Article pubs.acs.org/Langmuir
Developing Antifouling Biointerfaces Based on Bioinspired Zwitterionic Dopamine through pH-Modulated Assembly Chun-Jen Huang,*,†,‡ Lin-Chuan Wang,† Jing-Jong Shyue,§ and Ying-Chih Chang∥ †
Graduate Institute of Biomedical Engineering and ‡Chemical & Materials Engineering Department, National Central University, Jhong-Li, Taoyuan 320, Taiwan § Research Center for Applied Sciences and ∥Genomics Research Center, Academia Sinica, 128, Sec 2 Academia Road, Taipei 115, Taiwan S Supporting Information *
ABSTRACT: The use of synthetic biomaterials as implantable devices typically is accompanied by considerable nonspecific adsorption of proteins, cells, and bacteria. These may eventually induce adverse pathogenic problems in clinical practice, such as thrombosis and biomaterial-associated infection. Thus, an effective surface coating for medical devices has been pursued to repel nonspecific adsorption from surfaces. In this study, we employ an adhesive dopamine molecule conjugated with zwitterionic sulfobetaine moiety (SB-DA), developed based on natural mussels, as a surface ligand for the modification of TiO2. The electrochemical study shows that the SB-DA exhibits fully reversible reduction−oxidation behavior at pH 3, but it is irreversible at pH 8. A contact angle goniometer and X-ray photoelectron spectroscopy were utilized to explore the surface hydration, chemical states, and bonding mechanism of SB-DA. The results indicate that the binding between hydroxyl groups of SB-DA and TiO2 converts from hydrogen bonds to bidentate binding upon the pH transition from pH 3 to 8. In order to examine the antifouling properties of SB-DA thin films, the modified substrates were brought into contact with bovine serum albumin and bacteria solutions. The fouling levels were monitored using a quartz crystal microbalance with dissipation sensor and fluorescence optical microscope. Tests showed that the sample prepared via the pH transition approach provides the best resistance to nonspecific adsorption due to the high coverage and stability of the SB-DA films. These findings support the mechanism of the pH-modulated assembly of SB-DA molecules, and for the first time we demonstrate the antifouling properties of the SB-DA to be comparable with traditional thiol-based zwitterionic self-assemblies. The success of modification with SB-DA opens an avenue for developing a biologically inspired surface chemistry and can have applications over a wide spectrum of bioapplications. The strategy of the pH transition can also be applied to other functional dopamine derivatives. ditions.3−5 An emerging class of charged materials called zwitterionic polymers, however, provides effective biofouling repellence, long-term durability, and environmental stability.6−8 Its promising achievements have stimulated the development of methodologies based on zwitterionic materials for surface engineering. Surface-initiated polymerization (“graft-from”) has been the most popular approach for preparation of robust and well-controllable surface coatings.6,7 However, lack of applicability due to reliance on specialized chemical infrastructures and technicians makes unlikely for the wide-spreading industrial implementation. Currently, approaches based on “graft-to” strategy have been used to accomplish rapid and large-area surface coatings.8−12 Self-assembled monolayers (SAMs) have attracted substantial attention since their introduction.13 SAMs are thin films spontaneously formed by the adsorption of molecular building
1. INTRODUCTION Unwanted biofouling frequently occurs on medical devices, boats, heat exchangers, and houseware, causing substantial concerns about their applicability, function, and durability. These problems are especially critical in medical devices. For example, urinary tract infections, as the second most common type of healthcare-associated infection, next to surgical site infections, cause more than 13 000 deaths annually in the US.1 Even state-of-the-art catheters still are unable to effectively prevent infection. Therefore, fundamental understanding and control of nonspecific adsorption at interfaces are essential for developing biocompatible implant devices. In general, the adsorption process is thermodynamically determined as entropy gain and enthalpy loss in a system, leading to adsorption energetic preference. Poly(ethylene glycol) (PEG) is the most commonly employed material for fouling resistance due to its characteristics of nontoxicity, high water solubility, large exclusion volume, steric hindrance, and uncharged properties.2 However, its susceptibility to thermal and oxidative degradation limits its applications under physiologic con© 2014 American Chemical Society
Received: August 10, 2014 Revised: October 2, 2014 Published: October 6, 2014 12638
dx.doi.org/10.1021/la503191b | Langmuir 2014, 30, 12638−12646
Langmuir
Article
strategy allowing the formation of a compact and stable antifouling SB-DA coating. A contact angle goniometer electrochemical apparatus and X-ray photoelectron spectroscopy (XPS) are employed to develop an understanding for the interaction of the catecholic groups with TiO2 surfaces upon the pH transition. The antifouling properties of SB-DA are accessed by challenging with protein and bacterial solutions. The fouling levels are observed by the quartz crystal microbalance with dissipation (QCM-D) and fluorescence optical microscopy. In this study, we not only demonstrate the effective antifouling property of SB-DA adlayers prepared by a simple and effective approach but also provide insight into the binding mechanism of catecholic derivatives for their full exploitation in surface modification.
blocks from solution or from the gas phase onto a solid or liquid surface. They offer a facile, versatile, and well-defied approach for tailoring the interfacial properties for a wide spectrum of applications. The usability depends on the interaction of their anchoring groups with substrates of interest. The commonly used anchoring groups include thiols, silanes, and phosphonates.14 Recently, very large assemblies conjugated with functional groups have been developed, which constitute a toolkit for manipulating surface properties at a molecular level. A mussel-inspired adhesive chemistry based on catecholic derivatives has attracted widespread interest.15 Mussels can adhere to a variety of materials with high binding strength under wet conditions due to the 3,4-dihydroxyphenylalanin (DOPA) content in their foot proteins.16,17 Dopamine (DA) is the most frequently used catecholic compound that can be directly deposited on substrates to become bioactive surfaces for cell growth, anticorrosion treatments for microtribology, functional adlayers for postconjugation, and packing materials for encapsulating drugs or bioactive molecules.15,18−21 Additionally, many synthetic catecholic compounds have been developed as novel surface chemistries, indicating a direction for new coating strategies.22−24 As an antifouling treatment, Messersmith et al. developed a DA mimetic initiator on substrates under a mild basic condition for the preparation of polymer brushes of oligo(ethylene glycol) methyl ether methacrylate (OEGMEMA).25 From the same group, Lau et al. synthesized surface-grafted polysarcosine brushes coupled to a mussel adhesive protein-inspired DOPA-Lys pentapeptide to yield a long-term antifouling layer.26 Textor et al. conjugated OEG to DOPA residues to provide a building block for the formation of antifouling coatings.23,24,27,28 Jiang’s group developed a biomimetic zwitterionic carboxybetaine polymers growing from catecholic initiators via atom transfer radical polymerization.10,29,30 These works accomplished bioinert interfaces via either the “graft-from” or “graft-to” strategies based on catecholic chemistry. Moreover, the potency of the adlayers strongly depends on the grafting density of adhesive catechol groups. However, the adhesion mechanism of catecholic derivatives remains poorly understood. Because of the complex reduction−oxidation reaction of catechol groups, the formation of catecholic coatings is affected by their concentration in the solution, oxidant employed, oxygen concentration, pH, and temperature.22 Several models have been proposed to clarify the mechanism of catechol bonding, including coordination bonding,31 bidentate chelating bonding,32 bridged bidentate bonding,33 and mixed bonding.34 Very recently, a zwitterionic DA compound was synthesized as a surface ligand for modification of the contrast agent (e.g., superparamagnetic iron oxide nanoparticles) for magnetic resonance imaging.35−37 The modified nanoparticles have been demonstrated by their small hydrodynamic diameters and stability with respect to time, pH, and salinity, displaying the unique characteristics of zwitterions. However, for antifouling application, the zwitterionic SPIONs merely reduce nonspecific adsorption to a limited extent, which is incomparable with zwitterionic SAMs based on thiols on novel gold substrates.8,12,38,39 The possible reasons for this reduced effectiveness of the zwitterionic dopamine ligand include insufficient surface density and weak strength of catecholic bonding. In this work, we report a systematic investigation on the binding mechanism of zwitterionic sulfobetaine dopamine (SBDA) in response to pH transition, leading to a modification
2. EXPERIMENTAL PROCEDURES 2.1. Materials. Dopamine hydrochloride, absolute ethanol, 28% ammonium hydroxide, 1,3-propane sultone, dimethylformamide (DMF), anhydrous sodium carbonate, iodomethane, ethyl acetate, acetone, and bovine serum albumin (BSA) were purchased from Sigma-Aldrich (St. Louis, MO) and used as received. Sodium dodecyl sulfate (SDS) and phosphate buffered saline (PBS) were obtained from Acros Organics (Morris Plains, NJ). LB (Luria-Bertani) agar was obtained from BD (NJ). The LIVE/DEAD BacLight Bacterial Viability kit was purchased from Life Technologies (Carlsbad, CA). 2.2. Synthesis of SB-DA. The synthesis of SB-DA was as described in a previous publication.35 Briefly, dopamine hydrochloride (1.137 g, 6 mmol) was dissolved in absolute ethanol (70 mL) and mixed with 1,3-propane sultone (799 mg, 6.5 mmol) in ethanol (5 mL) in a round-bottom flask. 28% ammonium hydroxide (416 μL, 3 mmol) was slowly added to the round-bottom flask and stirred at room temperature for 10 min. The solution was heated to 65 °C and then stirred for 18 h. The white precipitate was filtered and washed with ethanol three times. The residual white solid was dried under reduced pressure. The product was dopamine sulfonate, and the yield was 80%. Dopamine sulfonate (0.3286 g, 1 mmol) was dissolved in dimethylformamide (DMF) (25 mL) in a round-bottom flask. Anhydrous sodium carbonate (0.2544 g, 2.4 mmol) was dissolved in DMF (50 mL) in protection of N2. The dopamine sulfonate solution was dropped into sodium carbonate solution and stirred at 0 °C, followed by the addition of iodomethane (836 μL, 35 mmol) and stirring at 0 °C for 10 min. The solution was then kept at 65 °C for 20 h, during which time the solution turned yellow due to methylation. The DMF was removed using a rotary evaporator at 50 °C, yielding an oily mixture. Then 50 mL of ethyl acetate was added to precipitate out a pale-yellow product. After this, 50 mL of acetone was added to the crude product, and the mixture was refluxed at 55 °C for 2 h. The solution mixture was filtered again, and the precipitate was collected. The white solid product of SB-DA was dried under reduced pressure, and the yield was 95%. The NMR spectrum of SB-DA is shown in Figure SI-1 of the Supporting Information. 2.3. Preparation of Thin Films by SB-DA. A metallic substrate was prepared by deposition of titanium onto a 20 mm × 20 mm glass slide in a high-vacuum thermal evaporator. A 50 nm thick layer of titanium was formed and oxidized in ambient conditions to have a layer of oxide. The TiO2 substrates were cleaned by thorough washing with 1% SDS detergent, deionized water, and absolute ethanol, followed by drying in a stream of nitrogen. The substrates were exposed to O2 plasma in a plasma cleaner (PDC-32G, Harrick Plasma, NY) for 10 min in order to remove final traces of organic contamination on the surface. Buffers were prepared by titrating PBS (1×) with hydrochloric acid (for pH 3, 4, and 6); the pH of Tris buffer at a concentration of 2 mM was adjusted to 8. The cleaned TiO2 substrates were immersed individually in buffers at pH of 3, 4, 6, and 8 containing SB-DA molecules at a concentration of 2 mg mL−1 for 15 h at room temperature. In the pH transition approach, the substrates were subsequently transferred to a pH 8 buffer and incubated for 20 12639
dx.doi.org/10.1021/la503191b | Langmuir 2014, 30, 12638−12646
Langmuir
Article
Figure 1. Repetitive CVs for the oxidation of (a) DA at pH 8, (b) SB-DA at pH 8, (c) DA at pH 3, and (d) SB-DA at pH 3. The DA and SB-DA solutions were prepared at a concentration of 5 mM. The baselines (black lines) were developed at buffer solutions without DA and SB-DA. CVs were run at a scan speed of 0.1 V s−1 at 5 s intervals. angles at solid−liquid interfaces. The droplets used were 5 μL with a microsyringe, and the measurements were performed at least three times at random positions on each sample. 2.6. Ellipsometric Thickness of SB-DA Thin Film. Ellipsometric measurements for the thin films on TiO2 were performed with a Gaertner LSE Stokes ellipsometer equipped with a He−Ne laser (λ = 632.8 nm) at a fixed incidence angle of 70°. The bare substrates were measured to find the Ns (2.5), Ks (−3.2), and refractive index (n = 1.00) of the ambient. The refractive index of the organic thin films on the substrates was fixed to n = 1.446, and the thicknesses were automatically calculated by the measurement program. The measurements were performed at least three times at random locations on each sample. 2.7. XPS Analysis. The chemical element spectra were detected by a PHI 5000 VersaProbe system (ULVAC-PHI, Chigasaki, Japan) with a microfocused, monochromatic Al Kα X-ray (25 W, 100 μm). The takeoff angles (with respect to the surface) of the photoelectron were set at 20° and 90°. The pressure of the system was maintained below 10−8 Pa using oil-less ultrahigh vacuum pumping systems. A dual beam charge neutralizer (7 V Ar+ and flooding 1 V electron beam) was employed to compensate for the charge-up effect. Spectra were collected with the pass energy set to 58.7 eV, while the binding energy measured was calibrated against the Ti 2p3/2 peak at 454.1 eV. The typical data acquisition time was around 1 h. The ratio of peak intensity converted to atomic percentage using the sensitivity factors and thickness simulation was analyzed with the MULTIPAK software package. 2.8. QCM-D for BSA Fouling Tests. The TiO2-covered QCM crystal chips (AT-cut quartz crystals, f 0 = 5 MHz) (Q-Sense AB, Gothenburg, Sweden) were cleaned with the protocol used in a previous publication.40 Before the measurement, the chamber was rinsed with phosphate buffered saline (PBS) and the temperature was stabilized at 25 °C. A 1 mg mL−1 BSA solution in PBS was brought in contact with the sensor chip at a flow rate of 1 mL min−1 for 15 min, followed by rinsing with PBS. All measurements were recorded at the third overtone (15 MHz), and the data shown here were normalized to the fundamental frequency (5 MHz) by dividing by the overtone
min. Afterward, they were washed with large amounts of deionized water and dried in a stream of nitrogen for further measurements. 2.4. Electrochemical Properties of DA and SB-DA. A potentiostat (Multi Autolab, Metrohm, Utrecht, Netherlands) with a three-electrode configuration was used for all electrochemical studies. Ag/AgCl served as reference electrode, with a Pt counter electrode. The working electrodes were as described above. The electrochemical properties of DA and SB-DA were investigated at ambient temperature by cyclic voltammetry. The sample solutions were prepared at a concentration of 5 mM in buffers at pH 3 and 8. At pH 8, the potential was scanned from −1 to +5 V; at pH 3, from −1 to +8 V for 10 times at a scanning rate of 0.1 V s−1. The parameters in cyclic voltammograms (CVs) were determined, including the formal potential for a redox couple, E0′, separation between the two peak potentials, ΔEp, and current ratio, Ip. E0′ is characteristic of a redox species and obtained by averaging the two peak potentials
E 0′ =
Ep,f + Ep,r 2
where Ep,f and Ep,r represent the forward and reverse peak potentials, respectively. The ΔEp in volts can be used to determine the electrochemical reversibility for a redox couple, with
ΔEp =
0.058 n
for the reversible case, where n is the number of transferred electron. The Ip is used to observe the stability of the electrochemically generated product. If the product is stable over the span of experiment, the peak current for the return potential scan (ip,r) should be equal to that for the forward potential scan (ip,f). It represents a reversible electron transfer reaction in both directions and is obtained as Ip =
i p,r i p,f
2.5. Contact Angle Measurements. A contact angle goniometer (PHOENIX MINI, Korea) was used to measure static water contact 12640
dx.doi.org/10.1021/la503191b | Langmuir 2014, 30, 12638−12646
Langmuir
Article
Scheme 1. Plausible Redox Mechanisms of DA and SB-DA in Solutions at pH 8 and 3
number. Owing to the globular and relatively rigid structure of BSA, the increased mass on the chip is related to changes in frequency of the oscillating crystal through the Sauerbrey relationship:
ΔmSauerbrey =
CQCMΔf n
where ΔmSauerbrey represents the mass adsorbed on the quartz sensor, Δf is resonance frequency, CQCM is the mass-sensitivity constant (= 17.7 ng cm−2 Hz−1 at f = 5 MHz), and n is the overtone number (= 1, 3, 5, and 7).41−43 2.9. Bacterial Fouling Tests. Pseudomonas aeruginosa (P. aeruginosa) and Staphylococcus epidermidis (S. epidermidis) were streaked out onto a LB (Luria-Bertani) agar plate from a bacteria 20% glycerol stock solution stored at −80 °C. A single bacteria colony was picked from the LB agar plate to inoculate 25 mL of liquid LB growth media. After 12 h inoculation at 37 °C shaking at 280 rpm, 1 μL of bacteria containing media from the first culture was used for a secondary culture also in LB in a conical flask. The bacteria secondary culture was subsequently incubated at 37 °C for another 12 h. The bacteria were then washed with sterile PBS solution three times through centrifugation at 4000 rpm for 5 min and resuspension in PBS. After the final wash, the bacterial samples in PBS were diluted to an optical density reading at 670 nm (OD670) of 0.1 to be tested for antifouling properties of substrates. Before the test, the substrates were modified with DA or SB-DA according to the preparation methods above. The bacterial solution was added to the incubate with substrates at 37 °C for 3 h, followed by replacing by sterile PBS and shaking at 100 rpm for 5 min for three times. The adsorbed bacteria on substrates were stained with 50 μL of bacterial viability dye and covered with paraffin for 15 min. Then substrates were washed with PBS and observed under a fluorescence microscope (ZEISS Microscope Axio Obserber A1, Germany) with a magnification of 400× and an excitation wavelength of 488 nm. The measurements were performed at five random locations on each sample and analyzed using ImageJ software package (developed at National Institutes of Health, Bethesda, MD).
Figure 2. Contact angle measurements for the samples of bare TiO2, DA(8), SB-DA(8), SB-DA(3), and SB-DA(3−8). The samples of DA(8), SB-DA(8), SB-DA(3), and SB-DA(3−8) were referred according to their reactants and pH values in preparation.
pH values was monitored using the repetitive cyclic voltammetry (CV) to consider the effect of the redox reaction on the coating formation. The experiments were carried out with 5 mM DA and SB-DA in buffers at pH 8 and pH 3. Ten CVs at a scan speed of 0.1 V s−1 were recorded at 5 s intervals, as shown in Figure 1 as well as the baseline (black lines) without reactants. At pH 8, the reverse peaks on the potential axis (Ep,r) for DA and SB-DA were hardly extrapolated, indicating that only a minute amount of reduction electrons were detected. The forward peaks for DA shifted from Ep,f = 0.342 to 0.435 V, and the forward peak currents changed from ip,f = 44.8 μA to ip,f 24.9 μA after 10 scans. The shift of Ep,f should be attributed to the progressive multistep electrochemical processes of DA switching from oxidized mediates to polydopamine.44,45 The decrease in ip,f reflects the consequences of consumption of DA monomers and deposition of a polydopamine film on the electrode that interferes with the electron transport. Interestingly, current densities for SB-DA were significant lower than those for DA over the course of cycling scans. In addition, no peak clearly appeared when the potential sweep was reversed. The results can be ascribed to the
3. RESULTS AND DISCUSSION 3.1. pH-Modulated Reduction−Oxidation of SB-DA. Since the reduction−oxidation (redox) mechanism of the catecholic molecules exhibits the pH-dependent behaviors,16,22 the electron transferring of DA or SB-DA dissolved at distinct 12641
dx.doi.org/10.1021/la503191b | Langmuir 2014, 30, 12638−12646
Langmuir
Article
Figure 4. QCM-D studies for the BSA adsorption on SB-DA SAMs prepared at the initial pH values from 3 to 8. The real-time kinetics of BSA adsorption is shown in the changes of resonance frequency (a) and dissipation (b). The calculated surface mass through the Sauerbrey relationship as a function of the initial pH values is present (c).
Figure 3. XPS spectra of N 1s (a), S 2p (b), and O 1s (c) for samples of SB-DA(3) and SB-DA(3−8).
Scheme 2. Binding Mechanism of the SB-DA Molecules with the pH Transition Approach and under Oxidation at pH 8
conversion of amine groups in DA to sulfobetaine moieties in SB-DA, which disabled the intramolecular cyclization upon the oxidation. Oxidation of SB-DA, therefore, merely involves quick irreversible electron transport to become SB-1,2- benzoquinone, as illustrated in Scheme 1. At pH 3, the resulting voltammograms showed two distinct peaks for both DA and SB-DA, and the shapes of the CVs remain almost unchanged during potential cycling. The important parameters in CVs were determined, including the formal potential for a redox couple, E0′, separation between the two peak potentials, ΔEp, and current ratio, Ip. The measurements and physical meanings of the parameters are described in the Experimental Procedures. Herein, the E0′, as characteristic of a redox species, for DA at pH 3 is 0.404 V, and that for SBDA is 0.519 V. The ΔEp values are 0.336 and 0.079 V for DA and SB-DA, corresponding to the numbers of transferred electrons as n = 0.17 and 0.73, respectively. The Ip value is used to directly observe the stability of the electrochemically generated product. For DA, the Ip value is about 0.603. The low Ip may be attributed to a few amine groups in DA undergoing irreversible cyclization. It is noteworthy that the Ip values for SB-DA at pH 3 had no noticeable change over the 12642
dx.doi.org/10.1021/la503191b | Langmuir 2014, 30, 12638−12646
Langmuir
Article
Figure 5. Bacterial adsorption tests for P. aeruginosa (a) and S. epidermidis (b) on samples. The bacteria were brought to contact with samples under a physiological condition at pH 7.4 and 37 °C in PBS. Patterned TiO2 surface with SB-DA prepared through the pH transition approach was incubated with P. aeruginosa solution and imaged under fluorescence microscope after washing (c). The bar present in image is 10 μm.
intermolecular interactions with surfaces of interest.33,48,49 According to the literature44,45 and electrochemical results in the current study, the chemical structure of the polydopamine film is complex, which it makes difficult to identify the attachment mechanism. Herein, at least to a certain extent, the low contact angles of DA and SB-DA films reflect the presence of SB moieties on TiO2, leading to the strong association with water molecules. For the SB-DA(8), however, its contact angle was significantly higher than other SB-DA samples, indicating insufficient surface coverage. At pH 8, the catecholic hydroxyl groups are oxidized to irreversibly transform to quinone groups, as indicated in Scheme 1. The low contact angle is most likely attributed to the disability of the quinone group to interact with TiO2. In contrast, the oxidation of catechol is inhibited at pH 3, and the TiO2 surface may thereby directly interact with the hydroxyl groups of SB-DA. Interestingly, no considerable difference was found in the contact angles for the samples of SB-DA(3) and SB-DA(3−8). Though the actual surface density of SB-DA cannot be determined from macroscopic contact angle measurements, adsorption of SB-DA onto TiO2 surfaces at pH 3 can be confirmed by surface wetting measurements. Since the formation of the covalent binding of the catechol group to the metal oxide surface is susceptible to pH,28,47,50 the results here indicate a possible approach to graft a highly packed and stable SB-DA film by pH transition in the preparation. 3.3. Elemental Composition and Chemical State of Thin Films by XPS. The elemental composition and chemical state of the samples of SB-DA(3) and SB-DA(3−8) were determined using XPS for scrutinizing the plausible molecular binding mechanism upon the pH transition. XPS signatures originating from N, S, and O atoms within the films were measured on the samples. As shown in Figure 3a, N peak components with BE at 402.5 eV appear in the XPS N 1s core-
cycles and are approximately 1, indicating fully reversible behavior in the redox reaction between catechol quinone. From the parallel electrochemical study, the zwitterionization to DA molecules affects their voltammetric signatures and blocks the intramolecular cyclization and subsequent polymerization. Although the redox characteristic of catechol groups in SB-DA generally remains intact, the quaternary ammonium cation is a strong electron-withdrawing substituent that likely alters the redox mechanism by changing the electron distribution and facilitating the electron transport of catecholic hydroxyl groups as shown in low ΔEp. The pH-dependent redox reaction observed here may occur in nitrodopamine derivatives, as developed by previous studies.23,28,46 These molecules have been proved to exhibit high stability on superparamagnetic iron oxide nanoparticles.24,27 It has been suggested that the nitro group, an electron-withdrawing substituent as well, can be attributed to the increased acidity of the catecholic hydroxyl groups and the subsequent strengthening of the molecular bonding.28 3.2. Surface Wetting of Thin Films. The modified TiO2 substrates were first examined using a contact angle goniometer for the surface wettability. The samples were prepared from solutions containing DA at pH 8, SB-DA at pH 8, SB-DA at pH 3, and SB-DA at pH values changing from pH 3 for incubation to pH 8 for washing. The resultant samples are referred to as DA(8), SB-DA(8), SB-DA(3), and SB-DA(3−8), according to their reactants and pH values. As shown in Figure 2, the contact angles of DA and SB-DA substrates were lower than bare TiO2. The contact angles of SB-DA(3) and SB-DA(3−8) reach a detectable minimum (
