafting density can be expressed using conventional units (nmol/cm2). The grafting density varies drastically according to reaction conditions. The grafting densities are quantified by colorimetric titration and are presented in Figure 3a,b. Different observations can be made. The first is that the grafting density of GEMA is superior or equal to the grafting density of AG and under optimal conditions GEMA polymerizes around 5 times more than does AG. For GEMA, the grafting densities are in the range of 50-250 nmol/cm2 (Figure 3a) whereas for AG the values are below 40 nmol/cm2 (Figure 2). This effect can be explained by the higher reactivity of GEMA bearing a methacrylic function compared to the less reactive allylic monomer. The conversion rate calculated as described in the Experimental Part is given in Table 1.
2
i
i)1
N
where Z h is the mean Z height. The Ra has been calculated for the total image sample after second-order flatness treatment of the raw data to take into account the curvature of the fiber surface. Images have been recorded in different zones in order to be representative of the total sample surface state. The infrared analyses were carried out on Brucker IFS 66 equipment with an ATR module with a diamond crystal from Pike Technologies. Sixty-four scans were performed for reference, and the samples had a resolution of 4 cm-1. The UV-visible spectroscopy analyses were carried out on Cary 1E equipment from Varian. The grafting density values are the result of the mean of at least five measurements, and the standard deviations were calculated.
(2)
(27) Dubois, M.; Gilles, K. A.; Hamilton, J. K. Anal. Chem. 1956, 28, 350356. (28) Lepoittevin, B.; Masson, S.; Huc, V.; Haut, C.; Roger, P. e-Polymer 2006, 32, 1-15.
Graft Polymerization on PET
Langmuir, Vol. 23, No. 20, 2007 10351
Figure 4. FTIR-ATR spectra of PET fibers: (A) native and (B) 10 mM and (C) 19 mM (C) grafted (GEMA).
Figure 3. Grafting density (nmol/cm2) of GEMA on the PET surface determined by colorimetric titration as a function of the surface area (cm2) of PET fibers exposed to plasma discharge with pretreatment (a) or without pretreatment (b). Table 1. Conversion Rate for AG (68 mM) and GEMA (33 mM) by Using Double Plasma Treatment Polymerization entry
monomer
mass of fibers (mg)
conversion rate (%)a
1 2 3 4 5 6
AG AG AG GEMA GEMA GEMA
0.50 1.00 2.00 0.50 1.00 2.00
3(1 1(1 1(1 71 ( 7 31 ( 5 36 ( 7
a
Calculated using eq 1.
As seen in Figure 2 and in Table 1, AG has a low reactivity compares to GEMA. Indeed, in the free radical polymerization of allylic monomers, terminations proceed by chain transfer to monomer, which leads to a partial inhibition of polymerization.29 The maximum value obtained for a fiber mass of 0.50 mg (entry 1) is equal to 3%. When the fiber mass increases, the conversion of AG decreases to 1% (entries 2 and 3). For GEMA, a higher conversion is observed, and the maximum value obtained for a fiber mass of 0.50 mg (entry 4) is equal to 71%. When the fiber mass increases, the conversion rate of GEMA decreases to around 30% (entries 5 and 6). The second observation is that the grafting density decreased when the surface of the fibers increased, whatever the monomer (29) Matsumoto, A. Prog. Polym. Sci. 2001, 26, 189-257.
concentration used. This could be explained by a screening effect occurring during the plasma treatment. The plasma is composed of a great number of reactive species such as electrons, free radicals, metastable species, ions, and UV photons. We have shown previously that outer fibers in the shell prevent the UV grafting of the fibers in the core by the very strong screening effect of PET for the far UV (
