Immobilization of polysaccharides on a fluorinated silicon surface

Colloids and Surfaces B: Biointerfaces 47 (2006) 57–63

Immobilization of polysaccharides on a fluorinated silicon surface Anfeng Wang a , Ting Cao a , Haiying Tang a , Xuemei Liang a , Carolyn Black a,b , Steven O. Salley a , James P. McAllister b , Gregory W. Auner c , K.Y. Simon Ng a,∗ a

Department of Chemical Engineering and Materials Science, Wayne State University, 5050 Anthony Wayne Drive, Detroit, MI 48202, United States b Department of Neurological Surgery, Wayne State University, 4201 Antoine Street, UHC-6E, Detroit, MI 48201, United States c Department of Electrical and Computer Engineering, Wayne State University, 5050 Anthony Wayne Drive, Detroit, MI 48202, United States Received 12 September 2005; received in revised form 16 November 2005; accepted 16 November 2005

Abstract A self-assembled monolayer (SAM) of fluoroalkyl silane (FAS) was deposited on a silicon surface by chemical vapor deposition (CVD) at room temperature under 1.01 × 105 Pa nitrogen. Using this new approach, the quality and reproducibility of the SAM are better than those prepared either in solution or by vapor phase deposition, and the deposition process is simpler. In this modified CVD process, the silane monomers, instead of the oligomeric species, are the primary reactants. Full coverage of the silicon surface by FAS molecules was achieved within 5 min. Heparin and hyaluronan, two naturally occurring biocompatible polysaccharides, were successfully covalently attached on the FAS SAM/Si surface by photo-immobilization. Atomic force microscopy (AFM) revealed the morphologic changes after the immobilization of heparin and hyaluronan, and X-ray photoelectron spectroscopy (XPS) confirmed the change in chemical compositions. Such combination of coatings is expected to enhance the stability and biocompatibility of the base material. © 2005 Elsevier B.V. All rights reserved. Keywords: Heparin; Hyaluronan; Silicon; Fluoroalkyl silane; Chemical vapor deposition; Photo-immobilization

1. Introduction Fluoroalkyl silane (FAS) coated surfaces are both highly hydrophobic and oleophobic [1], and they have demonstrated superb wear resistance and friction reduction. Thus modified surfaces have proven to be super-hydrophobic and super-stable on the surfaces of polymer and silica fillers (in denture composites) [2,3]. FAS coatings have found applications in adhesion control, low-soil coatings, lubricating treatment, textile treatments, and protective coatings [4–6]. They have also been found to be capable of improving the particle trapping efficiency of glass air filters [7], and to reduce bacterial infection [8] on biomaterials. FAS coatings have shown good biocompatibility, which has been attributed to their extreme low surface energies (∼8 mJ/m2 ) [9–13]. The surfaces of metals, semiconductive materials and polymers have often been modified by the deposition of a selfassembled monolayer (SAM) of FAS. Traditionally, such self-

∗

Corresponding author. Tel.: +1 313 577 3805; fax: +1 313 577 3810. E-mail address: [email protected] (K.Y.S. Ng).

0927-7765/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfb.2005.11.016

assembly on solid surfaces was performed in solution [14–18]. However, there are several shortcomings to this approach: (1) the quality of the SAM is strongly affected by the quality of the solvents, especially the water content, even at very low concentration; (2) most of the organic solvents are toxic to a certain extent; (3) some surfaces (e.g., polymers) are not compatible with organic solvents due to possible swelling, molecular rearrangements and even dissolution; (4) depositing SAMs on tiny delicate devices (e.g., AFM tips, silicon electrode arrays) is extremely challenging, since the coating solution tends to be trapped inside holes, gaps or edges, and many areas could not be coated at all [4]. Furthermore, some alternate structures, including inverse micelles and lamellar phases, can form in the deposition solutions, which are detrimental to the performance of the final SAMs [19]. Adsorption of silane oligomers on the solid surfaces, formed by hydrolysis and condensation reactions in solution, has often been found to occur, slower than but competitive with SAM formation at the liquid–solid interface [16]. These problems can be avoided by using a chemical vapor deposition (CVD) technique. In the literature, CVD of FAS SAMs has been performed under vacuum and/or at elevated temperature [2,7,20–26]. Sometimes the temperature can be

58

A. Wang et al. / Colloids and Surfaces B: Biointerfaces 47 (2006) 57–63

as high as 190 ◦ C [27]. However, high temperature (≥90 ◦ C) often resulted in SAMs with poor quality and reproducibility [4]. Organic solvents, which formed vapor together with FAS molecules, were even sometimes used as the dilution media during the CVD deposition process [2,11,28]. Excess amounts of FAS molecules on the surface after CVD were typically removed by subjecting the surfaces to a vacuum at elevated temperature for a few hours. Aggregates of FAS molecules were often observed on the FAS SAM modified surface when formed by the CVD process. We found that a very smooth, uniform and reproducible FAS SAM can be formed on silicon surfaces at room temperature without using vacuum and organic dilution solvents, as evidenced by the AFM and XPS results. This CVD process is simpler than those adopted previously as well [2,7,20–26]. In this paper, FAS SAM on a silicon surface prepared by this simplified CVD process is reported. The deposition process was performed under an ultra-dry nitrogen atmosphere to minimize water interference. The vapor pressure of FAS13 Cl (CF3 (CF2 )5 (CH2 )2 SiCl3 ) is approximately 40 Pa at room temperature [4], which is high enough to facilitate the SAM formation, and full surface coverage was achieved within only a few minutes. In the literature, plasma was often combined with CVD techniques to produce coatings on solid surfaces. Such plasmaenhanced coatings typically demonstrate less pure structures and much lower stability, although they are much thicker (hundreds of nanometers to tens of microns) than the SAMs (1–3 nm), furthermore the plasma sources often pose a high risk of damaging the samples to be coated [29]. Fluorinated surfaces prevent the adhesion of most chemical and biological substances [30]. Although fluorinated surfaces showed good biocompatibility, their biocompatibility still needs to be improved for in vivo applications [31–36]. Inertness and low surface energy are characteristic advantages of fluorinated surfaces. However, these very characteristics make it difficult to functionalize them with chemical or biological species. While there are published clinical results on heparin modified fluorinated surfaces to enhance their biocompatibility, the heparin is either ionically bound or claimed covalently bound without defined, reproducible procedures [31–34,36]. In this paper, heparin and hyaluronan were successfully covalently immobilized on a FAS SAM modified silicon surface. Heparin and hyaluronan were first modified by attaching photosensitive groups on them, and then under UV illumination, they were covalently immobilized on a FAS SAM coated silicon surface. So far such UV-activation based photo-immobilization has only been applied to organic surfaces with an abundance of C H, C N groups [37–40]. We demonstrate here that this photo-immobilization technique can also be applied to fluorinated surfaces, which is regarded as the most inert among all the organic surfaces. The combination of heparin (or hyaluronan) and FAS SAM is expected to bring enhanced stability and biocompatibility to the substrates [41,42]. Our objective is to make long-term implantable neuroprostheses, and satisfactory biostability and biocompatibility are two extremely important criteria. The in vivo experiments to evaluate the biostability and biocompatibility on animals are ongoing.

2. Experimental 2.1. Materials One-side polished N type silicon (1 1 1) wafers (test grade, with resistivity 1–2
cm and thickness 475–575 ␮m) were purchased from Wafer World Inc. (West Palm Beach, FL). Deionized water (DI water) with a resistivity of ≥18 M
cm was obtained by a Barnstead Nanopure Systems (Dubuque, Iowa). Trichloro (3,3,4,4,5,5,6,6,7,7,8,8,8tridecafluorooctyl)silane (FAS13 Cl, CF3 (CF2 )5 (CH2 )2 SiCl3 , 97%), isooctane (anhydrous), N-(3-dimethylaminopropyl)-N ethylcarbodiimide hydrochloride (water-soluble carbodiimide, WSC), 4-azidoaniline hydrochloride, heparin (sodium salt from porcine intestinal mucosa, ∼170 USP units/mg), and hyaluronic acid potassium salt (extracted from the human umbilical cord, named hyaluronan throughout this paper) were purchased from Sigma–Aldrich (St. Louis, MO). They were used as received. 2.2. FAS13 Cl SAM on silicon by CVD The pre-cut silicon pieces (approx. 1.2 cm × 1.2 cm) were rinsed with ethanol, acetone and then oxidized by the RCA (named after the company Radio Corporation of America, where the procedure was first developed) method [43], resulting in hydroxyl groups on top of the silicon oxide layer [44,45]. (Such treated silicon pieces are referred to as RCA Si hereafter.) The silicon pieces were then blown dry with nitrogen before SAM deposition. The CVD process was performed in a glove box purged with nitrogen (extra dry). Two drops of FAS13 Cl (∼50 ␮L) was placed in an enclosed glass container (∼500 mL in volume) with RCA Si inside, and the silicon pieces were taken out after predetermined durations. To remove excess FAS13 Cl molecules physically adsorbed on the surface, the modified RCA Si was rinsed with isooctane twice, and blown dry with nitrogen. 2.3. Photo-immobilization of heparin and hyaluronan on FAS13 Cl modified silicon Heparin, 4-azidoaniline hydrochloride and WSC at a weight ratio of 18.2:7.75:10 were dissolved in DI water to yield a 0.5% solution [38]. The pH of the solution was adjusted to 4.70–4.75 with 1N NaOH at room temperature. The solution was stirred at 4 ◦ C for 24 h, resulting in the aryl azido-modified heparin solution. 0.2% aryl azido-modified hyaluronan solution was prepared by the same means except that the weight ratio of hyaluronan, 4-azidoaniline hydrochloride and WSC was 42:17:28 [37]. FAS13 Cl SAM/Si pieces were immersed in an aryl azidomodified heparin (or hyaluronan) solution in a Teflon container with 3 mm height of solution above them. They were then illuminated with a mercury vapor UV lamp (175 W, Regent Lighting, Burlington, NC) for 5 min at a distance of 10 cm. The heparin or hyaluronan immobilized FAS13 Cl SAM/silicon samples were immersed for 48 h in DI water, which was refreshed every 12 h, and the samples were rinsed by DI water every 12 h.

A. Wang et al. / Colloids and Surfaces B: Biointerfaces 47 (2006) 57–63

59

2.4. Surface characterization The surface hydrophilicity was measured with a Rame-hart NRL contact angle goniometer (Model 100, Landing, NJ) in the laboratory atmosphere. A 20 ␮L DI water droplet was placed on the substrate and the static contact angles were measured on both sides of the droplet. Three droplets were placed at various spots on the substrate surface and the average readings are reported. XPS analysis of the surfaces was conducted on a PHI 5500 Spectrometer (Perkin Elmer, Wellesley, MA) equipped with an aluminum K␣ X-ray radiation source (1486.6 eV) and AugerScan system control (RBD Enterprises, Bend, OR). The pressure in the chamber was below 2 × 10−9 Torr (i.e. 2.67 × 10−7 Pa) before the data were taken, and the voltage and power of the anode were 15 kV and 200 W, respectively. The take-off angle was set at 45◦ . The pass energies for survey and multiplex scans were 117.40 and 23.50 eV, respectively. The binding energy scale was referenced by setting the C 1s peak maximum at 285.0 eV for RCA Si, while by setting the F 1s peak maximum at 689.0 eV for the other modified surfaces. Shirley baseline subtraction was applied to all the peaks before the atomic concentration calculation and curve-fitting by AugerScan software (Version 3, RBD Enterprises, Bend, OR). Each individual peak was of Gaussian shape after the fitting. AFM images were obtained with a Nanoscope IIIa (Digital Instruments, Santa Barbara, CA) in tapping mode, and only the height images are presented here. An E-scanner (maximum scan size of 16 ␮m × 16 ␮m) and a silicon probe (length: 125 ␮m, resonance frequency: between 244 and 366 kHz) were used. The probe was washed with ethanol and illuminated under UV light by a fiber optical illuminator (Dolan-Jenner Industries, Inc., Lawrence, MA) for more than 5 min before use, as recommended by the manufacturer. 3. Results and discussion 3.1. Hydrophobicity The static contact angles with DI water versus deposition time at room temperature and atmospheric pressure are shown in Fig. 1. The contact angle reaches a maximum of 102◦ and levels off after 5 min. This value is comparable to that observed for these surfaces prepared in solutions (100◦ prepared in isooctane solution [19] and 103◦ prepared in toluene solution [14]), and slightly lower than the contact angle for Teflon (108◦ ) [29], which is fully fluorinated. Radio-frequency (rf) plasmaenhanced CVD or microwave plasma-enhanced CVD produced fluorinated surfaces with contact angles as high as 108◦ and 160◦ , respectively [29,46]. However, those modified surfaces have multilayers of the fluoroalkyl silane (FAS) molecules (over hundreds of nanometers in thickness), and the chemical nature of the fluorinated silane molecules was altered as well. Moreover, the silane monomers were not oriented on the surfaces, and the surface roughness was increased significantly [29,46] due to the effect of plasma (generated by radio-frequency or microwave). The optical property of the base material was severely impaired by both plasma-enhanced CVD techniques [29], and the samples

Fig. 1. The water static contact angle and fluorine content (F%) of FAS13 Cl modified silicon surface versus CVD time. The CVD was performed at room temperature under nitrogen atmosphere (760 Torr). The contact angle levels off at the maximum of 102◦ after 5 min, and F% reaches the maximum of 32.7% in 5 min as well.

were subjected to a high risk of being damaged by the plasma source. The static contact angles with DI water at each preparative stage are listed in Table 1. After the photo-immobilization of heparin and hyaluronan, the hydrophobicity was reduced significantly, and the contact angles are 45◦ and 44◦ , respectively. The properties of a material’s outermost surface (only a few nanometers in thickness) play an important role in mediating the interactions between the physiological environments and the alien objects [47]. Hydrophilic surfaces typically demonstrate better biocompatibility, and they were also found to be able to reduce the chance of bacterial infection of implanted biomedical devices [48–50]. The deposition of FAS13 Cl SAM on RCA Si surface is much faster than its counterpart FAS13 OMe (CF3 (CF2 )5 (CH2 )2 Si(OCH3 )3 ), since the former has a significantly higher vapor pressure (40 Pa at room temperature [4]). It took 2 h for the fluorine content (F%) and contact angle to reach the maximal plateaus during CVD of FAS13 OMe on silicon at 100 ◦ C [21]. 3.2. XPS results Fig. 2 shows the XPS survey spectra for the surfaces at each preparation step, and the elemental compositions on the top surfaces are summarized in Table 2. A thin layer of SiOx (1.5–2 nm in thickness) was introduced on the silicon surface after the RCA treatment, which makes the oxygen content on Table 1 Static contact angle with DI water for the silicon surfaces at each preparative stage Substrates

Contact angle (◦ )

Si FAS/Si Heparin/FAS/Si Hyaluronan/FAS/Si

2 102 45 44

± ± ± ±

0.5 0.3 0.8 0.9

Aryl azido-modified heparin and hyaluronan concentrations were 0.5% and 0.2%, respectively, and the UV illumination time was 5 min.

A. Wang et al. / Colloids and Surfaces B: Biointerfaces 47 (2006) 57–63

60

Fig. 2. XPS survey spectra of: (a) RCA Si, (b) FAS13 Cl SAM/Si, (c) heparin/FAS13 Cl SAM/Si, and (d) hyaluronan/FAS13 Cl SAM/Si.

RCA Si 38.1%. Due to adventitious carbon contamination, carbon content on RCA Si surface is 9.8%. Deposition of FAS13 Cl SAM introduced fluorine (F) as the new element onto the surface, and increased the carbon (C) composition to 20.3% (Table 2). Chlorine (Cl) was not detected on the FAS13 Cl SAM modified surface, which originally existed in the FAS13 Cl monomers. Therefore, all the FAS13 Cl molecules on the RCA Si surface were linked with each other or onto the surface via Si O Si bonds. Subsequent heparin and hyaluronan immobilization introduced nitrogen (N) as the new element onto the surfaces, and further increased the carbon composition. The gradual decrease of silicon (Si) composition (Table 2) indicates that more and more organic substances are present on the surface after each coating step. The loadings of heparin and hyaluronan on the surfaces could be controlled by varying their concentrations and/or UV illumination time during the photoimmobilization process (data not shown). Photo-immobilization of heparin on FAS13 Cl SAM/Si surface should yield sulfur (S) as a new element. However, because of severe interference from the nearby Si 2s peak, multiplex scans of S 2p3 absorption gave unreliable results for the S content on the heparin FAS13 Cl SAM/Si surface. Fig. 3 shows the detailed C 1s spectra of the FAS13 Cl SAM/Si, heparin/FAS13 Cl SAM/Si and hyaluronan/FAS13 Cl SAM/Si surfaces. Fig. 3a is composed of six components centered at binding energies (BEs) of 283.6, 284.9, 286.3, 290.4, 291.3 and 293.6 eV, which correspond to C Si, CH2 CH2 , Table 2 Elemental compositions at the surfaces at each preparative stage obtained by XPS detailed scans Surfaces

C%

F%

N%

O%

Si%

Si FAS/Si Heparin/FAS/Si Hyaluronan/FAS/Si

9.8 20.3 41.2 41.2

0 29.1 13.1 12.6

0 0 6.3 5.3

38.1 21.0 21.6 24.2

52.1 29.6 17.9 16.7

Si and FAS represent RCA Si and FAS13 Cl SAM, respectively. FAS SAM was formed on the RCA Si by CVD for 2 h at room temperature and under atmosphere pressure with nitrogen protection. The aryl azido-modified heparin and hyaluronan concentrations were 0.5% and 0.2%, respectively. The UV illumination time is 5 min.

Fig. 3. Curve-fitting of C 1s spectra for: (a) FAS13 Cl SAM/Si, (b) heparin/FAS13 Cl SAM/Si, and (c) hyaluronan/FAS13 Cl SAM/Si. In addition to the six deconvoluted peaks in (a), (b) and (c) have two additional components, as indicated in the figures. Each component peak corresponds to carbon atoms at different oxidation state.

C O, CF2 CH2 , CF2 CF2 , and CF3 CF2 , respectively. This is in good agreement with Hozumi et al.’s results [21]. As shown in Fig. 3b and c, after the photo-immobilization of heparin and hyaluronan on FAS13 Cl SAM/Si, two more components appeared at 287.3–287.7 and 288.6–288.9 eV, which correspond to C O, and COO, respectively. The relative concentrations of carbon at different oxidation states based on the detailed XPS scans are summarized in Table 3. The portions of carbon at oxidation states of C O and COO among the total carbon are 7.2% and 4.6% for heparin/FAS13 Cl SAM/Si, and 11.6% and 3.2% for hyaluronan/FAS13 Cl SAM/Si. After the photo-immobilization of heparin and hyaluronan, the relative concentration of C O was also significantly increased, while the relative concentrations of CF2 CH2 , CF2 CF2 , and CF3 CF2 were dramatically decreased. All these data clearly indicate that each coating step is successful. As a comparison, our XPS results showed that FAS13 OMe on RCA Si by this simplified CVD process only generated the modified surface with fluorine content (F%) of 8.5% after 2 h,

A. Wang et al. / Colloids and Surfaces B: Biointerfaces 47 (2006) 57–63

61

Table 3 The relative concentrations (%) of carbon at different oxidation states based on the curve-fitting results of detailed XPS scans of C 1s Surfaces

C Si

CH2 CH2

C O

CF2 CH2

CF2 CF2

CF3 CF2

C O

COO

FAS13/Si Hep/FAS13/Si Hya/FAS13/Si

3.3 3.3 0.8

22.1 37.3 33.5

16.9 30.7 35.9

6.0 3.1 3.0

41.3 9.5 8.7

10.5 4.2 3.3

7.2 11.6

4.6 3.2

FAS13, Hep and Hya represent FAS13 Cl SAM, heparin and hyaluronan, respectively.

due to much lower vapor pressure of FAS13 OMe at room temperature. 3.3. AFM results Fig. 4 shows the AFM images of FAS13 Cl SAM/Si, heparin/FAS13 Cl SAM/Si and hyaluronan/FAS13 Cl SAM/Si surfaces. Their root-mean-square (RMS) values, which is a measure of the surface roughness, are 0.160, 0.234 and 0.892 nm, respectively. The FAS13 Cl SAM/Si surface was very uniform and smooth at the atomic level (Fig. 4a), while large particles (mostly in globular shape) were often found to be present by other researchers on their silicon surfaces modified with

FAS prepared in either solution or vapor phase (CVD) [4,19]. After the photo-immobilization of heparin, the surface roughness increased slightly, and the modified surface still appeared to be relatively smooth (Fig. 4b). However, after the photoimmobilization of hyaluronan, the surface roughness increased dramatically, and large aggregates ranging from 10 to 40 nm in diameter were observed (Fig. 4c). The increase of surface roughness was undoubtedly due to the immobilization of heparin and hyaluronan. The average molecular weights (by number) for heparin and hyaluronan are approximately 16,500 and 750,000, which results in the morphologic difference on the surfaces modified by them. 3.4. Mechanism of CVD

Fig. 4. AFM images of the surfaces of: (a) FAS13 Cl SAM/Si, (b) heparin/FAS13 Cl SAM/Si, and (c) hyaluronan/FAS13 Cl SAM/Si.

SAM is a robust film that is highly crosslinked both laterally and to the solid substrate surface, and the mechanism for its formation from solution phase is still under debate. Some believed that the hydrolysis of Si Cl has to precede the crosslinking among the monomers at the interfaces [51]. Therefore, Sung et al. suggested that the presence of water is very important in the SAM formation, even more important than the OH groups on the surface [52]. Some researchers have asserted that the oligomeric silanes are more effective for surface modification than the monomeric ones, which means that the SAM was partially pre-assembled in the solution before anchoring to the solid surfaces [53]. Wang et al. [54] reported that the oligomeric component of hydrolyzed silanes in solution adsorb onto glass surfaces before forming chemical bonds to the glass and simultaneous polymerization into a three-dimensional siloxane network. Similar findings were reported by Ikuta et al. [55]. They concluded that the true reactants in the surface modification are the oligomeric species, instead of the monomeric ones. Preparing SAMs in solution phase poses another problem. Self-assembly does not simply stop at the monolayer stage; instead, hydrolysis and oligomerization of the silane in solution and adsorption of oligomeric siloxane species also occur in competition with the SAM formation [19]. In our CVD process, the whole deposition procedure was performed in the vapor phase under a nitrogen atmosphere. The water content is significantly lower than when prepared in solutions. R-SiCl3 has a very high reactivity toward OH groups (which exists both in water and on RCA Si surface), and the FAS13 Cl monomers evaporating from the droplet in the deposition chamber should not be hydrolyzed. Oligomeric components can likely be ruled out as the reactants, due to their significantly lower vapor pressure at room temperature than the monomeric ones. The chemical structures of FAS13 Cl monomer, dimer

62

A. Wang et al. / Colloids and Surfaces B: Biointerfaces 47 (2006) 57–63

Fig. 5. Chemical structures of monomer, dimer and trimer of FAS13 Cl (CF3 (CF2 )5 (CH2 )2 SiCl3 ), and monomer of FAS17 Cl ((CF3 (CF2 )7 (CH2 )2 SiCl3 )).

and trimer, and FAS17 Cl (CF3 (CF2 )7 (CH2 )2 SiCl3 ) monomer are shown in Fig. 5. FAS17 Cl was found to have significantly lower vapor pressure than FAS13 Cl at room temperature and considerably longer process time was needed to achieve full SAM coverage [4]. The dimer and trimer of FAS13 Cl have even lower vapor pressure than FAS17 Cl monomer, due to longer combined carbon chain length and higher molecular weight. Oligomers of FAS13 Cl with more than three repeating units are even more unlikely to form the SAM on the solid surface during our CVD process. The mechanism for the FAS13 Cl SAM formation on RCA Si is illustrated in Fig. 6. Water is not necessary for the silane molecules to covalently attach to the RCA Si surface, which is the first step of the deposition reaction. Once this surface comes into contact with water vapor when the sample is exposed to the atmosphere, the other Si Cl bonds quickly hydrolyze (Step 2), and crosslink with one another laterally to form the SAM (Step 3). Hydrochloric acid (HCl) is the by-product of the reaction, which is also a catalyst for the SAM formation. Mayer et al. [4] reported that the density of FAS13 Cl monomers on the surface is approximately 3 × 1014 molecules/cm2 , therefore only a trace

amount of water (approximate 9 ng/cm2 ) is sufficient to facilitate the formation of a complete SAM. 4. Conclusions Self-assembled monolayer (SAM) of fluoroalkyl silane (FAS) could be formed on solid surfaces at room temperature and atmospheric pressure via chemical vapor deposition (CVD). An ultra-dry nitrogen environment can ensure the quality of the formed SAM by eliminating water interference. This new approach yielded an extremely uniform and smooth modified surface, while previous CVD processes prepared at high temperature and/or under vacuum often resulted in the formation of aggregates. The surface is fully covered by the FAS molecules (evidenced by the results of AFM and XPS) within 5 min and the modified surface is very hydrophobic. FAS monomers in the vapor phase are the reactants during the SAM deposition process, and it is very prohibitive for FAS oligomers to coexist in the vapor phase. Heparin and hyaluronan, two naturally existing biocompatible polymers were covalently attached to the fluorinated surface by photo-immobilization, and the hydrophobicity of the surface was dramatically reduced after such immobilization. The biocompatibility and biostability of such modified surfaces will be investigated in continuing studies in vivo on animals. Acknowledgments Financial support of this research from the Michigan Life Science Corridor (MLSC) is gratefully acknowledged. We would also like to thank Prof. Guangzhao Mao (Department of Chemical Engineering and Materials Science, Wayne State University) for the use of AFM and contact angle goniometer. References

Fig. 6. Reaction mechanism for FAS13 Cl SAM formation on silicon surface in our simplified CVD process. (1) The silane monomers in the vapor phase first anchor onto the hydroxylized silicon surface by forming covalent bond with the release of HCl; (2) the anchored silane monomers are hydrolyzed; and (3) they react with each other to form the crosslinked Si O Si network, once the surface is exposed to the atmosphere. Very trace amount of water is needed to facilitate the formation of a complete SAM, and approximately 9 ng water is needed for each square centimeter of RCA Si.

[1] T. Kawase, H. Sawada, J. Adhes. Sci. Technol. 16 (2002) 1121. [2] J. Genzer, K. Efimenko, Science (Washington, DC) 290 (2000) 2130. [3] T. Nihei, S. Kurata, Y. Kondo, K. Umemoto, N. Yoshino, T. Teranaka, J. Dent. Res. 81 (2002) 482. [4] T.M. Mayer, M.P. de Boer, N.D. Shinn, P.J. Clews, T.A. Michalske, J. Vac. Sci. Technol. B: Microelectr. Nanometer Struct. 18 (2000) 2433. [5] K. Ogawa, T. Nomura, Matsushita Electric Industrial Co., Ltd., Japan, Application EP EP 2001, p. 25. [6] M.J. Owen, D.E. Williams, in: K.L. Mittal (Ed.), Silanes and Other Coupling Agents, VSP, 1992, p. 67. [7] H. Tada, K. Shimoda, K. Goto, J. Electrochem. Soc. 142 (1995) L230.

A. Wang et al. / Colloids and Surfaces B: Biointerfaces 47 (2006) 57–63 [8] C. Price, M.G.J. Waters, D.W. Williams, M.A.O. Lewis, D. Stickler, J. Biomed. Mater. Res. 63 (2002) 122. [9] M.K. Chaudhury, Biosens. Bioelectron. 10 (1995) 785. [10] M.K. Chaudhury, Curr. Opin. Colloid Interf. Sci. 2 (1997) 65. [11] M.K. Chaudhury, G.M. Whitesides, Science (Washington, DC, United States) 255 (1992) 1230. [12] J.D. Swalen, D.L. Allara, J.D. Andrade, E.A. Chandross, S. Garoff, J. Israelachvili, T.J. McCarthy, R. Murray, R.F. Pease, et al., Langmuir 3 (1987) 932. [13] J. Wang, G. Mao, C.K. Ober, E.J. Kramer, Macromolecules 30 (1997) 1906. [14] R.E. Geer, D.A. Stenger, M.S. Chen, J.M. Calvert, R. Shashidhar, Y.H. Jeong, P.S. Pershan, Langmuir 10 (1994) 1171. [15] H.I. Kim, T. Koini, T.R. Lee, S.S. Perry, Tribol. Lett. 4 (1998) 137. [16] M.J. Pellerite, E.J. Wood, V.W. Jones, J. Phys. Chem. B 106 (2002) 4746. [17] H. Sugimura, N. Nakagiri, J. Vac. Sci. Technol. A: Vac. Surf. Films 14 (1996) 1223. [18] A. Ulman, An Introduction to Ultrathin Organic Films: From Langmuir–Blodgett to Self-assembly, Academic Press, 1991. [19] B.C. Bunker, R.W. Carpick, R.A. Assink, M.L. Thomas, M.G. Hankins, J.A. Voigt, D. Sipola, M.P. de Boer, G.L. Gulley, Langmuir 16 (2000) 7742. [20] P.W. Hoffmann, M. Stelzle, J.F. Rabolt, Langmuir 13 (1997) 1877. [21] A. Hozumi, K. Ushiyama, H. Sugimura, O. Takai, Langmuir 15 (1999) 7600. [22] H. Sugimura, A. Hozumi, T. Kameyama, O. Takai, Surf. Interf. Anal. 34 (2002) 550. [23] H. Sugimura, K. Ushiyama, A. Hozumi, O. Takai, J. Vac. Sci. Technol. B: Microelectron. Nanometer Struct. 20 (2002) 393. [24] H. Tada, J. Electrochem. Soc. 142 (1995) L11. [25] H. Tada, H. Nagayama, Langmuir 10 (1994) 1472. [26] H. Wu, V.R. Thalladi, S. Whitesides, G.M. Whitesides, J. Am. Chem. Soc. 124 (2002) 14495. [27] U. Joensson, G. Olofsson, M. Malmqvist, I. Roennberg, Thin Solid Films 124 (1985) 117. [28] D.G. Kurth, T. Bein, J. Phys. Chem. 96 (1992) 6707. [29] A. Hozumi, O. Takai, Thin Solid Films 303 (1997) 222.

63

[30] T.G. Vargo, J.A. Gardella Jr., J.M. Calvert, M.S. Chen, Science 262 (1993) 1711. [31] P.C. Begovac, R.C. Thomson, J.L. Fisher, A. Hughson, A. Gallhagen, Eur. J. Vasc. Endovasc. Surgery 25 (2003) 432. [32] C. Devine, B. Hons, C. McCollum, J. Vasc. Surgery 33 (2001) 533. [33] C. Devine, C. McCollum, J. Vasc. Surgery 40 (2004) 924. [34] C.O. Esquivel, C.G. Bjoerck, S.E. Bergentz, D. Bergqvist, R. Larsson, S.N. Carson, P. Dougan, B. Nilsson, Surgery 95 (1984) 102. [35] R.R. Richardson Jr., J.A. Miller, W.M. Reichert, Biomaterials 14 (1993) 627. [36] E.F. Ritter, Y.B. Kim, H.P. Reischl, D. Serafin, A.M. Rudner, B. Klitzman, Surgery 122 (1997) 888. [37] G. Chen, Y. Ito, Y. Imanishi, A. Magnani, S. Lamponi, R. Barbucci, Bioconjugate Chem. 8 (1997) 730. [38] Y. Ito, J. Inorg. Biochem. 79 (2000) 77. [39] T. Matsuda, T. Sugawara, Langmuir 11 (1995) 2267. [40] T. Matsuda, T. Sugawara, Langmuir 11 (1995) 2272. [41] H. Hammerle, K. Kobuch, K. Kohler, W. Nisch, H. Sachs, M. Stelzle, Biomaterials 23 (2002) 797. [42] D.H. Szarowski, M.D. Andersen, S. Retterer, A.J. Spence, M. Isaacson, H.G. Craighead, J.N. Turner, W. Shain, Brain Res. 983 (2003) 23. [43] W. Kern, J. Electrochem. Soc. 137 (1990) 1887. [44] O. Sneh, S.M. George, J. Phys. Chem. 99 (1995) 4639. [45] L.T. Zhuravlev, Langmuir 3 (1987) 316. [46] O. Takai, A. Hozumi, N. Sugimoto, J. Non-Cryst. Solids 218 (1997) 280. [47] J.-C. Lin, W.-H. Chuang, J. Biomed. Mater. Res. 51 (2000) 413. [48] M.J. Bridgett, M.C. Davies, S.P. Denyer, P.R. Eldridge, Biomaterials 14 (1993) 184. [49] L. Cen, K.G. Neoh, E.T. Kang, Langmuir 18 (2002) 8633. [50] Y. Ikada, Biomaterials 15 (1994) 725. [51] J.B. Brzoska, I.B. Azouz, F. Rondelez, Langmuir 10 (1994) 4367. [52] M.M. Sung, G.J. Kluth, R. Maboudian, J. Vac. Sci. Technol. A: Vac. Surf. Films 17 (1999) 540. [53] T. Kawase, H. Sawada, J. Adhes. Sci. Technol. 16 (2002) 1103. [54] D. Wang, F.R. Jones, P. Denison, in: K.L. Mittal (Ed.), Silanes and Other Coupling Agents, VSP, 1992. [55] N. Ikuta, T. Hori, H. Naitoh, Y. Kera, E. Nishio, I. Abe, Compos. Interf. 1 (1993) 455.