- Email: [email protected]
Surface characterization and platelet compatibility evaluation of the binary mixed self-assembled monolayers
Journal of Colloid and Interface Science 308 (2007) 474–484 www.elsevier.com/locate/jcis
Surface characterization and platelet compatibility evaluation of the binary mixed self-assembled monolayers Meng-Yen Tsai, Yu-Ting Sun, Jui-Che Lin ∗ Department of Chemical Engineering, National Cheng Kung University, Tainan 70101, Taiwan Received 17 October 2006; accepted 8 January 2007 Available online 12 January 2007
Abstract This report describes a technique that used mixed self-assembled monolayer (SAM) as a model surface to evaluate the effect of steric hindrance on the SAM packing quality and its platelet compatibility. Two series of binary mixed SAMs were formed by mixing the bulky terminated alkanethiol (HS(CH2 )10 PO3 H2 ) with a smaller terminated one (HS(CH2 )9 CH3 and HS(CH2 )11 OH) respectively. Surface characterization results showed the hydrophilicity on these two series of mixed SAMs changed with the solution mole fraction of –PO3 H2 terminated thiol, χ PO3 H2 ,soln , and reached to a nearly constant value as χ PO3 H2 ,soln was 0.6 for PO3 H2 + CH3 SAM and 0.4 for PO3 H2 + OH SAM. This finding should be due to the gradual saturation of surface –PO3 H2 functionality on these mixed SAMs. The XPS analysis indicated the addition of the –CH3 and –OH terminated thiol could reduce the steric hindrance effect of –PO3 H2 functionality on monolayer formation and, henceforth, improve the SAM packing quality. In vitro platelet adhesion assay revealed the platelet compatibility on the PO3 H2 + OH SAMs was better than that on the PO3 H2 + CH3 and the pure –PO3 H2 ones. Moreover, the PO3 H2 + OH SAM with a low χ PO3 H2 ,soln value exhibited the least platelet activating property of these two mixed SAM systems. These findings suggested that material’s surface wettability and surface charge density should act collectively in affecting its platelet compatibility. © 2007 Elsevier Inc. All rights reserved. Keywords: Mixed self-assembled monolayers; Phosphonic acid; Surface characterization; X-ray photoelectron spectroscopy; Platelet compatibility
1. Introduction The clinical application of biomedical materials or devices, such as cardiovascular implants, catheters, extracorporeal circulation, and biosensors, is very important in modern medicine. The contact of blood with foreign surfaces could induce thrombosis and trigger various cellular responses [1]. These reactions would potentially impair the therapeutic functions of the medical devices and even lead to the patients’ death. Thrombosis, which is induced by a series of protein adsorption and platelet adhesion and activation, remains as an obstacle for a biomaterial scientist to develop a truly blood compatible biomedical material. The interfacial interactions between the biological environment and the biomaterial are mediated by the surface properties of the artificial material, such as surface hydrophilicity [2], surface chemical composition [3], surface microphase * Corresponding author. Fax: +886 6 234 4496
E-mail address: [email protected] (J.-C. Lin). 0021-9797/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2007.01.015
separation [4], and surface electrical charge [5]. Henceforth, to optimize the surface properties of the artificial biomaterial has become the major research theme in improving its blood compatibility. Self-assembled monolayer (SAM) prepared by the longchain alkanethiols on gold [Au-S(CH2 )n X, n 10] can provide a densely packed, well-defined, and highly ordered surface model with a broad range of chemical configuration [6–8]. Therefore, after the 1983 pioneering work by Nuzzo and Allara, SAM material has been recognized of great potential in improving our knowledge with regard to the biological/biomedical investigations, including protein adsorption, cell growth, blood compatibility and biosensors’ antibiofouling applications [9–21]. Among these, Holmlin et al. [11] have studied a series of single-component zwitterionic SAMs and zwitterionic mixed SAMs that resist nonspecific adsorption of protein. They found the “inert” SAM surface containing hydrophilic and electrically neutral properties could effectively resist nonspecific adsorption of protein. Recently, Rodrigues
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
et al. [13] discovered the platelet adhesion and activation decreased with increase of surface hydrophilicity on OH + CH3 mixed SAMs in spite of the different chain length between these two mixed thiols. Sperling et al. [14] indicated platelet adhesion density on hydrophobic COOH + CH3 mixed SAM was greater than that of hydrophilic COOH + OH surface. Phosphate is an essential component in biology due to its pervasive existence in organs and tissues, such as cell membranes, bones and body fluids. For example, a lipid bilayer membrane is composed of phospholipids containing various phosphate head groups. In skeletal tissues, phosphates are the key building blocks of the bone component, hydroxyapatite (HA). Based on the biomimetic rationales, the phosphate terminated SAM has been studied [15–18]. Due to the phosphate-like structure, the PO3 H2 terminated SAM and its neutral counterpart, PO3 (C2 H5 )2 SAM, have also been employed to improve the platelet compatibility in our previous work [19]. The ionic PO3 H2 surface exhibited a lower platelet reactivity than the COOH control and, in addition, the electrically neutral PO3 (C2 H5 )2 SAM presented a better platelet compatibility than its ionic counterpart, PO3 H2 . However, less packing order structure was noted on the pure SAM with bulky PO3 H2 terminal end. Surface protein adsorption is a very important issue for the biomaterial design because the composition and conformation of the preadsorbed protein layer will mediate the subsequent interactions between the adhered cells and the artificial material [1,22,23]. Although nonspecific protein adsorption is complex and not well understood, it could be caused by the charge– charge interaction, hydrophobic interaction or a combination of these two effects [24]. Therefore, surface hydrophilicity and surface charge density of the artificial substrates are the key factors for resisting nonspecific protein adsorption and, henceforth, improving the blood compatibility. The mixed SAMs produced by the binary mixtures of alkanethiols with different terminal functionalities can offer the possibility to tailor the surface wettability of SAM substrates at the molecular level [26–31]. In this study, we will utilize the mixed SAM technique to control the packing quality of the SAM containing the bulky –PO3 H2 functionality, and to manipulate the surface wettability and surface charge density of this organic thin film. Therefore, the smaller functional alkanethiols (HS(CH2 )9 CH3 and HS(CH2 )11 OH) will be mixed with the bulky terminal alkanethiol (HS(CH2 )10 PO3 H2 ) respectively at different molar ratio to prepare two series of binary mixed SAMs. The contact angle measurements and X-ray photoelectron spectroscopy (XPS) analysis will be used to characterize the surface properties of these two mixed SAM systems. The platelet compatibility on the two series of mixed SAMs will be also evaluated by in vitro platelet adhesion assay. 2. Materials and methods 2.1. Synthesis of 10-mercaptodecanylphosphonic acid The synthesis of 10-mercaptodecanylphosphonic acid, illustrated in Scheme 1, is modified from our previous syn-
475
Scheme 1. Reagents and conditions: (a) P(OC2 H5 )3 , reflux, 20 h; H2 O, 2 h. (b) TMSBr, CH2 Cl2 , RT, 18 h; H2 O, 2 h. (c) Thiourea, EtOH/H2 O, reflux, 6 h; NH4 OH, reflux, 2 h.
thesis method [19]. All of the solvents and reagents, either HPLC/Spectro or reagent grade, were used without further purification. NMR analysis (Bruker AMX-300), mass spectrometer (JEOL JMS-700) and elemental analysis (Elementar vario EL III), which are managed by Instrumentation Center, National Science Council, Taiwan, were performed during these synthesis steps. 2.1.1. Diethyl 10-bromodecanylphosphonate, 1 95.9 mL (0.56 mol) of triethylphosphite (Riedel de Haën, Germany) was gradually added via syringe into the two-neck round bottom flask containing 156.8 mL (0.7 mol) of 1,10dibromodecane (Fluka, Germany), which was attached with a reflux condenser, a bubble trap and a septum. After being well mixed for 30 min, the reaction was refluxed for 20 h at 130 ◦ C oil bath with a steady N2 flow bubbling below the surface of the solution. After the solution was cooled, 20 mL of water was added and stirred for 2 h. The two-phase mixture was then extracted with dichloromethane (CH2 Cl2 ). The CH2 Cl2 extracts were gathered, washed with water and brine, and then dried over anhydrous MgSO4 (JT Baker, USA). The filtrate was concentrated in vacuo to a crude oil and was purified by medium pressure liquid chromatography (SEPACORE™ , Büchi, Switzerland) using 1/3 ethyl acetate/n-hexane as eluent to yield clear colorless oil. TLC: Rf = 0.61 in 1/3 ethyl acetate/n-hexane. 1 H NMR (300 MHz, CDCl3 ): δ 3.97–3.84 (m, 4H), 3.21 (t, J = 6.8 Hz, 2H), 1.70–1.00 (overlapped, 24H); 13 C NMR (300 MHz, CDCl ): δ 60.96–60.88 (d, 2C), 33.44 (s), 3 32.36 (s), 30.20–29.98 (d), 28.89–28.78 (d), 28.58 (s), 28.25 (s), 27.68 (s), 26.11 (s), 24.25 (s), 21.90 (s), 16.00 (d, 2C); 31 P NMR (300 MHz, CDCl3 ): δ 31.74 (s); FAB-MS m/z (nitrobenzene alcohol matrix): 357 (M+ , 100%). 2.1.2. 10-Bromodecanyl phosphonic acid, 2 5.2 mL (39.1 mmol) of bromotrimethylsilane (Fluka, Germany) was added via syringe under N2 flow to a solution of 4.66 g (13.1 mmol) 1 in CH2 Cl2 . The mixture was stirred at room temperature for 18 h and refluxed for two additional hours. The excess reagent and solvent were removed in vacuo to yield an orange oily liquid, and it was hydrolyzed by water for 2 h. The mixture was extracted with CH2 Cl2 , and then the organic extract was washed with water and dried over an-
476
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
hydrous MgSO4 . The solvent was evaporated in vacuo to obtain the white solid, and it was recrystallized with CH2 Cl2 /nhexane to form a colorless crystal solid (2.77 g, 71% yield). 1 H NMR (300 MHz, CDCl ): δ 7.26 (s, broad, 2H), 3.43– 3 3.38 (t, J = 6.72 Hz, 2H), 1.90–1.28 (overlapped, 18H); 13 C NMR (300 MHz, CDCl3 ): δ 34.00 (s), 32.80 (s), 30.52–30.29 (d), 29.35–29.24 (d), 29.00 (s), 28.72 (s), 28.15 (s), 26.31 (s), 24.39 (s), 22.00 (s); 31 P NMR (300 MHz, CDCl3 ): δ 36.44 (s); FAB-MS m/z (nitrobenzene alcohol matrix): 301 (M+ , 76.43%). ANAL. Calc. for C10 H22 BrO3 P: 39.88% C, 7.36% H. Found: 39.53% C, 7.27% H. 2.1.3. 10-Mercaptodecanylphosphonic acid, 3 A solution of 0.90 g (3.0 mmol) of 2 in ethanol and the aqueous solution of 1.03 g (13.5 mmol) thiourea (Riedel de Haën, Germany) were added into the round bottom flask and refluxed for 6 h with stirring. After removing ethanol in vacuo, the mixture was added with 9.8 mL (252 mmol) of the concentrated ammonium hydroxide (Tedia, USA) and heated for additional 2 h in the 75 ◦ C oil bath. The mixture was acidified to pH < 3 with 6 N HCl aqueous solution after cooling, and then extracted with CH2 Cl2 . The combined organic extracts were washed with 1 N HCl aqueous solution and dried over anhydrous MgSO4 . The solvent was evaporated in vacuo to obtain a white solid, and it was recrystallized with CH2 Cl2 /n-hexane to form the white crystals (0.66 g, 87% yield). 1 H NMR (300 MHz, CDCl3 ): δ 9.7 (s, broad, 2H), 2.55– 2.48 (q, J = 7.32 Hz, 2H), 1.74–1.27 (overlapped, 19H); 13 C NMR (300 MHz, CDCl3 ): δ 34.02 (s), 30.52–30.30 (d), 29.40 (s), 29.30 (s), 29.03 (s), 28.40 (s), 26.30 (s), 24.60 (s), 24.60 (s), 22.00 (s); 31 P NMR (300 MHz, CDCl3 ): δ 36.66 (s); FAB-MS m/z (nitrobenzene alcohol matrix): 255 (MH+ , 100%). ANAL. Calc. for C10 H23 O3 PS: 47.23% C, 12.61% H. Found: 47.25% C, 12.55% H. 2.2. Preparation of gold substrates Si (111) wafers (4 inch diameter, 525 µm thick, P type, Silicon Development International Corp) were first cut into 1.5 × 5 cm2 strips and then cleaned by immersion in piranha solution (7:3 vol. ratio of concentrated sulfuric acid and 30 wt% H2 O2 ) at 90 ◦ C for 1 h. Caution: piranha solution is violently corrosive and should be handled with great care. Upon removal, these cleaned Si strips were rinsed thoroughly with deionized water and 99.99% ethanol and blown dry with a stream of nitrogen. The gold substrate was prepared by predepositing Ti (100 Å, 1 Å/s) layer onto Si strips to improve the adhesion of Au layer. The Au (2000 Å, 2 Å/s) film was then deposited with a home made thermal vacuum evaporator which is preevacuated to a base pressure of 5×10−5 Torr. The thickness of Au film was monitored with a quartz crystal oscillator. Contact angle analysis and XPS measurement indicated the Au layer prepared is uniform consistently.
2.3. Preparation of mixed SAMs The freshly-coated Au chips were then immersed into the 2 mM mixed thiol solutions with various molar fractions of –PO3 H2 terminated thiol, including 0.0, 0.2, 0.4, 0.6, 0.8, 1.0, at room temperature for 24 h. The 2 mM mixed thiol solutions were prepared by mixing the lab-synthesized –PO3 H2 terminated alkanethiol with 1-decanethiol (>95% purity, Fluka, USA) and 11-mercapto-1-undecanol (97%, Aldrich, USA) in absolute ethanol, respectively. Upon removal, these samples were treated by different cleaning procedure as follows: (1) rinse thoroughly with absolute ethanol (referred as EtOHwash); (2) wash thoroughly with deionized water after the treatment as (1) and then cleanse completely with ethanol again (called H2 O-wash) in order to determine the possible deprotonation effect of the ionic –PO3 H2 terminal ends. The mixed SAM substrates were then blown dry by pure nitrogen after various cleaning treatments. Both EtOH-wash and H2 O-wash samples were provided for contact angle measurements while only EtOH-wash one was used in the XPS analysis and platelet adhesion assay. The samples were then stored in petri dishes with Parafilm™ sealing, which were stored in a humidity-controlled container until XPS analysis and the platelet adhesion assay within 1 day. Note: All glass containers used Teflon® sealing covers were thoroughly cleaned by piranha solution and deionized water before use for SAM preparation. 2.4. Contact angle measurements The surface wetting property was determined by the sessile drop technique (Face Contact Angle Meter, Model CA-A, Kyowa Kaimenkagaku Co. Ltd., Japan) using deionized water as the probe liquid. For each sample, the experiments were repeated for three times and each measurement was performed at least four times in different spots. 2.5. X-ray photoelectron spectroscopy (XPS) The surface chemical analysis was performed by X-ray photoelectron spectroscopy technique (PHI Quantera SXM, USA) with monochromatic AlKα source (hν = 1486.6 eV) at Hsinchu Regional Instrumentation Center, managed by the National Science Council, Taiwan. All XPS spectra were obtained at a take-off angle of 45◦ for each element. Using the data analysis software Origin® (OriginLab Co., USA), these high-resolution spectra were further resolved into individual peaks by a peak-fit algorithm assuming Gaussian distribution for each peak. Quantification for surface chemical composition was done on the basis of peak areas of the C1s, O1s, P2p, and S2p multiplex. XPS measurements for each sample were carried out for at least once and the bare Au substrate and pure –PO3 H2 SAM were used as the internal control in each experiment run. Generally, the error bar range is about 10–15% if multiple XPS runs could be performed.
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
2.6. In vitro platelet adhesion experiment Since the platelet adhesion is one of the important steps during the blood coagulation on biomaterial’s surface, in vitro platelet adhesion assay on the mixed SAMs with various molar fractions of –PO3 H2 terminated thiol could offer some information about the possible effects caused by the surface packing of the SAM with a bulky terminal end, such as phosphonic acid here. The experimental procedure used here followed our previous studies [19,20]. Human platelet-rich plasma (PRP, ∼3.5 × 1010 platelets/mL), a generous gift from the local Blood Donation Center, was used. The sample was first equilibrated with Hepes–Tyrodes buffer solution for 1 h. Platelet-rich plasma was gently added into the sample well. Platelet adhesion and spreading were taken place inside an incubator with CO2 flow (5%) at 37 ◦ C for 1 h. The nonadhered platelets were then gently rinsed off with Hepes–Tyrodes buffer. The adhered platelets were then fixed in 2 vol% of glutaraldehyde in Hepes solution for 30 min. After glutaraldehyde fixation, the sample was washed sequentially with Hepes–Tyrodes buffer of 100, 75, 50, 25%, and double deionized water each for 3 min by a shaker at 50 rpm. The sample was then dehydrated with ethanol of 25, 50, 75 and 100% sequentially for 3 min each. The sample was finally dried in a CO2 critical point dryer (Hitachi HCP-2, Japan) to preserve the 3D morphologies of the adhered platelets. The sample was immediately sputter-coated with a layer of gold and maintained in a desiccator for further SEM analysis. The platelet adhesion density was counted in three different regions for each sample at ×1000 magnification. The platelet adhesion experiments were repeated for three times for each sample. Because of the slight variation of the platelet concentration in the platelet-rich plasma among the different donors, it is essential to include bare Au substrate (Au-coated Si wafer strip) as the internal control for each platelet adhesion test.
477
–PO3 H2 terminated alkanethiol respectively at different molar ratios to prepare two series of mixed SAMs (termed as PO3 H2 + CH3 and PO3 H2 + OH SAMs) for further investigations. These alkanethiols, which contain smaller terminal groups, were likely to be inserted between the bulky alkanethiolate molecules (i.e., –PO3 H2 terminated thiolate). By acting like a spacer, these alkanethiols would affect the molecular arrangement of SAMs by changing the interactions among the terminal ends. The correlation between the solution mole fraction of the –PO3 H2 terminated thiol and surface characteristics of the mixed SAM prepared will be discussed as follows. 3.2.1. Contact angle measurements Fig. 1 shows the variation of surface hydrophilicity, as measured by the static contact angle, on the PO3 H2 + CH3 SAM as the function of solution mole fraction of the –PO3 H2 terminated thiol, χ PO3 H2 ,soln . The contact angle values of each mixed SAMs were shown as follows: χ PO3 H2 ,soln = 0 (EtOH-wash, 107.6 ± 0.4◦ ; H2 O-wash, 106.4 ± 0.5◦ ); χ PO3 H2 ,soln = 0.2 (EtOH-wash, 92.4 ± 1.3◦ ; H2 O-wash, 87.2 ± 0.8◦ ); χ PO3 H2 ,soln = 0.4 (EtOH-wash, 77.0 ± 1.4◦ ; H2 O-wash, 68.8 ± 2.0◦ ); χ PO3 H2 ,soln = 0.6 (EtOH-wash, 57.4 ± 1.1◦ ; H2 O-wash, 46.6 ± 1.8◦ ); χ PO3 H2 ,soln = 0.8 (EtOH-wash, 60.6 ± 1.9◦ ; H2 Owash, 43.8 ± 2.2◦ ); χ PO3 H2 ,soln = 1.0 (EtOH-wash, 64.4 ± 3.6◦ ; H2 O-wash, 45.8 ± 0.5◦ ). For both EtOH-wash and H2 O-wash samples, the surface contact angle value gradually decreased with the increase of the χ PO3 H2 ,soln , and reached to a nearly constant value when χ PO3 H2 ,soln was 0.6. This finding could be attributed to the addition of more hydrophilic alkanethiol (i.e., –PO3 H2 thiol) into the binary thiol solution that leads to the formation of a SAM with more hydrophilic terminal ends. However, when the χ PO3 H2 ,soln was greater than 0.6, the surface was gradually saturated with the bulky hydrophilic –PO3 H2 terminal ends, and, henceforth, the wettability of the mixed SAMs reached to a nearly constant value. In addition, a lower contact
3. Results and discussion 3.1. Synthesis of 10-mercaptodecanylphosphonic acid The synthesis route for the long chain alkanethiol with phosphonic acid terminal was slightly modified from our previous study [19] because compound 2 in Scheme 1 is a key starting material for further synthesis of another alkanethiol with phospholipid-like terminal end. All of the compounds synthesized in this procedure were fairly pure with the NMR, mass spectrometer and elemental analysis measurements. 3.2. Surface characterization of binary mixed SAMs with various surface compositions Because of the bulkiness nature associated with the terminal functionality, the –PO3 H2 terminated SAM with less packing order was found in our previous investigation [19]. Therefore, with an attempt to improve the packing quality of this –PO3 H2 terminated SAM, two alkanethiols that contain a smaller terminal end, such as –CH3 and –OH, were mixed with the bulky
Fig. 1. The relationship between surface hydrophilicity, as determined by the sessile drop technique, and solution mole fraction of the acidic –PO3 H2 functionalized thiol for the PO3 H2 + CH3 mixed SAM surfaces. EtOH-wash (solid circles): samples rinsed thoroughly by absolute ethanol. H2 O-wash (open circles): samples washed thoroughly with deionized water after the EtOH-wash treatment and then cleansed completely with ethanol again.
478
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
Table 1 Surface atomic percentage of the PO3 H2 + CH3 SAM as a function of solution mole fraction of the –PO3 H2 terminated thiol χ PO3 H2 ,soln
Atom % C1s
O1s
S2p
P2p
0 0.2 0.4 0.6 0.8 1
90.6 (90.9)a 73.4 (84.7) 70.1 (79.4) 65.2 (74.6) 64.3 (70.4) 63.1 (66.7)
4.3 (–) 19.1 (5.1) 22.0 (9.5) 26.6 (13.4) 27.3 (16.9) 27.5 (20)
5.1 (9.1) 3.8 (8.5) 3.6 (7.9) 3.1 (7.5) 2.8 (7.1) 3.0 (6.7)
– 3.7 (1.7) 4.4 (3.2) 5.6 (4.5) 5.6 (5.6) 6.5 (6.7)
a The values in parentheses correspond to the theoretical values based on the
stoichiometry of the thiol compounds.
Fig. 2. The relationship between surface hydrophilicity, as determined by the sessile drop technique, and solution mole fraction of the acidic –PO3 H2 functionalized thiol for the PO3 H2 + OH mixed SAM surfaces. EtOH-wash (solid circles): samples rinsed thoroughly by absolute ethanol. H2 O-wash (open circles): samples washed thoroughly with deionized water after the EtOH-wash treatment and then cleansed completely with ethanol again.
angle value for the H2 O-wash PO3 H2 + CH3 SAM than that for EtOH-wash counterpart was noted except for χ PO3 H2 ,soln being 0 (i.e., pure CH3 -terminated SAM). This should be attributed to the deprotonation of the acidic terminal group, –PO3 H2 , after water washing. The deprotonation/protonation of the surface functionality, such as the acidic or basic terminal groups, will change its surface hydrophilicity. These ionic surfaces are generally more hydrophilic than their neutral counterparts [32]. Moreover, the difference of surface hydrophilicity between the EtOH-wash and H2 O-wash samples increased with the solution composition of the –PO3 H2 terminated thiol, suggesting the ionizability (i.e., deprotonation capability) and/or the amount of the surface –PO3 H2 terminal functionalities was gradually increased with the χ PO3 H2 ,soln , although the static contact angle value of the mixed SAM remained similar when χ PO3 H2 ,soln is greater than 0.6. In contrast to the findings noted for PO3 H2 + CH3 series, surface static contact angle on PO3 H2 + OH SAM was gradually increased with χ PO3 H2 ,soln (Fig. 2). The contact angle values for PO3 H2 + OH SAMs were presented as follows: χ PO3 H2 ,soln = 0 (EtOH-wash, 27.6 ± 1.8◦ ; H2 O-wash, 25.8 ± 3.1◦ ); χ PO3 H2 ,soln = 0.2 (EtOH-wash, 42.8 ± 1.6◦ ; H2 Owash, 32.0 ± 2.5◦ ); χ PO3 H2 ,soln = 0.4 (EtOH-wash, 70.4 ± 4.0◦ ; H2 O-wash, 52.4±1.9◦ ); χ PO3 H2 ,soln = 0.6 (EtOH-wash, 63.1± 2.2◦ ; H2 O-wash, 45.5 ± 0.6◦ ); χ PO3 H2 ,soln = 0.8 (EtOH-wash, 67.8 ± 1.0◦ ; H2 O-wash, 56.8 ± 2.7◦ ); χ PO3 H2 ,soln = 1.0 (EtOHwash, 61.6 ± 1.8◦ ; H2 O-wash, 44.8 ± 2.9◦ ). This was resulted from the incorporation of the –OH terminal end, which is more hydrophilic than –PO3 H2 . In addition, the surface hydrophilicity did not vary significantly as the χ PO3 H2 ,soln was greater than 0.4. Similar to the trends shown for PO3 H2 + CH3 SAMs (Fig. 1), the formation of “ionized” –PO3 H2 terminal functionality on the mixed SAM surfaces has resulted in a higher surface hydrophilicity on the H2 O-wash PO3 H2 + OH SAMs than the EtOH-wash counterparts. However, in contrast to the observation for the PO3 H2 + CH3 SAMs, such a hydrophilicity
difference didn’t vary significantly as the solution composition of –PO3 H2 terminated alkanethiol was greater than 0.4. This might be related to the gradual saturation in the deprotonation capability of the surface –PO3 H2 functionalities on these PO3 H2 + OH SAM surfaces. 3.2.2. X-ray photoelectron spectroscopy measurements Table 1 demonstrates the surface atomic percentage observed by XPS for the PO3 H2 + CH3 SAMs prepared at different χ PO3 H2 ,soln values. The O1s and P2p atomic percentages increased with χ PO3 H2 ,soln whereas the C1s and S2p values decreased. In addition, the atomic percentage values did not vary significantly as χ PO3 H2 ,soln was greater than 0.6. This observation is in close resemblance to the finding in static contact angle measurement, in which the contact angle value reached to a nearly constant value as χ PO3 H2 ,soln was greater than 0.6 (Fig. 1). This provides further evidence that the surface of PO3 H2 + CH3 SAM was gradually saturated with the bulky –PO3 H2 functional ends. As compared to the theoretical values, a lower S2p atomic percentage and, in contrast, a higher O1s and P2p atomic percentage values were observed. In the proposed SAM structure, oxygen and phosphorus atoms of –PO3 H2 terminal groups were resided on the topmost layer of the SAM whereas the sulfur atom was bound to the Au substrate in bottom. The inelastic scattering effect for the photoelectrons generated from the bottom sulfur atoms will lead to the reduced atomic percentage value observed. In contrast, much less inelastic scattering effect for the photoelectrons from the top oxygen and phosphorus atoms would be expected. Since the total summation of surface atomic percentage values always equals to 100, considering both factors just mentioned, a higher than theoretically predicted atomic percentage value should be expected for the oxygen and phosphorus atoms. A small amount of oxygen was noticed on the pure –CH3 terminated SAM. It was possibly due to the existence of oxygen-containing contaminants on the sample surface. In Table 2, similar findings were noted on the PO3 H2 + OH SAM. Interestingly, the surface saturation effect occurred when the χ PO3 H2 ,soln value was about 0.4. This value was lower than that for the PO3 H2 + CH3 SAMs, likely due to the hydrogen-bond interactions between these two hydrophilic functionalities. Such interactions would enhance the surface saturation with bulky –PO3 H2 terminal end. The method for quantifying the surface composition of the mixed SAMs has been proposed by Whitesides et al. [26,28].
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
479
Table 2 Surface atomic percentage of the PO3 H2 + OH SAM as a function of solution mole fraction of the –PO3 H2 terminated thiol χ PO3 H2 ,soln
Atom % C1s
O1s
S2p
P2p
0 0.2 0.4 0.6 0.8 1
86.7 (84.6)a 69.1 (80.6) 64.8 (76.8) 64.1 (73.2) 63.0 (69.9) 62.2 (66.7)
8.9 (7.7) 20.2 (10.4) 25.1 (13.1) 25.7 (15.5) 26.6 (17.8) 27.5 (20)
4.4 (7.7) 4.0 (7.5) 3.9 (7.2) 3.7 (7.1) 3.5 (6.8) 3.1 (6.7)
– 6.6 (1.5) 6.2 (2.9) 6.6 (4.2) 6.9 (5.5) 7.1 (6.7)
a The values in parentheses correspond to the theoretical values based on the
stoichiometry of the thiol compounds.
In this study, the P2p intensity was used to estimate the surface compositions for the binary mixed SAMs, PO3 H2 + CH3 and PO3 H2 + OH, because there was less likely inelastic scattering effect occurred for the P2p photoelectrons while the O1s intensity could be interfered by the oxygen-containing contaminants. Therefore, the surface composition of the –PO3 H2 functionality, χ PO3 H2 ,surf , was determined by the following equation: (χPO3 H2 ,surf )χPO3 H2 ,soln ≡
(AP2p )χPO3 H2 ,soln , (AP2p )χPO3 H2 ,soln=1
(1)
where AP2p is defined as the normalized area for P2p photoelectrons. The normalized area was obtained from the peak area of the photoelectron divided by its sensitivity factor provided by the XPS manufacturer [33]. Fig. 3 presents the relationship between the surface composition of the –PO3 H2 functionality on the binary mixed SAMs and the solution mole fraction of the thiol mixtures used. The two binary mixed SAM series, PO3 H2 +CH3 and PO3 H2 +OH, were both PO3 H2 -rich on the surface layer, which could be due to the low solubility of the –PO3 H2 terminated thiol in ethanol. In addition, the PO3 H2 + OH SAM surface was more “PO3 H2 rich” than the PO3 H2 + CH3 counterpart, and this might be related to the hydrogen-bond interactions between these two hydrophilic terminated alkanethiols during the SAM formation. Previous studies have also suggested that the composition and coverage of mixed SAMs can be influenced by intermolecular interactions between the thiols, the solvent, and the surface during assembly [25,26,34]. Bain et al. [25,26] have studied four mixed SAM systems each composed of –CH3 terminated alkanethiol and an alkanethiol with the polar functionalized end (–COOH, –CH2 OH, –CH2 Br and –CN) and found the adsorption of polar terminal thiols is disfavored. For the COOH + CH3 mixed SAM system, the preferential adsorption of the –CH3 terminated thiol was found at all mixing ratios, which reflected better solvation of the carboxylic acid in the ethanolic solution than at the surface of monolayer surface. In addition, for the OH + CH3 mixed SAM system, the adsorption of the alkanethiol with the polar OH tail group was strongly disfavored only at low mixing ratios. The possible explanation was that, at a low mixing ratio, the OH groups on the monolayer are isolated and can only form hydrogen bonds to the solvent. At higher concentration, the OH groups started to aggregate and form H bonds within monolayer as well as with the solvent.
Fig. 3. The surface composition of the –PO3 H2 on the binary mixed SAMs was plotted against the solution mole fraction of the –PO3 H2 in the ethanolic mixtures of –PO3 H2 and –CH3 terminated thiols (open triangles), and –PO3 H2 and –OH terminated thiols (solid diamonds).
Henceforth, the specific interaction, such as hydrogen bonds, between the terminal group of the alkanethiol and the solvent or that among the alkanethiols controlled the surface composition and/or structure of the mixed SAMs. In other words, the relative solubility of the alkanethiols, which composed the binary mixed SAMs, in ethanol determined their adsorption rates and, henceforth, affected the surface composition of mixed SAMs as those observed in this study. Castner et al. [35] have studied the XPS S2p spectra for the monolayers of the alkanethiolate on gold. The S2p peak can be fitted into a doublet peak with S2p3/2 and S2p1/2 constituents in a 2:1 intensity ratio and a 1.2-eV splitting. For the PO3 H2 + CH3 SAM, Fig. 4 shows the S2p spectra were deconvoluted into two constituent doublet peaks representing the bound thiolate and unbound thiol. As mentioned by the previous publication [35–37], the S2p3/2 peak of the bound thiolate was centered at 162 eV while that of the unbound thiol was at 163.9 eV. In addition, the intensity ratio of the S2p3/2 and S2p1/2
Fig. 4. The XPS S2p spectra of the PO3 H2 + CH3 SAMs prepared by various solution mole fraction of the –PO3 H2 terminated thiol.
480
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
Fig. 5. The XPS S2p spectra of the PO3 H2 + OH SAMs prepared by various solution mole fraction of the –PO3 H2 terminated thiol.
peak is almost 2:1 in both cases. With these constraints, it was noted the peak area percentage of the unbound thiol increased from 19.2% for χ PO3 H2 ,soln = 0 to 29.5, 32.4, 37.2, 43.6, and 50.5% for χ PO3 H2 ,soln = 0.2, 0.4, 0.6, 0.8, and 1.0, respectively. This finding should be caused by the steric hindrance effect of the bulky –PO3 H2 terminal groups. As compared to the results shown by Castner et al. [35], the higher unbound thiol percentage noted for pure –CH3 SAM (χ PO3 H2 ,soln = 0) here is likely due to the higher alkanethiol concentration used for SAM formation (2 mM vs 1 mM) as well as the less effective ethanol washing scheme used. Fig. 5 shows similar results for the PO3 H2 + OH SAM. The percentage of unbound thiol increased from 35.0% for χ PO3 H2 ,soln = 0 to 36.2, 36.9, 47.0, 40.1, and 51.9% for χ PO3 H2 ,soln = 0.2, 0.4, 0.6, 0.8, and 1.0, respectively. Based on these findings, the addition of an alkanethiol containing a smaller functional group, such as –CH3 and –OH in this study, could improve the packing quality of the bulky –PO3 H2 terminated SAM as we proposed earlier. 3.3. In vitro platelet adhesion studies The scanning electron micrographs in Fig. 6 are the typical platelet morphology adherent on the Au control and the binary mixed SAMs, PO3 H2 + CH3 and PO3 H2 + OH, prepared at different χ PO3 H2 ,soln values. The morphology of the adherent platelets will be categorized according to the method described by Ko et al. [38]. For the PO3 H2 + CH3 SAMs, Fig. 7 has shown the platelet adhesion density increased in the following order: χ PO3 H2 ,soln = 0 (32 ± 0.6 platelets/1000 µm2 ) < Au control (43 ± 4.9 platelets/1000 µm2 ) ∼ χ PO3 H2 ,soln = 0.4 (44 ± 9.2 platelets/1000 µm2 ) ∼ χ PO3 H2 ,soln = 1 (47 ± 1.7 platelets/ 1000 µm2 ) ∼ χ PO3 H2 ,soln = 0.8 (49 ± 1.0 platelets/1000 µm2 ) ∼ χ PO3 H2 ,soln = 0.6 (51 ± 1.2 platelets/1000 µm2 ) < χ PO3 H2 ,soln = 0.2 (54 ± 1.7 platelets/1000 µm2 ). The adherent platelets on Au control were most activated and reached to spreading and fully spread forms. Some of them were aggregated and overlapped with each other. The lowest amount of ad-
herent platelets was noticed on the sample with the χ PO3 H2 ,soln value being 0 (i.e., pure CH3 -terminated SAM) and the platelet morphology was mostly classified as a less activating form as spread-dendritic to spreading as compared to the fully spread shape found during the platelet aggregation, such as that on the Au surface. For the binary mixed SAMs, the platelet adhesion density was higher than the pure CH3 -terminated SAM but, however, the adhered platelets were further less activated with a round to dendritic form. Moreover, when the χ PO3 H2 ,soln value was 0.2, a highest platelet adhesion density and mostly dendritic form for the adherent platelets were found. Henceforth, considering both the adhesion density and the morphological variation of the adhered platelets, the platelet reactivity was alleviated with the increase in χ PO3 H2 ,soln . Fig. 8 indicated that the platelet adhesion density on the PO3 H2 + OH mixed SAMs increased in the following order: χ PO3 H2 ,soln = 0 (8 ± 1.0 platelets/1000 µm2 ) < χ PO3 H2 ,soln = 0.4 (11 ± 1.0 platelets/1000 µm2 ) < χ PO3 H2 ,soln = 0.2 (19 ± 5.5 platelets/1000 µm2 ) < χ PO3 H2 ,soln = 0.6 (30 ± 4.4 platelets/1000 µm2 ) < χ PO3 H2 ,soln = 0.8 (37 ± 1.2 platelets/ 1000 µm2 ) < Au-control (43 ± 4.9 platelets/1000 µm2 ) ∼ χ PO3 H2 ,soln = 1 (47 ± 1.7 platelets/1000 µm2 ). Round to dendritic shaped platelets were adhered onto the pure OHterminated SAM (i.e., the χ PO3 H2 ,soln = 0, Fig. 6) in which the lowest platelet adhesion density was noted. In comparison to the pure OH-terminated SAM, a significantly higher amount of adherent platelets but with a similar morphology was noted on these binary mixed SAM surfaces (P < 0.005). The mixed SAMs with a low χ PO3 H2 ,soln value, such as 0.2 and 0.4, showed a significantly less amount of adherent platelets than their high χ PO3 H2 ,soln -value counterparts, such as 0.6 and 0.8 (P < 0.05). Moreover, the surface with the χ PO3 H2 ,soln value of 0.4 revealed the lowest platelet adhesion density among these binary mixed SAMs. In contrast to the observation for the PO3 H2 + CH3 SAMs, the platelet compatibility was deteriorated with the increase of the χ PO3 H2 ,soln value. This might be related to various surface densities of the negatively charged –PO3 H2 functionality on these hydrophilic mixed SAMs that resulted in the different extent of protein adsorptions prior to the platelet adhesion. Surface preadsorbed plasma protein has been considered as the important factors to affect the material’s thrombogenicity [1,39]. Adsorbed fibrinogen plays a major role in mediating platelet aggregation and adhesion via direct interactions with the receptor GPIIb/IIIa on the platelet cell membrane [40–42]. For examples, a low plasma protein deposition was noticed on the platelet compatible –OH terminal SAM while the platelet-binding protein fibrinogen was found on the more platelet-activating CH3 -terminated SAM [43]. Several works [24,44–47] have been done to measure the interaction between the functionalized SAM surfaces and proteins by using the atomic force microscopy and the surface plasmon resonance (SPR) spectroscopy. By both measuring techniques, it was discovered that proteins were less likely to interact with a hydrophilic surface. Moreover, Holmlin et al. [11] also proposed the protein-resistant surfaces contain four features: (1) hydrophilic; (2) electrically neutral; (3) hy-
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
481
Fig. 6. Representative scanning electron micrographs of platelets adhered on (a) Au control; on the PO3 H2 + CH3 SAMs prepared by various solution mole fraction of –PO3 H2 terminated thiol (χ PO3 H2 ,soln. ): (b) 0, (c) 0.2, (d) 0.4, (e) 0.6, and (f) 0.8; on the PO3 H2 + OH SAMs prepared by various solution mole fraction of –PO3 H2 terminated thiol (χ PO3 H2 ,soln. ): (g) 0, (h) 0.2, (i) 0.4, (j) 0.6, (k) 0.8; and on the (l) pure –PO3 H2 terminated SAM.
482
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
Fig. 6. (continued)
drogen bond acceptor; (4) not hydrogen bond donor. In this study, the PO3 H2 + OH SAM with a low χ PO3 H2 ,soln value was revealed as the least platelet-reacting surface over the
two binary mixed SAMs systems evaluated. This result implicated the likely protein-resistant, hydrophilic and charge neutral –OH functionality should play an important role in re-
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
Fig. 7. The platelet adhesion density on Au control and the PO3 H2 + CH3 SAMs prepared by various solution mole fraction of –PO3 H2 terminated thiol. *P < 0.05 as compared to χ PO3 H2 ,soln = 0.2.
483
and this value was depended on the choice of alkanethiol with the smaller terminal functionality (i.e., –CH3 or –OH terminated thiol). XPS analysis has shown the –PO3 H2 terminated thiol preferentially adsorbed on the surface of both series of mixed SAM. More “PO3 H2 -enrichment” was noted on the PO3 H2 + OH SAM, possibly due to the hydrogen-bond interactions between the –PO3 H2 and –OH functional ends. These findings concluded that the chemical characteristics of these smaller functional groups of the alkanethiol added would affect the interfacial properties of the mixed SAMs through varying the interaction among the alkanethiols in the ethanolic solution as well as that between the terminal group of the alkanethiol and the solvent molecule. Moreover, XPS S2p spectra revealed that the addition of alkanethiol with smaller functionality could alleviate the steric hindrance effect of the bulky –PO3 H2 ends on the monolayer formation. Henceforth, the packing quality of these mixed SAMs was better than that of the pure –PO3 H2 terminated one. In vitro platelet adhesion assay has shown the platelet compatibility on the PO3 H2 + OH SAMs was better than that on the PO3 H2 + CH3 and the pure –PO3 H2 ones. Moreover, the PO3 H2 + OH SAM with a low χ PO3 H2 ,soln value was indicated as the least platelet activating surface among the binary mixed SAMs studied. These findings suggested that the biomaterial’s platelet compatibility is significantly influenced by the collaborative effect of surface wettability and surface charge density. In addition, these findings would be helpful for future design of a platelet compatible, nonpassivated and antibiofouling biomedical device, such as implantable biosensor. Acknowledgments
Fig. 8. The platelet adhesion density on Au-control and the PO3 H2 + OH SAMs prepared by various solution mole fraction of –PO3 H2 terminated thiol. *P < 0.05 as compared to χ PO3 H2 ,soln = 0; # P < 0.05 as compared to χ PO3 H2 ,soln = 0.6.
ducing platelet activation. In contrast, the adherent platelets will be activated on the less protein-resistant, less hydrophilic and anionic –PO3 H2 terminal ends, and even more be activated on the more fibrinogen-adhered, hydrophobic and neutral –CH3 terminated surface. This finding provides further evidence that the material’s platelet compatibility is significantly influenced by both surface wettability and surface charge density. 4. Summary Two series of binary mixed SAMs have been prepared by mixing the bulky lab-synthesized –PO3 H2 terminated alkanethiol with the smaller –CH3 and –OH terminated ones respectively. Surface characterization indicated the surface of these mixed SAMs was gradually saturated with the –PO3 H2 functionality with the increase of χ PO3 H2 ,soln , and finally reached to a constant surface property at distinctive χ PO3 H2 ,soln value,
The financial support from the National Science Council, Taiwan under Grant NSC 93-2214-E-006-020 and NSC 942214-E-006-004 is gratefully appreciated. We also thank the Tainan Blood Donation Center for providing human plateletrich plasma. References [1] A.S. Hoffman, in: S.L. Cooper, N.A. Peppas (Eds.), Biomaterials: Interfacial Phenomena and Applications, in: Advances in Chemistry Series, vol. 199, American Chemical Society, Washington, DC, 1982, p. 3. [2] Y.-S. Lin, V. Hlady, J. Janavota, Biomaterials 13 (1992) 497. [3] P.W. Wojciechowski, J.L. Brash, Colloids Surf. B 1 (1993) 107. [4] S.L. Goodman, S.R. Simmons, S.L. Cooper, R.M. Albrecht, J. Colloid Interface Sci. 139 (1990) 561. [5] J.H. Silver, J.-C. Lin, F. Lim, V.A. Tegoulia, M.K. Chaudhury, S.L. Cooper, Biomaterials 20 (1999) 1533. [6] A. Ulman, Chem. Rev. 96 (1996) 1533. [7] A. Ulman, An Introduction to Ultrathin Organic Films: From Langmuir– Blodgett to Self-Assembly, Academic Press, San Diego, 1991. [8] J.C. Love, L.A. Estroff, J.K. Kriebel, R.G. Nuzzo, G.M. Whitesides, Chem. Rev. 105 (2005) 1103. [9] M. Mrksich, Chem. Soc. Rev. 29 (2000) 267. [10] Y.-Y. Luk, M. Kato, M. Mrksich, Langmuir 16 (2000) 9604.
484
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
[11] R.E. Holmlin, X. Chen, R.G. Chapman, S. Takayama, G.M. Whitesides, Langmuir 17 (2001) 2841. [12] M.C.L. Martins, B.D. Ratner, M.A. Barbosa, J. Biomed. Mater. Res. 67A (2003) 158. [13] S.N. Rodrigues, I.C. Gonçalves, M.C.L. Martins, M.A. Barbosa, B.D. Ratner, Biomaterials 27 (2006) 5357. [14] C. Sperling, R.B. Schweiss, U. Streller, C. Werner, Biomaterials 26 (2005) 6547. [15] J. Zhang, J. Kirkham, C. Robinson, M.L. Wallwork, D.A. Smith, A. Marsh, M. Wong, Anal. Chem. 72 (2000) 1973. [16] A. Borgh, J. Ekeroth, R.M. Petoral, K. Uvdal, P. Konradsson, B. Liedberg, J. Colloid Interface Sci. 295 (2006) 41. [17] M. Östblom, J. Ekeroth, P. Konradsson, B. Liedberg, J. Phys. Chem. B 110 (2006) 1695. [18] M. Tanahashi, T. Matsuda, J. Biomed. Mater. Res. 34 (1997) 305. [19] M.-Y. Tsai, J.-C. Lin, J. Biomed. Mater. Res. 55 (2001) 554. [20] J.-C. Lin, W.-H. Chuang, J. Biomed. Mater. Res. 51 (2000) 413. [21] R.L. Letsinger, R. Elghanian, G. Viswanadham, C.A. Mirkin, Bioconjugate Chem. 11 (2000) 289. [22] J.R. Capadona, D.M. Collard, A.J. García, Langmuir 19 (2003) 1847. [23] B.G. Keselowsky, D.M. Collard, A.J. García, J. Biomed. Mater. Res. 66 (2003) 247. [24] G.B. Sigal, M. Mrksich, G.M. Whitesides, J. Am. Chem. Soc. 120 (1998) 3464. [25] C.D. Bain, G.M. Whitesides, J. Am. Chem. Soc. 110 (1988) 6560. [26] C.D. Bain, J. Evall, G.M. Whitesides, J. Am. Chem. Soc. 111 (1989) 7155. [27] S. Imabayashi, N. Gon, T. Sasaki, D. Hobara, T. Kakiuchi, Langmuir 14 (1998) 2348. [28] J.P. Folkers, P.E. Laibinis, G.M. Whitesides, Langmuir 8 (1992) 1330.
[29] K.E. Nelson, L. Gamble, L.S. Jung, M.S. Boeckl, E. Naeemi, S.L. Golledge, T. Sasaki, D.G. Castner, C.T. Campbell, P.S. Stayton, Langmuir 17 (2001) 2807. [30] F. Frederix, K. Bonroy, W. Laureyn, G. Reekmans, A. Campitelli, W. Dehaen, G. Maes, Langmuir 19 (2003) 4351. [31] E. Ostuni, B.A. Grzybowski, M. Mrksich, C.S. Roberts, G.M. Whitesides, Langmuir 19 (2003) 1861. [32] V. Chechik, R.M. Crooks, C.J.M. Stirling, Adv. Mater. 12 (2000) 1161. [33] J.D. Andrade, Surface Chemistry and Physics, vol. 1, Plenum, New York, 1985. [34] D.L. Allara, Biosens. Bioelectron. 10 (1995) 771. [35] D.G. Castner, K. Hinds, D.W. Grainger, Langmuir 12 (1996) 5083. [36] H. Rieley, G.K. Kendall, F.W. Zemicael, T.L. Smith, S. Yang, Langmuir 14 (1998) 5147. [37] T. Ishida, N. Choi, W. Mizutani, H. Tokumoto, I. Kojima, H. Azehara, H. Hokari, U. Akiba, M. Fujihira, Langmuir 15 (1999) 6799. [38] T.-M. Ko, J.-C. Lin, S.L. Cooper, J. Colloid Interface Sci. 156 (1993) 207. [39] T.A. Horbett, Cardiovasc. Pathol. 2 (1993) 137S. [40] J.M. Anderson, Annu. Rev. Mater. Res. 31 (2001) 81. [41] W.-B. Tsai, J.M. Grunkemeier, C.D. McFarland, T.A. Horbett, J. Biomed. Mater. Res. 60 (2002) 348. [42] W.-B. Tsai, J.M. Grunkemeier, C.D. McFarland, T.A. Horbett, J. Biomed. Mater. Res. 67A (2003) 1255. [43] M. Lestelius, B. Liedberg, P. Tengvall, Langmuir 13 (1997) 5900. [44] E. Ostuni, R.G. Chapman, R.E. Holmlin, S. Takayama, G.M. Whitesides, Langmuir 17 (2001) 5605. [45] S. Kidoaki, T. Matsuda, Langmuir 15 (1999) 7639. [46] A. Sethuraman, M. Han, R.S. Kane, G. Belfort, Langmuir 20 (2004) 7779. [47] L. Li, S. Chen, J. Zheng, B.D. Ratner, S. Jiang, J. Phys. Chem. B 109 (2005) 2934.
Surface characterization and platelet compatibility evaluation of the binary mixed self-assembled monolayers Meng-Yen Tsai, Yu-Ting Sun, Jui-Che Lin ∗ Department of Chemical Engineering, National Cheng Kung University, Tainan 70101, Taiwan Received 17 October 2006; accepted 8 January 2007 Available online 12 January 2007
Abstract This report describes a technique that used mixed self-assembled monolayer (SAM) as a model surface to evaluate the effect of steric hindrance on the SAM packing quality and its platelet compatibility. Two series of binary mixed SAMs were formed by mixing the bulky terminated alkanethiol (HS(CH2 )10 PO3 H2 ) with a smaller terminated one (HS(CH2 )9 CH3 and HS(CH2 )11 OH) respectively. Surface characterization results showed the hydrophilicity on these two series of mixed SAMs changed with the solution mole fraction of –PO3 H2 terminated thiol, χ PO3 H2 ,soln , and reached to a nearly constant value as χ PO3 H2 ,soln was 0.6 for PO3 H2 + CH3 SAM and 0.4 for PO3 H2 + OH SAM. This finding should be due to the gradual saturation of surface –PO3 H2 functionality on these mixed SAMs. The XPS analysis indicated the addition of the –CH3 and –OH terminated thiol could reduce the steric hindrance effect of –PO3 H2 functionality on monolayer formation and, henceforth, improve the SAM packing quality. In vitro platelet adhesion assay revealed the platelet compatibility on the PO3 H2 + OH SAMs was better than that on the PO3 H2 + CH3 and the pure –PO3 H2 ones. Moreover, the PO3 H2 + OH SAM with a low χ PO3 H2 ,soln value exhibited the least platelet activating property of these two mixed SAM systems. These findings suggested that material’s surface wettability and surface charge density should act collectively in affecting its platelet compatibility. © 2007 Elsevier Inc. All rights reserved. Keywords: Mixed self-assembled monolayers; Phosphonic acid; Surface characterization; X-ray photoelectron spectroscopy; Platelet compatibility
1. Introduction The clinical application of biomedical materials or devices, such as cardiovascular implants, catheters, extracorporeal circulation, and biosensors, is very important in modern medicine. The contact of blood with foreign surfaces could induce thrombosis and trigger various cellular responses [1]. These reactions would potentially impair the therapeutic functions of the medical devices and even lead to the patients’ death. Thrombosis, which is induced by a series of protein adsorption and platelet adhesion and activation, remains as an obstacle for a biomaterial scientist to develop a truly blood compatible biomedical material. The interfacial interactions between the biological environment and the biomaterial are mediated by the surface properties of the artificial material, such as surface hydrophilicity [2], surface chemical composition [3], surface microphase * Corresponding author. Fax: +886 6 234 4496
E-mail address: [email protected] (J.-C. Lin). 0021-9797/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2007.01.015
separation [4], and surface electrical charge [5]. Henceforth, to optimize the surface properties of the artificial biomaterial has become the major research theme in improving its blood compatibility. Self-assembled monolayer (SAM) prepared by the longchain alkanethiols on gold [Au-S(CH2 )n X, n 10] can provide a densely packed, well-defined, and highly ordered surface model with a broad range of chemical configuration [6–8]. Therefore, after the 1983 pioneering work by Nuzzo and Allara, SAM material has been recognized of great potential in improving our knowledge with regard to the biological/biomedical investigations, including protein adsorption, cell growth, blood compatibility and biosensors’ antibiofouling applications [9–21]. Among these, Holmlin et al. [11] have studied a series of single-component zwitterionic SAMs and zwitterionic mixed SAMs that resist nonspecific adsorption of protein. They found the “inert” SAM surface containing hydrophilic and electrically neutral properties could effectively resist nonspecific adsorption of protein. Recently, Rodrigues
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
et al. [13] discovered the platelet adhesion and activation decreased with increase of surface hydrophilicity on OH + CH3 mixed SAMs in spite of the different chain length between these two mixed thiols. Sperling et al. [14] indicated platelet adhesion density on hydrophobic COOH + CH3 mixed SAM was greater than that of hydrophilic COOH + OH surface. Phosphate is an essential component in biology due to its pervasive existence in organs and tissues, such as cell membranes, bones and body fluids. For example, a lipid bilayer membrane is composed of phospholipids containing various phosphate head groups. In skeletal tissues, phosphates are the key building blocks of the bone component, hydroxyapatite (HA). Based on the biomimetic rationales, the phosphate terminated SAM has been studied [15–18]. Due to the phosphate-like structure, the PO3 H2 terminated SAM and its neutral counterpart, PO3 (C2 H5 )2 SAM, have also been employed to improve the platelet compatibility in our previous work [19]. The ionic PO3 H2 surface exhibited a lower platelet reactivity than the COOH control and, in addition, the electrically neutral PO3 (C2 H5 )2 SAM presented a better platelet compatibility than its ionic counterpart, PO3 H2 . However, less packing order structure was noted on the pure SAM with bulky PO3 H2 terminal end. Surface protein adsorption is a very important issue for the biomaterial design because the composition and conformation of the preadsorbed protein layer will mediate the subsequent interactions between the adhered cells and the artificial material [1,22,23]. Although nonspecific protein adsorption is complex and not well understood, it could be caused by the charge– charge interaction, hydrophobic interaction or a combination of these two effects [24]. Therefore, surface hydrophilicity and surface charge density of the artificial substrates are the key factors for resisting nonspecific protein adsorption and, henceforth, improving the blood compatibility. The mixed SAMs produced by the binary mixtures of alkanethiols with different terminal functionalities can offer the possibility to tailor the surface wettability of SAM substrates at the molecular level [26–31]. In this study, we will utilize the mixed SAM technique to control the packing quality of the SAM containing the bulky –PO3 H2 functionality, and to manipulate the surface wettability and surface charge density of this organic thin film. Therefore, the smaller functional alkanethiols (HS(CH2 )9 CH3 and HS(CH2 )11 OH) will be mixed with the bulky terminal alkanethiol (HS(CH2 )10 PO3 H2 ) respectively at different molar ratio to prepare two series of binary mixed SAMs. The contact angle measurements and X-ray photoelectron spectroscopy (XPS) analysis will be used to characterize the surface properties of these two mixed SAM systems. The platelet compatibility on the two series of mixed SAMs will be also evaluated by in vitro platelet adhesion assay. 2. Materials and methods 2.1. Synthesis of 10-mercaptodecanylphosphonic acid The synthesis of 10-mercaptodecanylphosphonic acid, illustrated in Scheme 1, is modified from our previous syn-
475
Scheme 1. Reagents and conditions: (a) P(OC2 H5 )3 , reflux, 20 h; H2 O, 2 h. (b) TMSBr, CH2 Cl2 , RT, 18 h; H2 O, 2 h. (c) Thiourea, EtOH/H2 O, reflux, 6 h; NH4 OH, reflux, 2 h.
thesis method [19]. All of the solvents and reagents, either HPLC/Spectro or reagent grade, were used without further purification. NMR analysis (Bruker AMX-300), mass spectrometer (JEOL JMS-700) and elemental analysis (Elementar vario EL III), which are managed by Instrumentation Center, National Science Council, Taiwan, were performed during these synthesis steps. 2.1.1. Diethyl 10-bromodecanylphosphonate, 1 95.9 mL (0.56 mol) of triethylphosphite (Riedel de Haën, Germany) was gradually added via syringe into the two-neck round bottom flask containing 156.8 mL (0.7 mol) of 1,10dibromodecane (Fluka, Germany), which was attached with a reflux condenser, a bubble trap and a septum. After being well mixed for 30 min, the reaction was refluxed for 20 h at 130 ◦ C oil bath with a steady N2 flow bubbling below the surface of the solution. After the solution was cooled, 20 mL of water was added and stirred for 2 h. The two-phase mixture was then extracted with dichloromethane (CH2 Cl2 ). The CH2 Cl2 extracts were gathered, washed with water and brine, and then dried over anhydrous MgSO4 (JT Baker, USA). The filtrate was concentrated in vacuo to a crude oil and was purified by medium pressure liquid chromatography (SEPACORE™ , Büchi, Switzerland) using 1/3 ethyl acetate/n-hexane as eluent to yield clear colorless oil. TLC: Rf = 0.61 in 1/3 ethyl acetate/n-hexane. 1 H NMR (300 MHz, CDCl3 ): δ 3.97–3.84 (m, 4H), 3.21 (t, J = 6.8 Hz, 2H), 1.70–1.00 (overlapped, 24H); 13 C NMR (300 MHz, CDCl ): δ 60.96–60.88 (d, 2C), 33.44 (s), 3 32.36 (s), 30.20–29.98 (d), 28.89–28.78 (d), 28.58 (s), 28.25 (s), 27.68 (s), 26.11 (s), 24.25 (s), 21.90 (s), 16.00 (d, 2C); 31 P NMR (300 MHz, CDCl3 ): δ 31.74 (s); FAB-MS m/z (nitrobenzene alcohol matrix): 357 (M+ , 100%). 2.1.2. 10-Bromodecanyl phosphonic acid, 2 5.2 mL (39.1 mmol) of bromotrimethylsilane (Fluka, Germany) was added via syringe under N2 flow to a solution of 4.66 g (13.1 mmol) 1 in CH2 Cl2 . The mixture was stirred at room temperature for 18 h and refluxed for two additional hours. The excess reagent and solvent were removed in vacuo to yield an orange oily liquid, and it was hydrolyzed by water for 2 h. The mixture was extracted with CH2 Cl2 , and then the organic extract was washed with water and dried over an-
476
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
hydrous MgSO4 . The solvent was evaporated in vacuo to obtain the white solid, and it was recrystallized with CH2 Cl2 /nhexane to form a colorless crystal solid (2.77 g, 71% yield). 1 H NMR (300 MHz, CDCl ): δ 7.26 (s, broad, 2H), 3.43– 3 3.38 (t, J = 6.72 Hz, 2H), 1.90–1.28 (overlapped, 18H); 13 C NMR (300 MHz, CDCl3 ): δ 34.00 (s), 32.80 (s), 30.52–30.29 (d), 29.35–29.24 (d), 29.00 (s), 28.72 (s), 28.15 (s), 26.31 (s), 24.39 (s), 22.00 (s); 31 P NMR (300 MHz, CDCl3 ): δ 36.44 (s); FAB-MS m/z (nitrobenzene alcohol matrix): 301 (M+ , 76.43%). ANAL. Calc. for C10 H22 BrO3 P: 39.88% C, 7.36% H. Found: 39.53% C, 7.27% H. 2.1.3. 10-Mercaptodecanylphosphonic acid, 3 A solution of 0.90 g (3.0 mmol) of 2 in ethanol and the aqueous solution of 1.03 g (13.5 mmol) thiourea (Riedel de Haën, Germany) were added into the round bottom flask and refluxed for 6 h with stirring. After removing ethanol in vacuo, the mixture was added with 9.8 mL (252 mmol) of the concentrated ammonium hydroxide (Tedia, USA) and heated for additional 2 h in the 75 ◦ C oil bath. The mixture was acidified to pH < 3 with 6 N HCl aqueous solution after cooling, and then extracted with CH2 Cl2 . The combined organic extracts were washed with 1 N HCl aqueous solution and dried over anhydrous MgSO4 . The solvent was evaporated in vacuo to obtain a white solid, and it was recrystallized with CH2 Cl2 /n-hexane to form the white crystals (0.66 g, 87% yield). 1 H NMR (300 MHz, CDCl3 ): δ 9.7 (s, broad, 2H), 2.55– 2.48 (q, J = 7.32 Hz, 2H), 1.74–1.27 (overlapped, 19H); 13 C NMR (300 MHz, CDCl3 ): δ 34.02 (s), 30.52–30.30 (d), 29.40 (s), 29.30 (s), 29.03 (s), 28.40 (s), 26.30 (s), 24.60 (s), 24.60 (s), 22.00 (s); 31 P NMR (300 MHz, CDCl3 ): δ 36.66 (s); FAB-MS m/z (nitrobenzene alcohol matrix): 255 (MH+ , 100%). ANAL. Calc. for C10 H23 O3 PS: 47.23% C, 12.61% H. Found: 47.25% C, 12.55% H. 2.2. Preparation of gold substrates Si (111) wafers (4 inch diameter, 525 µm thick, P type, Silicon Development International Corp) were first cut into 1.5 × 5 cm2 strips and then cleaned by immersion in piranha solution (7:3 vol. ratio of concentrated sulfuric acid and 30 wt% H2 O2 ) at 90 ◦ C for 1 h. Caution: piranha solution is violently corrosive and should be handled with great care. Upon removal, these cleaned Si strips were rinsed thoroughly with deionized water and 99.99% ethanol and blown dry with a stream of nitrogen. The gold substrate was prepared by predepositing Ti (100 Å, 1 Å/s) layer onto Si strips to improve the adhesion of Au layer. The Au (2000 Å, 2 Å/s) film was then deposited with a home made thermal vacuum evaporator which is preevacuated to a base pressure of 5×10−5 Torr. The thickness of Au film was monitored with a quartz crystal oscillator. Contact angle analysis and XPS measurement indicated the Au layer prepared is uniform consistently.
2.3. Preparation of mixed SAMs The freshly-coated Au chips were then immersed into the 2 mM mixed thiol solutions with various molar fractions of –PO3 H2 terminated thiol, including 0.0, 0.2, 0.4, 0.6, 0.8, 1.0, at room temperature for 24 h. The 2 mM mixed thiol solutions were prepared by mixing the lab-synthesized –PO3 H2 terminated alkanethiol with 1-decanethiol (>95% purity, Fluka, USA) and 11-mercapto-1-undecanol (97%, Aldrich, USA) in absolute ethanol, respectively. Upon removal, these samples were treated by different cleaning procedure as follows: (1) rinse thoroughly with absolute ethanol (referred as EtOHwash); (2) wash thoroughly with deionized water after the treatment as (1) and then cleanse completely with ethanol again (called H2 O-wash) in order to determine the possible deprotonation effect of the ionic –PO3 H2 terminal ends. The mixed SAM substrates were then blown dry by pure nitrogen after various cleaning treatments. Both EtOH-wash and H2 O-wash samples were provided for contact angle measurements while only EtOH-wash one was used in the XPS analysis and platelet adhesion assay. The samples were then stored in petri dishes with Parafilm™ sealing, which were stored in a humidity-controlled container until XPS analysis and the platelet adhesion assay within 1 day. Note: All glass containers used Teflon® sealing covers were thoroughly cleaned by piranha solution and deionized water before use for SAM preparation. 2.4. Contact angle measurements The surface wetting property was determined by the sessile drop technique (Face Contact Angle Meter, Model CA-A, Kyowa Kaimenkagaku Co. Ltd., Japan) using deionized water as the probe liquid. For each sample, the experiments were repeated for three times and each measurement was performed at least four times in different spots. 2.5. X-ray photoelectron spectroscopy (XPS) The surface chemical analysis was performed by X-ray photoelectron spectroscopy technique (PHI Quantera SXM, USA) with monochromatic AlKα source (hν = 1486.6 eV) at Hsinchu Regional Instrumentation Center, managed by the National Science Council, Taiwan. All XPS spectra were obtained at a take-off angle of 45◦ for each element. Using the data analysis software Origin® (OriginLab Co., USA), these high-resolution spectra were further resolved into individual peaks by a peak-fit algorithm assuming Gaussian distribution for each peak. Quantification for surface chemical composition was done on the basis of peak areas of the C1s, O1s, P2p, and S2p multiplex. XPS measurements for each sample were carried out for at least once and the bare Au substrate and pure –PO3 H2 SAM were used as the internal control in each experiment run. Generally, the error bar range is about 10–15% if multiple XPS runs could be performed.
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
2.6. In vitro platelet adhesion experiment Since the platelet adhesion is one of the important steps during the blood coagulation on biomaterial’s surface, in vitro platelet adhesion assay on the mixed SAMs with various molar fractions of –PO3 H2 terminated thiol could offer some information about the possible effects caused by the surface packing of the SAM with a bulky terminal end, such as phosphonic acid here. The experimental procedure used here followed our previous studies [19,20]. Human platelet-rich plasma (PRP, ∼3.5 × 1010 platelets/mL), a generous gift from the local Blood Donation Center, was used. The sample was first equilibrated with Hepes–Tyrodes buffer solution for 1 h. Platelet-rich plasma was gently added into the sample well. Platelet adhesion and spreading were taken place inside an incubator with CO2 flow (5%) at 37 ◦ C for 1 h. The nonadhered platelets were then gently rinsed off with Hepes–Tyrodes buffer. The adhered platelets were then fixed in 2 vol% of glutaraldehyde in Hepes solution for 30 min. After glutaraldehyde fixation, the sample was washed sequentially with Hepes–Tyrodes buffer of 100, 75, 50, 25%, and double deionized water each for 3 min by a shaker at 50 rpm. The sample was then dehydrated with ethanol of 25, 50, 75 and 100% sequentially for 3 min each. The sample was finally dried in a CO2 critical point dryer (Hitachi HCP-2, Japan) to preserve the 3D morphologies of the adhered platelets. The sample was immediately sputter-coated with a layer of gold and maintained in a desiccator for further SEM analysis. The platelet adhesion density was counted in three different regions for each sample at ×1000 magnification. The platelet adhesion experiments were repeated for three times for each sample. Because of the slight variation of the platelet concentration in the platelet-rich plasma among the different donors, it is essential to include bare Au substrate (Au-coated Si wafer strip) as the internal control for each platelet adhesion test.
477
–PO3 H2 terminated alkanethiol respectively at different molar ratios to prepare two series of mixed SAMs (termed as PO3 H2 + CH3 and PO3 H2 + OH SAMs) for further investigations. These alkanethiols, which contain smaller terminal groups, were likely to be inserted between the bulky alkanethiolate molecules (i.e., –PO3 H2 terminated thiolate). By acting like a spacer, these alkanethiols would affect the molecular arrangement of SAMs by changing the interactions among the terminal ends. The correlation between the solution mole fraction of the –PO3 H2 terminated thiol and surface characteristics of the mixed SAM prepared will be discussed as follows. 3.2.1. Contact angle measurements Fig. 1 shows the variation of surface hydrophilicity, as measured by the static contact angle, on the PO3 H2 + CH3 SAM as the function of solution mole fraction of the –PO3 H2 terminated thiol, χ PO3 H2 ,soln . The contact angle values of each mixed SAMs were shown as follows: χ PO3 H2 ,soln = 0 (EtOH-wash, 107.6 ± 0.4◦ ; H2 O-wash, 106.4 ± 0.5◦ ); χ PO3 H2 ,soln = 0.2 (EtOH-wash, 92.4 ± 1.3◦ ; H2 O-wash, 87.2 ± 0.8◦ ); χ PO3 H2 ,soln = 0.4 (EtOH-wash, 77.0 ± 1.4◦ ; H2 O-wash, 68.8 ± 2.0◦ ); χ PO3 H2 ,soln = 0.6 (EtOH-wash, 57.4 ± 1.1◦ ; H2 O-wash, 46.6 ± 1.8◦ ); χ PO3 H2 ,soln = 0.8 (EtOH-wash, 60.6 ± 1.9◦ ; H2 Owash, 43.8 ± 2.2◦ ); χ PO3 H2 ,soln = 1.0 (EtOH-wash, 64.4 ± 3.6◦ ; H2 O-wash, 45.8 ± 0.5◦ ). For both EtOH-wash and H2 O-wash samples, the surface contact angle value gradually decreased with the increase of the χ PO3 H2 ,soln , and reached to a nearly constant value when χ PO3 H2 ,soln was 0.6. This finding could be attributed to the addition of more hydrophilic alkanethiol (i.e., –PO3 H2 thiol) into the binary thiol solution that leads to the formation of a SAM with more hydrophilic terminal ends. However, when the χ PO3 H2 ,soln was greater than 0.6, the surface was gradually saturated with the bulky hydrophilic –PO3 H2 terminal ends, and, henceforth, the wettability of the mixed SAMs reached to a nearly constant value. In addition, a lower contact
3. Results and discussion 3.1. Synthesis of 10-mercaptodecanylphosphonic acid The synthesis route for the long chain alkanethiol with phosphonic acid terminal was slightly modified from our previous study [19] because compound 2 in Scheme 1 is a key starting material for further synthesis of another alkanethiol with phospholipid-like terminal end. All of the compounds synthesized in this procedure were fairly pure with the NMR, mass spectrometer and elemental analysis measurements. 3.2. Surface characterization of binary mixed SAMs with various surface compositions Because of the bulkiness nature associated with the terminal functionality, the –PO3 H2 terminated SAM with less packing order was found in our previous investigation [19]. Therefore, with an attempt to improve the packing quality of this –PO3 H2 terminated SAM, two alkanethiols that contain a smaller terminal end, such as –CH3 and –OH, were mixed with the bulky
Fig. 1. The relationship between surface hydrophilicity, as determined by the sessile drop technique, and solution mole fraction of the acidic –PO3 H2 functionalized thiol for the PO3 H2 + CH3 mixed SAM surfaces. EtOH-wash (solid circles): samples rinsed thoroughly by absolute ethanol. H2 O-wash (open circles): samples washed thoroughly with deionized water after the EtOH-wash treatment and then cleansed completely with ethanol again.
478
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
Table 1 Surface atomic percentage of the PO3 H2 + CH3 SAM as a function of solution mole fraction of the –PO3 H2 terminated thiol χ PO3 H2 ,soln
Atom % C1s
O1s
S2p
P2p
0 0.2 0.4 0.6 0.8 1
90.6 (90.9)a 73.4 (84.7) 70.1 (79.4) 65.2 (74.6) 64.3 (70.4) 63.1 (66.7)
4.3 (–) 19.1 (5.1) 22.0 (9.5) 26.6 (13.4) 27.3 (16.9) 27.5 (20)
5.1 (9.1) 3.8 (8.5) 3.6 (7.9) 3.1 (7.5) 2.8 (7.1) 3.0 (6.7)
– 3.7 (1.7) 4.4 (3.2) 5.6 (4.5) 5.6 (5.6) 6.5 (6.7)
a The values in parentheses correspond to the theoretical values based on the
stoichiometry of the thiol compounds.
Fig. 2. The relationship between surface hydrophilicity, as determined by the sessile drop technique, and solution mole fraction of the acidic –PO3 H2 functionalized thiol for the PO3 H2 + OH mixed SAM surfaces. EtOH-wash (solid circles): samples rinsed thoroughly by absolute ethanol. H2 O-wash (open circles): samples washed thoroughly with deionized water after the EtOH-wash treatment and then cleansed completely with ethanol again.
angle value for the H2 O-wash PO3 H2 + CH3 SAM than that for EtOH-wash counterpart was noted except for χ PO3 H2 ,soln being 0 (i.e., pure CH3 -terminated SAM). This should be attributed to the deprotonation of the acidic terminal group, –PO3 H2 , after water washing. The deprotonation/protonation of the surface functionality, such as the acidic or basic terminal groups, will change its surface hydrophilicity. These ionic surfaces are generally more hydrophilic than their neutral counterparts [32]. Moreover, the difference of surface hydrophilicity between the EtOH-wash and H2 O-wash samples increased with the solution composition of the –PO3 H2 terminated thiol, suggesting the ionizability (i.e., deprotonation capability) and/or the amount of the surface –PO3 H2 terminal functionalities was gradually increased with the χ PO3 H2 ,soln , although the static contact angle value of the mixed SAM remained similar when χ PO3 H2 ,soln is greater than 0.6. In contrast to the findings noted for PO3 H2 + CH3 series, surface static contact angle on PO3 H2 + OH SAM was gradually increased with χ PO3 H2 ,soln (Fig. 2). The contact angle values for PO3 H2 + OH SAMs were presented as follows: χ PO3 H2 ,soln = 0 (EtOH-wash, 27.6 ± 1.8◦ ; H2 O-wash, 25.8 ± 3.1◦ ); χ PO3 H2 ,soln = 0.2 (EtOH-wash, 42.8 ± 1.6◦ ; H2 Owash, 32.0 ± 2.5◦ ); χ PO3 H2 ,soln = 0.4 (EtOH-wash, 70.4 ± 4.0◦ ; H2 O-wash, 52.4±1.9◦ ); χ PO3 H2 ,soln = 0.6 (EtOH-wash, 63.1± 2.2◦ ; H2 O-wash, 45.5 ± 0.6◦ ); χ PO3 H2 ,soln = 0.8 (EtOH-wash, 67.8 ± 1.0◦ ; H2 O-wash, 56.8 ± 2.7◦ ); χ PO3 H2 ,soln = 1.0 (EtOHwash, 61.6 ± 1.8◦ ; H2 O-wash, 44.8 ± 2.9◦ ). This was resulted from the incorporation of the –OH terminal end, which is more hydrophilic than –PO3 H2 . In addition, the surface hydrophilicity did not vary significantly as the χ PO3 H2 ,soln was greater than 0.4. Similar to the trends shown for PO3 H2 + CH3 SAMs (Fig. 1), the formation of “ionized” –PO3 H2 terminal functionality on the mixed SAM surfaces has resulted in a higher surface hydrophilicity on the H2 O-wash PO3 H2 + OH SAMs than the EtOH-wash counterparts. However, in contrast to the observation for the PO3 H2 + CH3 SAMs, such a hydrophilicity
difference didn’t vary significantly as the solution composition of –PO3 H2 terminated alkanethiol was greater than 0.4. This might be related to the gradual saturation in the deprotonation capability of the surface –PO3 H2 functionalities on these PO3 H2 + OH SAM surfaces. 3.2.2. X-ray photoelectron spectroscopy measurements Table 1 demonstrates the surface atomic percentage observed by XPS for the PO3 H2 + CH3 SAMs prepared at different χ PO3 H2 ,soln values. The O1s and P2p atomic percentages increased with χ PO3 H2 ,soln whereas the C1s and S2p values decreased. In addition, the atomic percentage values did not vary significantly as χ PO3 H2 ,soln was greater than 0.6. This observation is in close resemblance to the finding in static contact angle measurement, in which the contact angle value reached to a nearly constant value as χ PO3 H2 ,soln was greater than 0.6 (Fig. 1). This provides further evidence that the surface of PO3 H2 + CH3 SAM was gradually saturated with the bulky –PO3 H2 functional ends. As compared to the theoretical values, a lower S2p atomic percentage and, in contrast, a higher O1s and P2p atomic percentage values were observed. In the proposed SAM structure, oxygen and phosphorus atoms of –PO3 H2 terminal groups were resided on the topmost layer of the SAM whereas the sulfur atom was bound to the Au substrate in bottom. The inelastic scattering effect for the photoelectrons generated from the bottom sulfur atoms will lead to the reduced atomic percentage value observed. In contrast, much less inelastic scattering effect for the photoelectrons from the top oxygen and phosphorus atoms would be expected. Since the total summation of surface atomic percentage values always equals to 100, considering both factors just mentioned, a higher than theoretically predicted atomic percentage value should be expected for the oxygen and phosphorus atoms. A small amount of oxygen was noticed on the pure –CH3 terminated SAM. It was possibly due to the existence of oxygen-containing contaminants on the sample surface. In Table 2, similar findings were noted on the PO3 H2 + OH SAM. Interestingly, the surface saturation effect occurred when the χ PO3 H2 ,soln value was about 0.4. This value was lower than that for the PO3 H2 + CH3 SAMs, likely due to the hydrogen-bond interactions between these two hydrophilic functionalities. Such interactions would enhance the surface saturation with bulky –PO3 H2 terminal end. The method for quantifying the surface composition of the mixed SAMs has been proposed by Whitesides et al. [26,28].
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
479
Table 2 Surface atomic percentage of the PO3 H2 + OH SAM as a function of solution mole fraction of the –PO3 H2 terminated thiol χ PO3 H2 ,soln
Atom % C1s
O1s
S2p
P2p
0 0.2 0.4 0.6 0.8 1
86.7 (84.6)a 69.1 (80.6) 64.8 (76.8) 64.1 (73.2) 63.0 (69.9) 62.2 (66.7)
8.9 (7.7) 20.2 (10.4) 25.1 (13.1) 25.7 (15.5) 26.6 (17.8) 27.5 (20)
4.4 (7.7) 4.0 (7.5) 3.9 (7.2) 3.7 (7.1) 3.5 (6.8) 3.1 (6.7)
– 6.6 (1.5) 6.2 (2.9) 6.6 (4.2) 6.9 (5.5) 7.1 (6.7)
a The values in parentheses correspond to the theoretical values based on the
stoichiometry of the thiol compounds.
In this study, the P2p intensity was used to estimate the surface compositions for the binary mixed SAMs, PO3 H2 + CH3 and PO3 H2 + OH, because there was less likely inelastic scattering effect occurred for the P2p photoelectrons while the O1s intensity could be interfered by the oxygen-containing contaminants. Therefore, the surface composition of the –PO3 H2 functionality, χ PO3 H2 ,surf , was determined by the following equation: (χPO3 H2 ,surf )χPO3 H2 ,soln ≡
(AP2p )χPO3 H2 ,soln , (AP2p )χPO3 H2 ,soln=1
(1)
where AP2p is defined as the normalized area for P2p photoelectrons. The normalized area was obtained from the peak area of the photoelectron divided by its sensitivity factor provided by the XPS manufacturer [33]. Fig. 3 presents the relationship between the surface composition of the –PO3 H2 functionality on the binary mixed SAMs and the solution mole fraction of the thiol mixtures used. The two binary mixed SAM series, PO3 H2 +CH3 and PO3 H2 +OH, were both PO3 H2 -rich on the surface layer, which could be due to the low solubility of the –PO3 H2 terminated thiol in ethanol. In addition, the PO3 H2 + OH SAM surface was more “PO3 H2 rich” than the PO3 H2 + CH3 counterpart, and this might be related to the hydrogen-bond interactions between these two hydrophilic terminated alkanethiols during the SAM formation. Previous studies have also suggested that the composition and coverage of mixed SAMs can be influenced by intermolecular interactions between the thiols, the solvent, and the surface during assembly [25,26,34]. Bain et al. [25,26] have studied four mixed SAM systems each composed of –CH3 terminated alkanethiol and an alkanethiol with the polar functionalized end (–COOH, –CH2 OH, –CH2 Br and –CN) and found the adsorption of polar terminal thiols is disfavored. For the COOH + CH3 mixed SAM system, the preferential adsorption of the –CH3 terminated thiol was found at all mixing ratios, which reflected better solvation of the carboxylic acid in the ethanolic solution than at the surface of monolayer surface. In addition, for the OH + CH3 mixed SAM system, the adsorption of the alkanethiol with the polar OH tail group was strongly disfavored only at low mixing ratios. The possible explanation was that, at a low mixing ratio, the OH groups on the monolayer are isolated and can only form hydrogen bonds to the solvent. At higher concentration, the OH groups started to aggregate and form H bonds within monolayer as well as with the solvent.
Fig. 3. The surface composition of the –PO3 H2 on the binary mixed SAMs was plotted against the solution mole fraction of the –PO3 H2 in the ethanolic mixtures of –PO3 H2 and –CH3 terminated thiols (open triangles), and –PO3 H2 and –OH terminated thiols (solid diamonds).
Henceforth, the specific interaction, such as hydrogen bonds, between the terminal group of the alkanethiol and the solvent or that among the alkanethiols controlled the surface composition and/or structure of the mixed SAMs. In other words, the relative solubility of the alkanethiols, which composed the binary mixed SAMs, in ethanol determined their adsorption rates and, henceforth, affected the surface composition of mixed SAMs as those observed in this study. Castner et al. [35] have studied the XPS S2p spectra for the monolayers of the alkanethiolate on gold. The S2p peak can be fitted into a doublet peak with S2p3/2 and S2p1/2 constituents in a 2:1 intensity ratio and a 1.2-eV splitting. For the PO3 H2 + CH3 SAM, Fig. 4 shows the S2p spectra were deconvoluted into two constituent doublet peaks representing the bound thiolate and unbound thiol. As mentioned by the previous publication [35–37], the S2p3/2 peak of the bound thiolate was centered at 162 eV while that of the unbound thiol was at 163.9 eV. In addition, the intensity ratio of the S2p3/2 and S2p1/2
Fig. 4. The XPS S2p spectra of the PO3 H2 + CH3 SAMs prepared by various solution mole fraction of the –PO3 H2 terminated thiol.
480
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
Fig. 5. The XPS S2p spectra of the PO3 H2 + OH SAMs prepared by various solution mole fraction of the –PO3 H2 terminated thiol.
peak is almost 2:1 in both cases. With these constraints, it was noted the peak area percentage of the unbound thiol increased from 19.2% for χ PO3 H2 ,soln = 0 to 29.5, 32.4, 37.2, 43.6, and 50.5% for χ PO3 H2 ,soln = 0.2, 0.4, 0.6, 0.8, and 1.0, respectively. This finding should be caused by the steric hindrance effect of the bulky –PO3 H2 terminal groups. As compared to the results shown by Castner et al. [35], the higher unbound thiol percentage noted for pure –CH3 SAM (χ PO3 H2 ,soln = 0) here is likely due to the higher alkanethiol concentration used for SAM formation (2 mM vs 1 mM) as well as the less effective ethanol washing scheme used. Fig. 5 shows similar results for the PO3 H2 + OH SAM. The percentage of unbound thiol increased from 35.0% for χ PO3 H2 ,soln = 0 to 36.2, 36.9, 47.0, 40.1, and 51.9% for χ PO3 H2 ,soln = 0.2, 0.4, 0.6, 0.8, and 1.0, respectively. Based on these findings, the addition of an alkanethiol containing a smaller functional group, such as –CH3 and –OH in this study, could improve the packing quality of the bulky –PO3 H2 terminated SAM as we proposed earlier. 3.3. In vitro platelet adhesion studies The scanning electron micrographs in Fig. 6 are the typical platelet morphology adherent on the Au control and the binary mixed SAMs, PO3 H2 + CH3 and PO3 H2 + OH, prepared at different χ PO3 H2 ,soln values. The morphology of the adherent platelets will be categorized according to the method described by Ko et al. [38]. For the PO3 H2 + CH3 SAMs, Fig. 7 has shown the platelet adhesion density increased in the following order: χ PO3 H2 ,soln = 0 (32 ± 0.6 platelets/1000 µm2 ) < Au control (43 ± 4.9 platelets/1000 µm2 ) ∼ χ PO3 H2 ,soln = 0.4 (44 ± 9.2 platelets/1000 µm2 ) ∼ χ PO3 H2 ,soln = 1 (47 ± 1.7 platelets/ 1000 µm2 ) ∼ χ PO3 H2 ,soln = 0.8 (49 ± 1.0 platelets/1000 µm2 ) ∼ χ PO3 H2 ,soln = 0.6 (51 ± 1.2 platelets/1000 µm2 ) < χ PO3 H2 ,soln = 0.2 (54 ± 1.7 platelets/1000 µm2 ). The adherent platelets on Au control were most activated and reached to spreading and fully spread forms. Some of them were aggregated and overlapped with each other. The lowest amount of ad-
herent platelets was noticed on the sample with the χ PO3 H2 ,soln value being 0 (i.e., pure CH3 -terminated SAM) and the platelet morphology was mostly classified as a less activating form as spread-dendritic to spreading as compared to the fully spread shape found during the platelet aggregation, such as that on the Au surface. For the binary mixed SAMs, the platelet adhesion density was higher than the pure CH3 -terminated SAM but, however, the adhered platelets were further less activated with a round to dendritic form. Moreover, when the χ PO3 H2 ,soln value was 0.2, a highest platelet adhesion density and mostly dendritic form for the adherent platelets were found. Henceforth, considering both the adhesion density and the morphological variation of the adhered platelets, the platelet reactivity was alleviated with the increase in χ PO3 H2 ,soln . Fig. 8 indicated that the platelet adhesion density on the PO3 H2 + OH mixed SAMs increased in the following order: χ PO3 H2 ,soln = 0 (8 ± 1.0 platelets/1000 µm2 ) < χ PO3 H2 ,soln = 0.4 (11 ± 1.0 platelets/1000 µm2 ) < χ PO3 H2 ,soln = 0.2 (19 ± 5.5 platelets/1000 µm2 ) < χ PO3 H2 ,soln = 0.6 (30 ± 4.4 platelets/1000 µm2 ) < χ PO3 H2 ,soln = 0.8 (37 ± 1.2 platelets/ 1000 µm2 ) < Au-control (43 ± 4.9 platelets/1000 µm2 ) ∼ χ PO3 H2 ,soln = 1 (47 ± 1.7 platelets/1000 µm2 ). Round to dendritic shaped platelets were adhered onto the pure OHterminated SAM (i.e., the χ PO3 H2 ,soln = 0, Fig. 6) in which the lowest platelet adhesion density was noted. In comparison to the pure OH-terminated SAM, a significantly higher amount of adherent platelets but with a similar morphology was noted on these binary mixed SAM surfaces (P < 0.005). The mixed SAMs with a low χ PO3 H2 ,soln value, such as 0.2 and 0.4, showed a significantly less amount of adherent platelets than their high χ PO3 H2 ,soln -value counterparts, such as 0.6 and 0.8 (P < 0.05). Moreover, the surface with the χ PO3 H2 ,soln value of 0.4 revealed the lowest platelet adhesion density among these binary mixed SAMs. In contrast to the observation for the PO3 H2 + CH3 SAMs, the platelet compatibility was deteriorated with the increase of the χ PO3 H2 ,soln value. This might be related to various surface densities of the negatively charged –PO3 H2 functionality on these hydrophilic mixed SAMs that resulted in the different extent of protein adsorptions prior to the platelet adhesion. Surface preadsorbed plasma protein has been considered as the important factors to affect the material’s thrombogenicity [1,39]. Adsorbed fibrinogen plays a major role in mediating platelet aggregation and adhesion via direct interactions with the receptor GPIIb/IIIa on the platelet cell membrane [40–42]. For examples, a low plasma protein deposition was noticed on the platelet compatible –OH terminal SAM while the platelet-binding protein fibrinogen was found on the more platelet-activating CH3 -terminated SAM [43]. Several works [24,44–47] have been done to measure the interaction between the functionalized SAM surfaces and proteins by using the atomic force microscopy and the surface plasmon resonance (SPR) spectroscopy. By both measuring techniques, it was discovered that proteins were less likely to interact with a hydrophilic surface. Moreover, Holmlin et al. [11] also proposed the protein-resistant surfaces contain four features: (1) hydrophilic; (2) electrically neutral; (3) hy-
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
481
Fig. 6. Representative scanning electron micrographs of platelets adhered on (a) Au control; on the PO3 H2 + CH3 SAMs prepared by various solution mole fraction of –PO3 H2 terminated thiol (χ PO3 H2 ,soln. ): (b) 0, (c) 0.2, (d) 0.4, (e) 0.6, and (f) 0.8; on the PO3 H2 + OH SAMs prepared by various solution mole fraction of –PO3 H2 terminated thiol (χ PO3 H2 ,soln. ): (g) 0, (h) 0.2, (i) 0.4, (j) 0.6, (k) 0.8; and on the (l) pure –PO3 H2 terminated SAM.
482
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
Fig. 6. (continued)
drogen bond acceptor; (4) not hydrogen bond donor. In this study, the PO3 H2 + OH SAM with a low χ PO3 H2 ,soln value was revealed as the least platelet-reacting surface over the
two binary mixed SAMs systems evaluated. This result implicated the likely protein-resistant, hydrophilic and charge neutral –OH functionality should play an important role in re-
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
Fig. 7. The platelet adhesion density on Au control and the PO3 H2 + CH3 SAMs prepared by various solution mole fraction of –PO3 H2 terminated thiol. *P < 0.05 as compared to χ PO3 H2 ,soln = 0.2.
483
and this value was depended on the choice of alkanethiol with the smaller terminal functionality (i.e., –CH3 or –OH terminated thiol). XPS analysis has shown the –PO3 H2 terminated thiol preferentially adsorbed on the surface of both series of mixed SAM. More “PO3 H2 -enrichment” was noted on the PO3 H2 + OH SAM, possibly due to the hydrogen-bond interactions between the –PO3 H2 and –OH functional ends. These findings concluded that the chemical characteristics of these smaller functional groups of the alkanethiol added would affect the interfacial properties of the mixed SAMs through varying the interaction among the alkanethiols in the ethanolic solution as well as that between the terminal group of the alkanethiol and the solvent molecule. Moreover, XPS S2p spectra revealed that the addition of alkanethiol with smaller functionality could alleviate the steric hindrance effect of the bulky –PO3 H2 ends on the monolayer formation. Henceforth, the packing quality of these mixed SAMs was better than that of the pure –PO3 H2 terminated one. In vitro platelet adhesion assay has shown the platelet compatibility on the PO3 H2 + OH SAMs was better than that on the PO3 H2 + CH3 and the pure –PO3 H2 ones. Moreover, the PO3 H2 + OH SAM with a low χ PO3 H2 ,soln value was indicated as the least platelet activating surface among the binary mixed SAMs studied. These findings suggested that the biomaterial’s platelet compatibility is significantly influenced by the collaborative effect of surface wettability and surface charge density. In addition, these findings would be helpful for future design of a platelet compatible, nonpassivated and antibiofouling biomedical device, such as implantable biosensor. Acknowledgments
Fig. 8. The platelet adhesion density on Au-control and the PO3 H2 + OH SAMs prepared by various solution mole fraction of –PO3 H2 terminated thiol. *P < 0.05 as compared to χ PO3 H2 ,soln = 0; # P < 0.05 as compared to χ PO3 H2 ,soln = 0.6.
ducing platelet activation. In contrast, the adherent platelets will be activated on the less protein-resistant, less hydrophilic and anionic –PO3 H2 terminal ends, and even more be activated on the more fibrinogen-adhered, hydrophobic and neutral –CH3 terminated surface. This finding provides further evidence that the material’s platelet compatibility is significantly influenced by both surface wettability and surface charge density. 4. Summary Two series of binary mixed SAMs have been prepared by mixing the bulky lab-synthesized –PO3 H2 terminated alkanethiol with the smaller –CH3 and –OH terminated ones respectively. Surface characterization indicated the surface of these mixed SAMs was gradually saturated with the –PO3 H2 functionality with the increase of χ PO3 H2 ,soln , and finally reached to a constant surface property at distinctive χ PO3 H2 ,soln value,
The financial support from the National Science Council, Taiwan under Grant NSC 93-2214-E-006-020 and NSC 942214-E-006-004 is gratefully appreciated. We also thank the Tainan Blood Donation Center for providing human plateletrich plasma. References [1] A.S. Hoffman, in: S.L. Cooper, N.A. Peppas (Eds.), Biomaterials: Interfacial Phenomena and Applications, in: Advances in Chemistry Series, vol. 199, American Chemical Society, Washington, DC, 1982, p. 3. [2] Y.-S. Lin, V. Hlady, J. Janavota, Biomaterials 13 (1992) 497. [3] P.W. Wojciechowski, J.L. Brash, Colloids Surf. B 1 (1993) 107. [4] S.L. Goodman, S.R. Simmons, S.L. Cooper, R.M. Albrecht, J. Colloid Interface Sci. 139 (1990) 561. [5] J.H. Silver, J.-C. Lin, F. Lim, V.A. Tegoulia, M.K. Chaudhury, S.L. Cooper, Biomaterials 20 (1999) 1533. [6] A. Ulman, Chem. Rev. 96 (1996) 1533. [7] A. Ulman, An Introduction to Ultrathin Organic Films: From Langmuir– Blodgett to Self-Assembly, Academic Press, San Diego, 1991. [8] J.C. Love, L.A. Estroff, J.K. Kriebel, R.G. Nuzzo, G.M. Whitesides, Chem. Rev. 105 (2005) 1103. [9] M. Mrksich, Chem. Soc. Rev. 29 (2000) 267. [10] Y.-Y. Luk, M. Kato, M. Mrksich, Langmuir 16 (2000) 9604.
484
M.-Y. Tsai et al. / Journal of Colloid and Interface Science 308 (2007) 474–484
[11] R.E. Holmlin, X. Chen, R.G. Chapman, S. Takayama, G.M. Whitesides, Langmuir 17 (2001) 2841. [12] M.C.L. Martins, B.D. Ratner, M.A. Barbosa, J. Biomed. Mater. Res. 67A (2003) 158. [13] S.N. Rodrigues, I.C. Gonçalves, M.C.L. Martins, M.A. Barbosa, B.D. Ratner, Biomaterials 27 (2006) 5357. [14] C. Sperling, R.B. Schweiss, U. Streller, C. Werner, Biomaterials 26 (2005) 6547. [15] J. Zhang, J. Kirkham, C. Robinson, M.L. Wallwork, D.A. Smith, A. Marsh, M. Wong, Anal. Chem. 72 (2000) 1973. [16] A. Borgh, J. Ekeroth, R.M. Petoral, K. Uvdal, P. Konradsson, B. Liedberg, J. Colloid Interface Sci. 295 (2006) 41. [17] M. Östblom, J. Ekeroth, P. Konradsson, B. Liedberg, J. Phys. Chem. B 110 (2006) 1695. [18] M. Tanahashi, T. Matsuda, J. Biomed. Mater. Res. 34 (1997) 305. [19] M.-Y. Tsai, J.-C. Lin, J. Biomed. Mater. Res. 55 (2001) 554. [20] J.-C. Lin, W.-H. Chuang, J. Biomed. Mater. Res. 51 (2000) 413. [21] R.L. Letsinger, R. Elghanian, G. Viswanadham, C.A. Mirkin, Bioconjugate Chem. 11 (2000) 289. [22] J.R. Capadona, D.M. Collard, A.J. García, Langmuir 19 (2003) 1847. [23] B.G. Keselowsky, D.M. Collard, A.J. García, J. Biomed. Mater. Res. 66 (2003) 247. [24] G.B. Sigal, M. Mrksich, G.M. Whitesides, J. Am. Chem. Soc. 120 (1998) 3464. [25] C.D. Bain, G.M. Whitesides, J. Am. Chem. Soc. 110 (1988) 6560. [26] C.D. Bain, J. Evall, G.M. Whitesides, J. Am. Chem. Soc. 111 (1989) 7155. [27] S. Imabayashi, N. Gon, T. Sasaki, D. Hobara, T. Kakiuchi, Langmuir 14 (1998) 2348. [28] J.P. Folkers, P.E. Laibinis, G.M. Whitesides, Langmuir 8 (1992) 1330.
[29] K.E. Nelson, L. Gamble, L.S. Jung, M.S. Boeckl, E. Naeemi, S.L. Golledge, T. Sasaki, D.G. Castner, C.T. Campbell, P.S. Stayton, Langmuir 17 (2001) 2807. [30] F. Frederix, K. Bonroy, W. Laureyn, G. Reekmans, A. Campitelli, W. Dehaen, G. Maes, Langmuir 19 (2003) 4351. [31] E. Ostuni, B.A. Grzybowski, M. Mrksich, C.S. Roberts, G.M. Whitesides, Langmuir 19 (2003) 1861. [32] V. Chechik, R.M. Crooks, C.J.M. Stirling, Adv. Mater. 12 (2000) 1161. [33] J.D. Andrade, Surface Chemistry and Physics, vol. 1, Plenum, New York, 1985. [34] D.L. Allara, Biosens. Bioelectron. 10 (1995) 771. [35] D.G. Castner, K. Hinds, D.W. Grainger, Langmuir 12 (1996) 5083. [36] H. Rieley, G.K. Kendall, F.W. Zemicael, T.L. Smith, S. Yang, Langmuir 14 (1998) 5147. [37] T. Ishida, N. Choi, W. Mizutani, H. Tokumoto, I. Kojima, H. Azehara, H. Hokari, U. Akiba, M. Fujihira, Langmuir 15 (1999) 6799. [38] T.-M. Ko, J.-C. Lin, S.L. Cooper, J. Colloid Interface Sci. 156 (1993) 207. [39] T.A. Horbett, Cardiovasc. Pathol. 2 (1993) 137S. [40] J.M. Anderson, Annu. Rev. Mater. Res. 31 (2001) 81. [41] W.-B. Tsai, J.M. Grunkemeier, C.D. McFarland, T.A. Horbett, J. Biomed. Mater. Res. 60 (2002) 348. [42] W.-B. Tsai, J.M. Grunkemeier, C.D. McFarland, T.A. Horbett, J. Biomed. Mater. Res. 67A (2003) 1255. [43] M. Lestelius, B. Liedberg, P. Tengvall, Langmuir 13 (1997) 5900. [44] E. Ostuni, R.G. Chapman, R.E. Holmlin, S. Takayama, G.M. Whitesides, Langmuir 17 (2001) 5605. [45] S. Kidoaki, T. Matsuda, Langmuir 15 (1999) 7639. [46] A. Sethuraman, M. Han, R.S. Kane, G. Belfort, Langmuir 20 (2004) 7779. [47] L. Li, S. Chen, J. Zheng, B.D. Ratner, S. Jiang, J. Phys. Chem. B 109 (2005) 2934.