- Email: [email protected]
Polydopamine-melanin initiators for Surface-initiated ATRP
Polymer 52 (2011) 2141e2149
Contents lists available at ScienceDirect
Polymer journal homepage: www.elsevier.com/locate/polymer
Polydopamine-melanin initiators for Surface-initiated ATRP Bocheng Zhu, Steve Edmondson* Department of Materials, Loughborough University, Loughborough LE11 3TU, UK
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 November 2010 Received in revised form 4 March 2011 Accepted 16 March 2011 Available online 23 March 2011
Thin films of polydopamine, a melanin-like material formed by the oxidative polymerization of dopamine in aqueous solution, have attracted much recent interest as functional coatings. Here we demonstrate the use of modified polydopamine as an initiator layer for surface-initiated polymerization, synthesized by reacting dopamine with 2-bromoisobutyryl bromide (BIBB) before deposition. Modified polydopamine films grew at a comparable rate to unmodified polydopamine, with a 45 nm being grown in 24 h. The presence of initiator groups was confirmed by XPS and FTIR and by the growth of PMMA and PHEMA polymer brushes. ARGET, the recently-developed oxygen-tolerant variant of ATRP, was used for surface polymerization with sodium ascorbate as the reducing agent. Polymerization was rapid, with PMMA brushes of nearly 240 nm thickness being grown in 24 h. A range of substrates could be coated in this way, including metals and hydrophobic polystyrene, demonstrating the broad applicability of this initiator. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Polydopamine Brushes ATRP
1. Introduction Surface-initiated polymerization (SIP) [1e3] is the growth of polymer from initiators immobilized on a surface, producing layers of densely end-grafted chains known as polymer brushes. Using SIP to grow these thin polymer films has become a very popular way to tailor interfacial properties for a range of applications such as responsive “smart” surfaces [4,5], antibacterial coatings [6e8], biocorrosion resistance [9,10] and antifouling surfaces [11e13]. Polymer films produced in this way show excellent solvent stability due to the strong grafting of each polymer chain to the surface. A range of controlled and uncontrolled polymerization chemistries have been used for SIP [2]. However, in the majority of studies, copper-mediated atom transfer radical polymerization (ATRP) has been applied. This is likely due to the relative ease with which fairly well-controlled polymerization can be obtained using relatively simple ligands for the copper and commercially available or easilysynthesized initiators. More recently, several studies have reported the use of ARGET (activators regenerated by electron transfer) ATRP, a variant of conventional ATRP, for SIP [14e16]. In ARGET ATRP, excess reducing agent added to the polymerization solution generates the Cu(I) activator species, reducing the sensitivity of the system to oxygen and allowing greatly reduced catalyst quantities to be employed [17].
* Corresponding author. E-mail address: [email protected] (S. Edmondson). 0032-3861/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymer.2011.03.027
Initiator immobilization is the inevitable first step in the process of SIP. Different strategies and chemistries are required to immobilize initiators onto different surfaces (e.g., alkanethiols on noble metals, silanes on metal oxides) and the synthesis of surfacespecific small-molecules initiators is usually required. This requirement for chemical specificity between the initiators and surfaces complicates the practical application of SIP and limits its use for high surface area substrates (since large amounts of initiator may need to be synthesized). Developing a simple and versatile strategy for initiator immobilization applicable to a wide range of surfaces is desirable for extending the technological application of SIP. Some progress has been made towards this goal with the development of the macroinitiator approach to SIP [18]. Macroinitiators for SIP are polymers containing many initiator sites (e.g. as pendant groups on repeat unit) which are adsorbed to the surface. Whereas small molecule initiators are typically immobilised using surface-specific covalent-bond-forming reactions, macroinitiators are immobilised using a large number of weaker non-covalent interactions (e.g. electrostatic attraction [19,20], hydrogen bonding [21] or hydrophobic interaction [5]) with cooperativity of the many binding sites ensuring overall strong attachment. Macroinitiators can be synthesized on a large scale allowing their application to high surface area substrates. Moreover, each macroinitiator can be applied to a broad class of surfaces (e.g. cationic macroinitiators can be applied to all anionic surfaces). However, they do not represent a truly universal initiator for all surfaces. The results presented here present a promising step in the development of such an initiator.
2142
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
Recently, Messersmith and co-workers have reported that the bio-inspired polymerization of dopamine allows the formation of adherent polydopamine films on a very wide range of substrates [22]. The polymerization system is simple e a dilute aqueous solution of dopamine is buffered to pH 8.5 by TRIS, and substrates are coated by simply being dipped into the polymerization solution. Cross-linked hydrophilic films of polydopamine, which is structurally similar to natural eumelanin pigments [23], can be formed in this way with thicknesses of up to tens of nanometers. This research was inspired by the adhesive protein secreted by mussels [24e26], which have been shown to attach to virtually all types of inorganic and organic surfaces, even to conventionally non-adhesive materials such as poly(tetrafluoroethylene) (PTFE), in marine environments. This adhesive versatility was proposed to lie in the co-existence of catechol and amine groups in the protein structure [22]. Dopamine is a small-molecule chemical that contains both functionalities, and thus was chosen by Messersmith as the monomer for a simple synthetic polymer mimic for mussel adhesives. The broad applicability of polydopamine coating has led us to explore this technology as a platform for producing initiator for SIP that can be applied to a broad range of substrates. The mechanism of dopamine polymerization is proposed to occur in a manner reminiscent of melanin formation [23,27,28], involving oxidation of catechol to quinone, resulting in the formation of dopaminechrome, which can isomerize into 5,6-dihydroxylindole [29]. Dopaminechromes and 5,6-dihydroxylindoles are then proposed to react with themselves or with each other to form the adherent polydopamine film. There is much recent interest in the technological application of these coatings. For example, polydopamine has been deposited on porous membranes to improve hydrophilicity [30], to control pervaporation [31], or as an adhesion layer in multi-layer membranes [32]. Films grown on Nafion membranes enhanced methanol barrier properties for fuel-cell applications [33], and deposition of polydopamine inside layer-bylayer polyelectrolyte multilayers improved the mechanical properties allowing free-standing film to be produced [34]. Polydopamine capsules have also been synthesized for future biomedical applications [35] and have been shown to have interesting uptake/release behaviour [36]. On planar surfaces, semiconductor nanocrystals have been embedded in a polydopamine coating for sensor applications [37]. Polydopamine coatings have also been explored as a versatile platform for secondary reactions such as electroless deposition of silver [38,39], in-situ gold nanoparticle synthesis [40] and biomolecule immobilisation [41]. Of particular interest for this study are reports by Messersmith and co-workers [42e45] of a monomeric dopamine-based initiator for surface-initiated ATRP. This initiator is a functionalised dopamine which binds to metal oxide surfaces using specific interaction with the catechol group. Since these initiators do not polymerise into a polydopamine, polymer brush chains grown from these initiators may suffer stability problems since surface attachment is only through a single non-covalent interaction. Although a useful addition to the “arsenal” of initiators available for SIP, this system does not take advantage of the cross-linking inherent in a film made from a polydopamine. We expect this cross-linking to improve the robustness of the coating even on surfaces where it is only weakly adhered (since the only route to degrafting the polymer chains is to delaminate the whole initiator layer). However, the presence of polar functional groups in a polydopamine films (hydroxyls and amines) suggests that good adhesion should be expected for many substrates. In this work, we report a polydopamine-based initiator for surface-initiated ATRP systems, which we hope could be applied to a wide range of substrates. These initiator films were synthesized by functionalising a fraction of dopamine monomers with
2-bromoisobutyryl bromide (BIBB) before polymerization, introducing 2-bromoisobutyrate ATRP initiator groups into the polydopamine films. Grafted polymer brushes could then be grown from these polydopamine initiators using surface-initiated ARGET ATRP. Since this work was conducted, other reports of the use of polydopamine films for surface-initiated ATRP have appeared, in which the polydopamine is modified after deposition. This approach either requires the synthesis of nucleophilic thiol-functional initiators [46] or exposing the sample to solvents which may damage some substrates (e.g. polymers) [47,48]. Our approach uses only commercial reagents and avoids the need for two discrete surface-modification steps, introducing initiators in a one-step process consisting of simple immersion into the modified dopamine solution. 2. Experimental 2.1. Materials All reagents and solvents were purchased from SigmaeAldrich (Gillingham, UK) or Fisher Scientific (Loughborough, UK) and used as received. Water was deionized and obtained from a Elga Option 4 system. Silicon wafers (‹100› orientation, boron doped, 1e100 U cm, polished one side) were purchased from Compart Technology Ltd. (Peterborough, UK). Aluminium (Al) was commercial foil for food use, containing 98.6% aluminium. Glass slides were 0.13e0.17 mm thick microscope cover glass. Steel chips were 1.15 mm thick stainless steel. Polystyrene chips are 0.70 mm thick and cut from polystyrene petri dishes (Fisher Scientific, UK). 2.2. Synthesis 2.2.1. Polydopamine deposition Silicon wafers pieces (w1 cm2) were cleaned and rendered hydrophilic by first washing with acetone, methanol and water, and then immersed for 15 min in a mixture of ammonia solution (28 mL, 35% wt), hydrogen peroxide solution (28 mL, 30% wt) and water (142 mL) at 75 C. Wafers were removed, rinsed thoroughly with water and dried under a stream of nitrogen. Al foil and glass slides were used after washing with acetone and methanol. Polystyrene chips were used after washing with methanol alone. Stainless steel chips were cleaned and rendered hydrophilic by first washing with acetone, methanol and water, and then treated in a UV-Ozone reactor for 1000 s. The substrate was then immersed in a solution of 3-hydroxytyramine hydrochloride (dopamine hydrochloride, 200 mg, 1.05 mmol) and tris(hydroxymethyl) aminomethane (TRIS, 120 mg, 1.0 mmol) in deionised water (100 mL) in a glass dish open to the air. This solution was continuously magnetically stirred at a speed of 200 rpm at room temperature. The clear solution initially became pink coloured, then darkened as stirring continued, finally becoming black after around 10 min. Polydopamine-coated substrates were removed from the solution after various deposition times, washed with deionised water and dried with compressed air. Polydopamine particles were obtained from a polymerization solution prepared as above which was continuously magnetically stirred at a speed of 200 rpm at room temperature for 24 h while open to the air. Particles were collected by filtration and dried in a vacuum oven. 2.2.2. Polydopamine initiator deposition Dopamine (400 mg, 2.10 mmol) was placed in a flask which was degassed by purging with dry N2 for 5 min. To this flask was added N,N0 -dimethylformamide (DMF, 20 mL), 2-bromoisobutyryl
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
bromide (BIBB, 0.13 mL, 1.05 mmol), and triethylamine (0.15 mL, 1.05 mmol) under dry N2. After stirring under dry N2 at room temperature for 3 h, this mixture was transferred to a glass dish to which tris(hydroxymethyl) aminomethane (TRIS) (480 mg, 4.0 mmol) and deionised water (100 mL) were added. Cleaned silicon wafer sections (w1 cm2), polystyrene chips, Al foil and stainless steel chips were then immersed in this new mixture which was continuously magnetically stirred at a speed of 200 rpm while open to the air. Polydopamine initiator-coated substrates were removed from the solution after various deposition times washed with deionised water and dried with compressed air. Samples with varying BIBB:dopamine ratios were prepared as above, but with the BIBB and triethylamine concentrations being altered according to the desired ratio. Polydopamine initiator particles were obtained from a polymerization solution prepared as above which was continuously magnetically stirred at a speed of 200 rpm at room temperature for 24 h while open to the air. Particles were collected by filtration and dried in a vacuum oven. 2.2.3. Surface-initiated ARGET ATRP from polydopamine initiators A solution of methyl methacrylate (MMA, 20 mL, 18.72 g, 187.0 mmol) in methanol (20 mL) was degassed by bubbling through dry N2 for 15 min in a flask sealed with a septum. To this solution was added copper (II) bromide (4 mg, 0.018 mmol), (þ)-sodium L-ascorbate (354 mg, 1.787 mmol) and 2,20 -dipyridyl (6 mg, 0.038 mmol). To dissolve all solids, the mixture was magnetically stirred for 5 min while degassing continued, giving a dark brown solution. In each glass tube was placed an initiatorcoated silicon wafer section (w1 cm2) and the tube was sealed with a septum. The tube was degassed by purging with dry N2 for 1 min and the monomer solution was then syringed over the wafer. After the polymerization was allowed to proceed at room temperature for various times the wafer was removed and washed sequentially with methanol and water, and dried under a compressed air stream. ARGET ATRP growth of MMA from polydopamine initiatorcoated steel chips and Al foil was conducted as above except that the methanol (20 mL) was replaced with a mixture of methanol (20 mL) and deionized water (10 mL). ARGET ATRP growth of HEMA from polydopamine initiator-coated steel chips and polystyrene chips was also conducted as above except that methyl methacrylate (20 mL, 18.72 g, 187.0 mmol) and methanol (20 mL) in the polymerization solution was replaced with 2-hydroxyethyl methacrylate (20 mL, 20.74 g, 164.8 mmol) and a solvent mixture of methanol (10 mL) and deionized water (10 mL).
2143
Scheme 1. Schematic illustration of the polymerization of dopamine hydrochloride into adherent polydopamine coatings on silicon wafers.
quality. Thus, all fitting assumed no optical absorption for polydopamine in order to simplify modelling. For PMMA grown from polydopamine initiator films a four-layer model was used, consisting of silicon, silicon dioxide (2 nm), polydopamine (n ¼ 1.6, 58 nm), PMMA (thickness fitted). The polydopamine layer thickness was measured before PMMA growth, and the PMMA refractive index was assumed to be n ¼ 1.5 at all wavelengths. XPS spectra were obtained using a VG Scientific Escalab 5 Mark II X-ray photoelectron spectrophotometer, with an unmonochromated Al source. A Shirley background was used, and peaks were fitted using XPSPEAK 4.1. FTIR spectra were obtained using a Shimadzu FTIR-8400S Fourier transform infrared spectrophotometer. Measurements on silicon samples were taken in transmittance mode using a blank silicon wafer as a background. The measurements of other substrates (steel, Al foil and PS) samples were taken in reflectance mode (i.e. ATR-FTIR). 3. Results and discussion 3.1. Deposition of unmodified polydopamine Before investigating the synthesis of polydopamine initiators, the deposition of unmodified polydopamine on cleaned silicon wafers was attempted (Scheme 1). By simply immersing a cleaned silicon wafer in aqueous solution of dopamine and TRIS, a 47 nm polydopamine layer was grown in 24 h (Fig. 1). This is consistent with the results reported by Messersmith and co-workers [22] who measured a 50 nm polydopamine film after 24 h. continuing the deposition beyond 24 h leads to a continued increase in thickness, but at a reduced growth rate. This is presumably due to dopamine monomer consumption (since polymerization is also occurring in solution). To demonstrate the broader applicability of polydopamine deposition, films were grown on four other substrates: glass slides,
2.3. Characterisation Ellipsometric measurements were conducted using a phasemodulated spectroscopic ellipsometer (Uvisel, Jobin Yvon) at 10 nm intervals from 500 nm to 700 nm at an angle of incidence of 70 . This wavelength range was chosen to avoid the strong optical absorption displayed by polydopamine at shorter wavelengths [28]. Modelling was conducted using the WVASE32 software package (J. A. Woolam Co., USA). For polydopamine and polydopamine initiator films a three-layer model was used, consisting of silicon, silicon dioxide (2 nm) and polydopamine (thickness fitted). Softwaresupplied refractive indices were used for silicon and silicon dioxide, and the refractive index of polydopamine was assumed to be n ¼ 1.6 at all wavelengths with no optical absorption (i.e. the non-zero imaginary part of the refractive index, k ¼ 0). Adding an absorption with k ¼ 0.02 at all wavelengths, as measured previously for polydopamine at 589 nm [28], produced only very small changes in the fitted thickness (<1%), and negligible improvement in fit
Fig. 1. Ellipsometric thickness against deposition time for unmodified (:) and BIBBmodified (-) polydopamine. Error bars are from ellipsometric fitting and are smaller than 0.5 nm where not shown.
2144
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
Al foil, stainless steel and polystyrene (PS) chips. On these substrates, a striking colour change is observed with increasing dopamine deposition time (Fig. 2). Polydopamine has a strong broad-band UVevisible absorption [28] and the darkening colour with growth time is indicative of polydopamine thickness. To assess the strength of adhesion between the polydopamine and various substrates, a tape peel test was conducted with polydopamine-coated silicon, steel, glass, aluminium, polystyrene (PS) and low-density polyethylene (LDPE). On every substrate, polydopamine remained on the surface after peeling (as judged by the remaining brown colour). Every sample did show some transfer of
material from the sample to the tape, however. This is likely to be a layer of weakly-bonded polydopamine particles formed in solution during the coating process. On all substrates, a second peel test (after removal of weakly-bonded material) did not remove the underlying polydopamine film, demonstrating good adhesion to the substrate. Ellipsometric measurements on polydopamine layers of various thicknesses (Supporting Information, Fig S1) confirm strong adhesion. In all cases, no significant change in polydopamine thickness was observed after tape peeling. Although a small amount of weakly bound particles was always transferred to the tape, this change is not detected by ellipsometry. Dopamine
Fig. 2. Photographs showing various polydopamine-coated substrates. A: Glass and Al foil pieces with increasing polydopamine deposition time. Polydopamine has a strong broadband UV/Vis absorption, giving rise to a brown colour increasing in intensity with deposited film thickness. B: Steel and polystyrene pieces with increasing polydopamine deposition time. C: A polydopamine-coated polystyrene sample is subjected to a tape peel test. Comparison with uncoated polystyrene reveals that a brown colour is still present on the peeled section, indicating that most polydopamine is not removed. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
monomer contains polar amine and hydroxyl groups, so strong adhesion to polar surfaces (e.g. silica and alumina) is not surprising [49]. More surprising is the adhesion to non-polar surfaces (PS and LDPE), demonstrating the broad applicability of polydopamine as a surface coating. 3.2. Deposition of ATRP-initiator-modified polydopamine The next step in the synthesis of a polydopamine-based initiator is the introduction of 2-bromoisobutyrate ATRP-initiating groups into the film. Initially, we sought to react the secondary amines and catechol hydroxyl groups of pre-formed polydopamine films with BIBB under basic catalysis. However, after reaction these films showed no capacity to initiate SI-ATRP, and in some cases the films were completely destroyed during the reaction. It is likely that our initial assumption that the polydopamine films would contain much nucleophilic functionality, and react with an electrophilic acid bromide, was incorrect. It has been shown that these films are in fact electrophilic, existing in the quinone form in basic solution, allowing reaction with nucleophilic amines and thiols [22,41]. Indeed, a recently reported alternative approach to ATRP-initiating films relies on the reaction of polydopamime with a thiol-functional ATRP initiator [46]. Nucleophilic hydroxyl groups can be incorporated into polydopamine-like films, but an alternative monomer containing an additional hydroxyl group (norepinephrine) must be used [50]. Since this work was completed, two studies have been reported by Wang and co-workers in which pre-formed polydopamine has been successfully reacted with BIBB, contrary to other reports of the electrophilic nature of polydopamine [47,48]. This could be a consequence of differences in the precise composition of the BIBB esterification mix (e.g. amount of excess base) or sample history (e.g. degree of exposure to oxygen). This suggests that with a thorough investigation, this reaction could be successfully applied. However, in light of reports of the electrophilic reactions of polydopamine, we chose to abandon this route. In our alternative approach, the dopamine monomer is reacted with BIBB under base catalysis before polymerization into polydopamine (Scheme 2). This reaction is carried out in polar aprotic solvent (DMF) to allow the dilution into water for the polymerization, and in the absence of air to prevent premature polymerization. After this reaction was allowed to proceed for 3 h, the mixture was diluted into water with TRIS buffer and the polydopamine polymerization allowed to proceed in the presence of air, as for unmodified dopamine. A greater amount of TRIS was required to increase the pH and initiate polymerization, presumably since
2145
unreacted acid bromide (BIBB) forms carboxylic acid upon addition of water, decreasing the pH. Our route to ATRP-initiating dopamine (pre-deposition dopamine modification) has advantages over those recently reported by others requiring post-deposition modification of polydopamine. Unlike the work of Hu et al. [46] our route does not require the synthesis and purification of a nucleophilic thiol-functional initiator. The work of Wang and co-workers [47,48] requires exposing the substrate to a THF solution of BIBB, obviously making this route incompatible with solvent-sensitive substrates such as polystyrene. Using our route the substrate is only exposed to a solvent of dilute DMF in water, which will be compatible with a much wider range of substrates (including polystyrene, as we show below). It is not known if the BIBB will preferentially react with the hydroxyl or amine groups on the dopamine (our attempts to use 13 C NMR to elucidate the reaction site have so far not been successful). However, in either case it is likely that this modification of dopamine will hinder or completely prevent polymerization. Thus, the mole ratio of BIBB to dopamine was chosen as 0.5, statistically leaving much of the dopamine unmodified so that polymerization could still proceed, but allowing modified dopamine to be incorporated into the polydopamine layer. It appears that the modification of a fraction of the dopamine with BIBB does not greatly affect the polymerization, with film growth proceeding at a similar rate to that for unmodified dopamine (Fig. 1). Substrates coated with polydopamine initiator display the same characteristic brown colour as unmodified polydopamine coatings, increasing in intensity with deposition time (Supporting Information, Fig S2). In order to assess the incorporation of 2-bromoisobutyrate groups into the polydopamine film, XPS was conducted on both unmodified and BIBB-modified films on silicon wafers (Fig. 3). An unmodified film deposited for 24 h showed strong C 1 s, O 1 s and N 1 s signals, consistent with polydopamine. No silicon peaks are observed (e.g. Si 2 s at 155 eV or Si 2p at 103 eV), indicating that the polydopamine film is both continuous (with no defects revealing the underlying wafer) and thicker than the w10 nm sampling depth of XPS, in agreement with ellipsometry (47 nm thickness). From relative peak areas, the surface composition (in atom %) was measured, giving the C:O:N atomic ratio from XPS as 75.4: 17.5: 7.1, close to the calculated ratio for dopamine monomer C:O:N ¼ 8:2:1 ¼ 72.7: 18.2: 9.1. In the BIBB-modified polydopamine film, the presence of a weak Br 3d signal (1.0 atom%) in the XPS spectrum confirms the incorporation of the initiator groups. The Br/N ratio allows us to estimate
Scheme 2. Schematic illustration of the reaction of dopamine with 2-bromoisobutyryl bromide (BIBB) and the subsequent deposition of initiator-functionalised polydopamine onto silicon wafers. PMMA polymer brushes are then grown by surface-initiated ARGET ATRP. The exact structure of BIBB-modified polydopamine is not known; in particular, BIBBfunctionalisation may also occur at the amine position.
2146
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
Fig. 3. XPS spectra for polydopamine and polydopamine initiator coatings with a deposition time of 24 h on silicon wafers. Data for BIBB-modified polydopamine has been vertically offset for clarity. Inset: Expanded 0e200 eV region.
the BIBB/dopamine ratio. For this sample, Br/N ¼ 0.20 compared to a target of 0.5, determined by the reaction stoichiometry. This apparent low yield could be due to incomplete reaction or poor incorporation of BIBB-modified dopamine monomer into the polymer. Ester formation at the catechol hydroxyl group will prevent oxidation into the quinone form, whereas amide formation may hinder cyclisation into the indole; in both cases, polymerization may be hindered. A small amount of silicon is also observed in the XPS spectrum. Since the ellipsometric thickness of this film was 45 nm (greater than the sampling depth of XPS), this may indicate some small defects in the coating, revealing the underlying wafer. XPS narrow-scan spectra (C 1 s, N 1 s, O 1 s and Br 3d) for both unmodified and initiator-modified polydopamine are included in the Supporting Information (Fig S3). Transmission FTIR on free polydopamine and polydopamine initiator powders (precipitated from solution during the growth of the coatings) provides further confirmation of the incorporation of initiator groups (Fig. 4). These powders are polymerised from exactly the same solution as the coatings and so should very similar compositions; the free powders were used to achieve sufficiently strong absorbances. The spectrum of unmodified polydopamine is consistent with the results reported by Fei et al. [40], with peaks at w1250 cm1, w1500 cm1 and w1600 cm1 deriving from the aromatic rings in the polymer and a broad peak at w3300 cm1 from hydroxyls, amines and water absorbed in this hydrophilic material. The FTIR spectrum of the BIBB-modified polydopamine displays two significant differences from unmodified polydopamine (Fig. 4, inset). A weak absorbance at around 1710 cm1 could be indicative of esters formed by reaction of BIBB with catechol hydroxyl groups, whereas a shoulder at w1650 cm1 on the aromatic peak at 1600 cm1 suggests the presence of amides formed by reaction of BIBB with the dopamine primary amine. In both spectra, peaks at w2300 cm1 are from atmospheric carbon dioxide, and peaks at 2800e3000 cm1 are attributed to CeH bonds. The increased intensity of the CeH peak in the initiator sample is consistent with the incorporation of initiator sites (which contain six CeH bonds each).
3.3. Surface-initiated ARGET ATRP from initiator-modified polydopamine on silicon wafers The ability to grow surface-initiated polymer from the BIBBmodified polydopamine provides the most important test of the presence of ATRP-initiating groups. No other functionality in the polydopamine film is known to initiate ATRP. When surface-initiated ARGET ATRP was attempted from unmodified polydopamine on silicon wafers, we found no increase in film thickness after 24 h. Indeed, a 15% decrease in polydopamine thickness was measured, indicating some degradation of the coating by the polymerization solution or desorption of weakly-bonded polydopamine at the surface. XPS measurements support the conclusion that unmodified polydopamine is not capable of initiating polymerization (Supporting Information, Figs S4 & S5). In contrast, BIBB-modified polydopamine films on silicon wafers were used to successfully grow surface-initiated PMMA brushes using ARGET ATRP (Scheme 2). Using methanol as a solvent, 72 nm of PMMA was grown from a 58 nm thick BIBB-modified polydopamine initiator (giving a total thickness of 130 nm) in 24 h (Fig. 5). At 72 h, a PMMA thickness of 239 nm was measured (296 nm total thickness). Such large PMMA layer thicknesses indicate that the grafting density is sufficiently high to be in the brush regime. In unpublished results, we have found that surface-initiated ARGET ATRP of PMMA from more conventional polyelectrolyte macroinitiators [18] can produce similarly thick brush layers, with thicknesses of over 350 nm possible after 24 h1 XPS of PMMA grown from polydopamine initiator for 24 h displays a surface atomic composition consistent with pure PMMA
1 Although ARGET ATRP is significantly less sensitive to oxygen than conventional ATRP [14], we have found that care is still needed for polymerization over long times. Improper sealing or deoxygenation of the reaction vessel or the use of nondeoxygenated solvent leads to a reduction in polymerization rate at long times, although the polymerization still proceeds. This slowing is presumably due to consumption of the reservoir of reducing agent (and so a reduction in Cu(I) concentration) before the end of the polymerization.
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
2147
Fig. 4. FTIR spectra for the polydopamine initiator and polydopamine particles with a polymerization time of 24 h. Data for BIBB-modified polydopamine has been vertically offset for clarity. Inset: Expanded 1400e2000 cm1 region, highlighting differences between the spectra.
(see Supporting Information, Fig S6). In particular, the absence of an N 1s signal from the underlying polydopamine layer suggests the PMMA layer is thick and continuous. Transmission FTIR of the same sample reveals features characteristic of both the polydopamine initiator and the PMMA brushes (see Supporting Information, Fig S7), consistent with the presence of both layers with a similar thickness. Initially, a mole ratio of BIBB : dopamine of 0.5:1 was chosen as a compromise between introducing sufficient initiator sites to produce high grafting density for surface-initiated polymerization, and inhibition of the dopamine polymerisation mechanism by functionalizing too many hydroxyl or amine sites. A subsequent study of the effect of BIBB : dopamine ratio (Supporting Information, Fig S8) indeed revealed that of a number of ratios tested (0.25:1, 0.5:1, 1:1, 2:1), 0.5:1 did indeed produce the thickest films after surface-initiated ARGET ATRP of MMA. For very high BIBB:dopamine (3:1), the dopamine polymerization was completely inhibited. To confirm strong grafting of these PMMA brushes to the surface, a silicon wafer sample consisting of 62 4 nm grown from a 25 nm polydopamine initiator was subjected to Soxhlet extraction with THF (a solvent for PMMA) for 9 h. After this aggressive washing procedure, the measured PMMA thickness was almost unchanged at 58 4 nm, confirming the covalent tethering of the PMMA to the polydopamine initiator, and the strong adhesion of the initiator to the surface.
consistent with initiator deposition, including the presence of bromine atoms from the 2-bromoisobutyrate initiator group. PMMA was then grown by ARGET ATRP for 24 h from steel and Al foil substrates. Although ellipsometry is not possible on these substrates, ATR-FTIR spectra were consistent with a thick PMMA coating, including the strong ester peak at 1720 cm1 confirming the successful surface-initiated polymerization (Fig. 6). Peaks from CeH stretching (2800e3050 cm1), ester C-O stretching (1140 cm1) and CeH bending (1430e1490 cm1) are also consistent with the presence of PMMA. XPS on these substrates was also consistent with a continuous, thick PMMA film (see Supporting Information, Fig S9). PMMA could not be grown from polystyrene, since the monomer solution caused swelling of the substrate. However, a more hydrophilic monomer, 2-hydroxyethyl methacrylate (HEMA), could be used to grow PHEMA polymer films from polystyrene. ATR-FTIR on this sample (Fig. 7) shows peaks originating from both the PHEMA and the underlying PS (notably
3.4. Surface-initiated ARGET ATRP from initiator-modified polydopamine on other substrates Our main motivation for the development of polydopaminebased ATRP initiators is substrate-independence; the same initiator can be applied to a wide range of substrates. To demonstrate this principle, polydopamine initiator was deposited for 24 h on steel, Al foil and polystyrene substrates. On all substrates, XPS analyses were
Fig. 5. Ellipsometric PMMA thickness against growth time for SI-ARGET ATRP of MMA from silicon wafers coated with a 58 5 nm polydopamine initiator layer.
2148
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
Fig. 6. ATR-FTIR spectra for PMMA grown for 24 h from initiator-functionalised polydopamine on steel (top) and Al foil (middle) surfaces. The spectrum of bulk PMMA (bottom) is included for comparison. For clarity, data for steel and Al foil has been vertically offset and the data for bulk PMMA has been vertically scaled.
aromatic CeH stretching at 3000e3100 cm1 and aromatic ring stretching modes at 1600 cm1, 1450 cm1 and 1490 cm1). Growing the PHEMA from the same initiator on steel allows the peaks originating from the PHEMA to be more clearly identified (although the steel can contribute some peaks in this range due to the thin oxide layer and adsorbed water [51], the peaks from the thick polymer film dominate the spectrum), including the strong ester peak at 1720 cm1. Peaks from CeH stretching (2800e3050 cm1), ester CeO stretching (1140 cm1), alcohol CeO stretching (1070 cm1) and OeH stretching (3000e3600 cm1) are also consistent with the presence of PHEMA. This grafted PHEMA spectrum is very similar to that of bulk PHEMA, with the exception of peaks relating to water present in the bulk PHEMA structure (notably 1660 cm1, HeOeH bend). On both steel and polystyrene, XPS analysis is consistent with a continuous, thick PHEMA film (see Supporting Information, Fig S9). XPS narrow-scan analysis of the C 1 s peak for polystyrene substrates confirms the changes at the surface after polydopamine initiator coating and subsequent growth of PHEMA (Fig. 8). Of particular interest is the pronounced sp2 C]O peak from the polydopamine initiator coating. This could be partially attributed to the oxidised quinone form of polydopamine [30], but could also be indicative of ester or amide groups formed during reaction with BIBB. Fig. 8. XPS narrow-scan spectra and fitted peaks for bare polystyrene substrates (top), after coating with polydopamine initiator for 24 h (middle), and after surface-initiated polymerization of PHEMA brushes for 24 h (bottom). Peaks have been assigned based on prior literature [30,52]. To compensate for sample charging, the fitted peak with the lowest binding energy has been corrected to 285 eV.
4. Conclusions
Fig. 7. ATR-FTIR spectra for PHEMA grown for 24 h from initiator-functionalised polydopamine on polystyrene (top) and steel (middle) surfaces.. The spectrum of bulk PHEMA (bottom) is included for comparison. For clarity, data for steel and polystyrene has been vertically offset and the data for bulk PHEMA has been vertically scaled.
In summary, we have demonstrated that polydopamine could be deposited on various surfaces by self-polymerization of dopamine. Bromoester initiating groups for ATRP were incorporated into polydopamine coatings by reacting a fraction of the dopamine monomer with 2-bromoisobutyryl bromide (BIBB) before polymerization. This modification did not appear to significantly impede coating deposition, although by XPS the incorporation of initiator groups into the film appeared to be lower than targeted. PMMA and PHEMA brushes could then be grown from these initiators by surface-initiated ARGET ATRP, with brush thicknesses of up to 239 nm possible after 72 h growth. A range of substrates could be coated, including metals and hydrophobic polystyrene,
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
showing the versatility of this initiator system. We are currently investigating a broader range of substrates in order to assess this system as a possible “universal initiator” for surface-initiated ATRP. Appendix. Supplementary Data Supplementary data related to this article can be found online at doi:10.1016/j.polymer.2011.03.027. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]
Zhao B. Progr Polymer Sci 2000;25(5):677e710. Edmondson S, Osborne VL, Huck WTS. Chem Soc Rev 2004;33(1):14e22. Tsujii Y, Ohno K, Yamamoto S, Goto A, Fukuda T. Adv Polymer Sci 2006;197:1e45. Lindqvist J, Nyström D, Ostmark E, Antoni P, Carlmark A, Johansson M, et al. Biomacromolecules 2008;9(8):2139e45. Mizutani A, Kikuchi A, Yamato M, Kanazawa H, Okano T. Biomaterials 2008; 29(13):2073e81. Cheng Z, Zhu X, Shi ZL, Neoh KG, Kang ET. Ind Eng Chem Res 2005;44(18): 7098e104. Huang J, Murata H, Koepsel RR, Russell AJ, Matyjaszewski K. Biomacromolecules 2007;8(5):1396e9. Li L, Ke Z, Yan G, Wu J. Polymer Int 2008;57(11):1275e80. Yuan SJ, Pehkonen SO, Ting YP, Neoh KG, Kang ET. ACS Appl Mater Interfaces 2009;1(3):640e52. Yuan SJ, Xu FJ, Pehkonen SO, Ting YP, Neoh KG, Kang ET. Biotechnol Bioeng 2009;103(2):268e81. Chen Y, Liu D, Deng Q, He X, Wang X. J Polymer Sci Polymer Chem 2006; 44(11):3434e43. Bozukova D, Pagnoulle C, De Pauw-Gillet M-C, Ruth N, Jérôme R, Jérôme C. Langmuir 2008;24(13):6649e58. Li L, Yan G, Wu J. J Appl Polymer Sci 2009;111(4):1942e6. Matyjaszewski K, Dong H, Jakubowski W, Pietrasik J, Kusumo A. Langmuir 2007;23(8):4528e31. Hansson S, Ostmark E, Carlmark A, Malmström E. ACS Appl Mater Interfaces 2009;1(11):2651e9. Cao L, Kruk M. Polym Chem 2010;1(1):97e101. Matyjaszewski K, Jakubowski W, Min K, Tang W, Huang J, Braunecker WA, et al. Proc Natl Acad Sci USA 2006;103(42):15309e14. Edmondson S, Armes SP. Polymer Int 2009;58(3):307e16. Edmondson S, Vo C-D, Armes SP, Unali G-F. Macromolecules 2007;40(15):5271e8. Edmondson S, Vo C-D, Armes SP, Unali G-F, Weir MP. Langmuir 2008;24(14): 7208e15. Stöhr T, Rühe J. Macromolecules 2000;33(12):4501e11. Lee H, Dellatore SM, Miller WM, Messersmith PB. Science 2007;318(5849): 426e30.
2149
[23] d’Ischia M, Napolitano A, Pezzella A, Meredith P, Sarna T. Angewandte Chemie International Edition 2009;48(22):3914e21. [24] Lin Q, Gourdon D, Sun C, Holten-Andersen N, Anderson TH, Waite JH, et al. Proc Natl Acad Sci USA 2007;104(10):3782e6. [25] Lee H, Scherer NF, Messersmith PB. Proc Natl Acad Sci USA 2006;103(35): 12999e3003. [26] Holten-Andersen N, Waite JH. J Dent Res 2008;87(8):701e9. [27] Park TJ, Kim J, Kim TK, Park HM, Choi SS. Bull Korean Chem Soc 2008;29(12): 2459e64. [28] Bernsmann F, Ponche A, Ringwald C, Hemmerle c J, Raya J, Bechinger B, et al. J Phys Chem C 2009;113(19):8234e42. [29] Li Y, Liu M, Xiang C, Xie Q, Yao S. Thin Solid Films 2006;497(1e2):270e8. [30] Xi Z-Y, Xu Y-Y, Zhu L-P, Wang Y, Zhu B-K. J Membr Sci 2009;327(1e2): 244e53. [31] Li B, Liu W, Jiang Z, Dong X, Wang B, Zhong Y. Langmuir 2009;25(13): 7368e74. [32] Chen J, Chen X, Yin X, Ma J, Jiang Z. J Membr Sci 2009;344(1e2):136e43. [33] Wang J, Xiao L, Zhao Y, Wu H, Jiang Z, Hou W. J Power Sources 2009;192(2): 336e43. [34] Bernsmann F, Richert L, Senger B, Lavalle P, Voegel J-C, Schaaf P, et al. Soft Matter 2008;4(8):1621e4. [35] Postma A, Yan Y, Wang Y, Zelikin AN, Tjipto E, Caruso F. Chem Mater 2009; 21(14):3042e4. [36] Yu B, Wang DA, Ye Q, Zhou F, Liu W. Chem Commun 2009;44:6789e91. [37] Zhang Y, Wang H, Nie J, Zhou H, Shen G, Yu R. Electrochem Commun 2009; 11(10):1936e9. [38] Liao Y, Wang Y, Feng X, Wang W, Xu F, Zhang L. Mater Chem Phys 2010; 121(3):534e40. [39] Liao Y, Cao B, Wang W-C, Zhang L, Wu D, Jin R. Appl Surf Sci 2009;255(19): 8207e12. [40] Fei B, Qian B, Yang Z, Wang R, Liu WC, Mak CL, et al. Carbon 2008;46(13): 1795e7. [41] Lee H, Rho J, Messersmith PB. Adv Mater 2009;21(4):431e4. [42] Hamming L, Fan X, Messersmith P, Brinson L. Compos Sci Technol 2008;68(9): 2042e8. [43] Fan X, Lin L, Messersmith P. Compos Sci Technol 2006;66(9):1198e204. [44] Dalsin JL, Messersmith PB. Mater Today 2005;8(9):38e46. [45] Fan X, Lin L, Dalsin JL, Messersmith PB. J Am Chem Soc 2005;127(45): 15843e7. [46] Hu H, Yu B, Ye Q, Gu Y, Zhou F. Carbon 2010;48(8):2347e53. [47] Wang WC, Wang J, Liao Y, Zhang L, Cao B, Song G, et al. J Appl Polym Sci 2010; 117(1):534e41. [48] Li CY, Wang WC, Xu FJ, Zhang LQ, Yang WT. J Membr Sci 2011;367(1e2):7e13. [49] Ryu J, Ku SH, Lee H, Park CB. Adv Funct Mater 2010;20(13):2132e9. [50] Kang SM, Rho J, Choi IS, Messersmith PB, Lee H. J Am Chem Soc 2009;131(37): 13224e5. [51] Mueller HJ, Freeman D. Mater Charact 1995;35(2):113e26. [52] Beamson G, Briggs D. High resolution XPS of organic polymers. Chichester: John Wiley & Sons; 1992.
Contents lists available at ScienceDirect
Polymer journal homepage: www.elsevier.com/locate/polymer
Polydopamine-melanin initiators for Surface-initiated ATRP Bocheng Zhu, Steve Edmondson* Department of Materials, Loughborough University, Loughborough LE11 3TU, UK
a r t i c l e i n f o
a b s t r a c t
Article history: Received 25 November 2010 Received in revised form 4 March 2011 Accepted 16 March 2011 Available online 23 March 2011
Thin films of polydopamine, a melanin-like material formed by the oxidative polymerization of dopamine in aqueous solution, have attracted much recent interest as functional coatings. Here we demonstrate the use of modified polydopamine as an initiator layer for surface-initiated polymerization, synthesized by reacting dopamine with 2-bromoisobutyryl bromide (BIBB) before deposition. Modified polydopamine films grew at a comparable rate to unmodified polydopamine, with a 45 nm being grown in 24 h. The presence of initiator groups was confirmed by XPS and FTIR and by the growth of PMMA and PHEMA polymer brushes. ARGET, the recently-developed oxygen-tolerant variant of ATRP, was used for surface polymerization with sodium ascorbate as the reducing agent. Polymerization was rapid, with PMMA brushes of nearly 240 nm thickness being grown in 24 h. A range of substrates could be coated in this way, including metals and hydrophobic polystyrene, demonstrating the broad applicability of this initiator. Ó 2011 Elsevier Ltd. All rights reserved.
Keywords: Polydopamine Brushes ATRP
1. Introduction Surface-initiated polymerization (SIP) [1e3] is the growth of polymer from initiators immobilized on a surface, producing layers of densely end-grafted chains known as polymer brushes. Using SIP to grow these thin polymer films has become a very popular way to tailor interfacial properties for a range of applications such as responsive “smart” surfaces [4,5], antibacterial coatings [6e8], biocorrosion resistance [9,10] and antifouling surfaces [11e13]. Polymer films produced in this way show excellent solvent stability due to the strong grafting of each polymer chain to the surface. A range of controlled and uncontrolled polymerization chemistries have been used for SIP [2]. However, in the majority of studies, copper-mediated atom transfer radical polymerization (ATRP) has been applied. This is likely due to the relative ease with which fairly well-controlled polymerization can be obtained using relatively simple ligands for the copper and commercially available or easilysynthesized initiators. More recently, several studies have reported the use of ARGET (activators regenerated by electron transfer) ATRP, a variant of conventional ATRP, for SIP [14e16]. In ARGET ATRP, excess reducing agent added to the polymerization solution generates the Cu(I) activator species, reducing the sensitivity of the system to oxygen and allowing greatly reduced catalyst quantities to be employed [17].
* Corresponding author. E-mail address: [email protected] (S. Edmondson). 0032-3861/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.polymer.2011.03.027
Initiator immobilization is the inevitable first step in the process of SIP. Different strategies and chemistries are required to immobilize initiators onto different surfaces (e.g., alkanethiols on noble metals, silanes on metal oxides) and the synthesis of surfacespecific small-molecules initiators is usually required. This requirement for chemical specificity between the initiators and surfaces complicates the practical application of SIP and limits its use for high surface area substrates (since large amounts of initiator may need to be synthesized). Developing a simple and versatile strategy for initiator immobilization applicable to a wide range of surfaces is desirable for extending the technological application of SIP. Some progress has been made towards this goal with the development of the macroinitiator approach to SIP [18]. Macroinitiators for SIP are polymers containing many initiator sites (e.g. as pendant groups on repeat unit) which are adsorbed to the surface. Whereas small molecule initiators are typically immobilised using surface-specific covalent-bond-forming reactions, macroinitiators are immobilised using a large number of weaker non-covalent interactions (e.g. electrostatic attraction [19,20], hydrogen bonding [21] or hydrophobic interaction [5]) with cooperativity of the many binding sites ensuring overall strong attachment. Macroinitiators can be synthesized on a large scale allowing their application to high surface area substrates. Moreover, each macroinitiator can be applied to a broad class of surfaces (e.g. cationic macroinitiators can be applied to all anionic surfaces). However, they do not represent a truly universal initiator for all surfaces. The results presented here present a promising step in the development of such an initiator.
2142
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
Recently, Messersmith and co-workers have reported that the bio-inspired polymerization of dopamine allows the formation of adherent polydopamine films on a very wide range of substrates [22]. The polymerization system is simple e a dilute aqueous solution of dopamine is buffered to pH 8.5 by TRIS, and substrates are coated by simply being dipped into the polymerization solution. Cross-linked hydrophilic films of polydopamine, which is structurally similar to natural eumelanin pigments [23], can be formed in this way with thicknesses of up to tens of nanometers. This research was inspired by the adhesive protein secreted by mussels [24e26], which have been shown to attach to virtually all types of inorganic and organic surfaces, even to conventionally non-adhesive materials such as poly(tetrafluoroethylene) (PTFE), in marine environments. This adhesive versatility was proposed to lie in the co-existence of catechol and amine groups in the protein structure [22]. Dopamine is a small-molecule chemical that contains both functionalities, and thus was chosen by Messersmith as the monomer for a simple synthetic polymer mimic for mussel adhesives. The broad applicability of polydopamine coating has led us to explore this technology as a platform for producing initiator for SIP that can be applied to a broad range of substrates. The mechanism of dopamine polymerization is proposed to occur in a manner reminiscent of melanin formation [23,27,28], involving oxidation of catechol to quinone, resulting in the formation of dopaminechrome, which can isomerize into 5,6-dihydroxylindole [29]. Dopaminechromes and 5,6-dihydroxylindoles are then proposed to react with themselves or with each other to form the adherent polydopamine film. There is much recent interest in the technological application of these coatings. For example, polydopamine has been deposited on porous membranes to improve hydrophilicity [30], to control pervaporation [31], or as an adhesion layer in multi-layer membranes [32]. Films grown on Nafion membranes enhanced methanol barrier properties for fuel-cell applications [33], and deposition of polydopamine inside layer-bylayer polyelectrolyte multilayers improved the mechanical properties allowing free-standing film to be produced [34]. Polydopamine capsules have also been synthesized for future biomedical applications [35] and have been shown to have interesting uptake/release behaviour [36]. On planar surfaces, semiconductor nanocrystals have been embedded in a polydopamine coating for sensor applications [37]. Polydopamine coatings have also been explored as a versatile platform for secondary reactions such as electroless deposition of silver [38,39], in-situ gold nanoparticle synthesis [40] and biomolecule immobilisation [41]. Of particular interest for this study are reports by Messersmith and co-workers [42e45] of a monomeric dopamine-based initiator for surface-initiated ATRP. This initiator is a functionalised dopamine which binds to metal oxide surfaces using specific interaction with the catechol group. Since these initiators do not polymerise into a polydopamine, polymer brush chains grown from these initiators may suffer stability problems since surface attachment is only through a single non-covalent interaction. Although a useful addition to the “arsenal” of initiators available for SIP, this system does not take advantage of the cross-linking inherent in a film made from a polydopamine. We expect this cross-linking to improve the robustness of the coating even on surfaces where it is only weakly adhered (since the only route to degrafting the polymer chains is to delaminate the whole initiator layer). However, the presence of polar functional groups in a polydopamine films (hydroxyls and amines) suggests that good adhesion should be expected for many substrates. In this work, we report a polydopamine-based initiator for surface-initiated ATRP systems, which we hope could be applied to a wide range of substrates. These initiator films were synthesized by functionalising a fraction of dopamine monomers with
2-bromoisobutyryl bromide (BIBB) before polymerization, introducing 2-bromoisobutyrate ATRP initiator groups into the polydopamine films. Grafted polymer brushes could then be grown from these polydopamine initiators using surface-initiated ARGET ATRP. Since this work was conducted, other reports of the use of polydopamine films for surface-initiated ATRP have appeared, in which the polydopamine is modified after deposition. This approach either requires the synthesis of nucleophilic thiol-functional initiators [46] or exposing the sample to solvents which may damage some substrates (e.g. polymers) [47,48]. Our approach uses only commercial reagents and avoids the need for two discrete surface-modification steps, introducing initiators in a one-step process consisting of simple immersion into the modified dopamine solution. 2. Experimental 2.1. Materials All reagents and solvents were purchased from SigmaeAldrich (Gillingham, UK) or Fisher Scientific (Loughborough, UK) and used as received. Water was deionized and obtained from a Elga Option 4 system. Silicon wafers (‹100› orientation, boron doped, 1e100 U cm, polished one side) were purchased from Compart Technology Ltd. (Peterborough, UK). Aluminium (Al) was commercial foil for food use, containing 98.6% aluminium. Glass slides were 0.13e0.17 mm thick microscope cover glass. Steel chips were 1.15 mm thick stainless steel. Polystyrene chips are 0.70 mm thick and cut from polystyrene petri dishes (Fisher Scientific, UK). 2.2. Synthesis 2.2.1. Polydopamine deposition Silicon wafers pieces (w1 cm2) were cleaned and rendered hydrophilic by first washing with acetone, methanol and water, and then immersed for 15 min in a mixture of ammonia solution (28 mL, 35% wt), hydrogen peroxide solution (28 mL, 30% wt) and water (142 mL) at 75 C. Wafers were removed, rinsed thoroughly with water and dried under a stream of nitrogen. Al foil and glass slides were used after washing with acetone and methanol. Polystyrene chips were used after washing with methanol alone. Stainless steel chips were cleaned and rendered hydrophilic by first washing with acetone, methanol and water, and then treated in a UV-Ozone reactor for 1000 s. The substrate was then immersed in a solution of 3-hydroxytyramine hydrochloride (dopamine hydrochloride, 200 mg, 1.05 mmol) and tris(hydroxymethyl) aminomethane (TRIS, 120 mg, 1.0 mmol) in deionised water (100 mL) in a glass dish open to the air. This solution was continuously magnetically stirred at a speed of 200 rpm at room temperature. The clear solution initially became pink coloured, then darkened as stirring continued, finally becoming black after around 10 min. Polydopamine-coated substrates were removed from the solution after various deposition times, washed with deionised water and dried with compressed air. Polydopamine particles were obtained from a polymerization solution prepared as above which was continuously magnetically stirred at a speed of 200 rpm at room temperature for 24 h while open to the air. Particles were collected by filtration and dried in a vacuum oven. 2.2.2. Polydopamine initiator deposition Dopamine (400 mg, 2.10 mmol) was placed in a flask which was degassed by purging with dry N2 for 5 min. To this flask was added N,N0 -dimethylformamide (DMF, 20 mL), 2-bromoisobutyryl
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
bromide (BIBB, 0.13 mL, 1.05 mmol), and triethylamine (0.15 mL, 1.05 mmol) under dry N2. After stirring under dry N2 at room temperature for 3 h, this mixture was transferred to a glass dish to which tris(hydroxymethyl) aminomethane (TRIS) (480 mg, 4.0 mmol) and deionised water (100 mL) were added. Cleaned silicon wafer sections (w1 cm2), polystyrene chips, Al foil and stainless steel chips were then immersed in this new mixture which was continuously magnetically stirred at a speed of 200 rpm while open to the air. Polydopamine initiator-coated substrates were removed from the solution after various deposition times washed with deionised water and dried with compressed air. Samples with varying BIBB:dopamine ratios were prepared as above, but with the BIBB and triethylamine concentrations being altered according to the desired ratio. Polydopamine initiator particles were obtained from a polymerization solution prepared as above which was continuously magnetically stirred at a speed of 200 rpm at room temperature for 24 h while open to the air. Particles were collected by filtration and dried in a vacuum oven. 2.2.3. Surface-initiated ARGET ATRP from polydopamine initiators A solution of methyl methacrylate (MMA, 20 mL, 18.72 g, 187.0 mmol) in methanol (20 mL) was degassed by bubbling through dry N2 for 15 min in a flask sealed with a septum. To this solution was added copper (II) bromide (4 mg, 0.018 mmol), (þ)-sodium L-ascorbate (354 mg, 1.787 mmol) and 2,20 -dipyridyl (6 mg, 0.038 mmol). To dissolve all solids, the mixture was magnetically stirred for 5 min while degassing continued, giving a dark brown solution. In each glass tube was placed an initiatorcoated silicon wafer section (w1 cm2) and the tube was sealed with a septum. The tube was degassed by purging with dry N2 for 1 min and the monomer solution was then syringed over the wafer. After the polymerization was allowed to proceed at room temperature for various times the wafer was removed and washed sequentially with methanol and water, and dried under a compressed air stream. ARGET ATRP growth of MMA from polydopamine initiatorcoated steel chips and Al foil was conducted as above except that the methanol (20 mL) was replaced with a mixture of methanol (20 mL) and deionized water (10 mL). ARGET ATRP growth of HEMA from polydopamine initiator-coated steel chips and polystyrene chips was also conducted as above except that methyl methacrylate (20 mL, 18.72 g, 187.0 mmol) and methanol (20 mL) in the polymerization solution was replaced with 2-hydroxyethyl methacrylate (20 mL, 20.74 g, 164.8 mmol) and a solvent mixture of methanol (10 mL) and deionized water (10 mL).
2143
Scheme 1. Schematic illustration of the polymerization of dopamine hydrochloride into adherent polydopamine coatings on silicon wafers.
quality. Thus, all fitting assumed no optical absorption for polydopamine in order to simplify modelling. For PMMA grown from polydopamine initiator films a four-layer model was used, consisting of silicon, silicon dioxide (2 nm), polydopamine (n ¼ 1.6, 58 nm), PMMA (thickness fitted). The polydopamine layer thickness was measured before PMMA growth, and the PMMA refractive index was assumed to be n ¼ 1.5 at all wavelengths. XPS spectra were obtained using a VG Scientific Escalab 5 Mark II X-ray photoelectron spectrophotometer, with an unmonochromated Al source. A Shirley background was used, and peaks were fitted using XPSPEAK 4.1. FTIR spectra were obtained using a Shimadzu FTIR-8400S Fourier transform infrared spectrophotometer. Measurements on silicon samples were taken in transmittance mode using a blank silicon wafer as a background. The measurements of other substrates (steel, Al foil and PS) samples were taken in reflectance mode (i.e. ATR-FTIR). 3. Results and discussion 3.1. Deposition of unmodified polydopamine Before investigating the synthesis of polydopamine initiators, the deposition of unmodified polydopamine on cleaned silicon wafers was attempted (Scheme 1). By simply immersing a cleaned silicon wafer in aqueous solution of dopamine and TRIS, a 47 nm polydopamine layer was grown in 24 h (Fig. 1). This is consistent with the results reported by Messersmith and co-workers [22] who measured a 50 nm polydopamine film after 24 h. continuing the deposition beyond 24 h leads to a continued increase in thickness, but at a reduced growth rate. This is presumably due to dopamine monomer consumption (since polymerization is also occurring in solution). To demonstrate the broader applicability of polydopamine deposition, films were grown on four other substrates: glass slides,
2.3. Characterisation Ellipsometric measurements were conducted using a phasemodulated spectroscopic ellipsometer (Uvisel, Jobin Yvon) at 10 nm intervals from 500 nm to 700 nm at an angle of incidence of 70 . This wavelength range was chosen to avoid the strong optical absorption displayed by polydopamine at shorter wavelengths [28]. Modelling was conducted using the WVASE32 software package (J. A. Woolam Co., USA). For polydopamine and polydopamine initiator films a three-layer model was used, consisting of silicon, silicon dioxide (2 nm) and polydopamine (thickness fitted). Softwaresupplied refractive indices were used for silicon and silicon dioxide, and the refractive index of polydopamine was assumed to be n ¼ 1.6 at all wavelengths with no optical absorption (i.e. the non-zero imaginary part of the refractive index, k ¼ 0). Adding an absorption with k ¼ 0.02 at all wavelengths, as measured previously for polydopamine at 589 nm [28], produced only very small changes in the fitted thickness (<1%), and negligible improvement in fit
Fig. 1. Ellipsometric thickness against deposition time for unmodified (:) and BIBBmodified (-) polydopamine. Error bars are from ellipsometric fitting and are smaller than 0.5 nm where not shown.
2144
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
Al foil, stainless steel and polystyrene (PS) chips. On these substrates, a striking colour change is observed with increasing dopamine deposition time (Fig. 2). Polydopamine has a strong broad-band UVevisible absorption [28] and the darkening colour with growth time is indicative of polydopamine thickness. To assess the strength of adhesion between the polydopamine and various substrates, a tape peel test was conducted with polydopamine-coated silicon, steel, glass, aluminium, polystyrene (PS) and low-density polyethylene (LDPE). On every substrate, polydopamine remained on the surface after peeling (as judged by the remaining brown colour). Every sample did show some transfer of
material from the sample to the tape, however. This is likely to be a layer of weakly-bonded polydopamine particles formed in solution during the coating process. On all substrates, a second peel test (after removal of weakly-bonded material) did not remove the underlying polydopamine film, demonstrating good adhesion to the substrate. Ellipsometric measurements on polydopamine layers of various thicknesses (Supporting Information, Fig S1) confirm strong adhesion. In all cases, no significant change in polydopamine thickness was observed after tape peeling. Although a small amount of weakly bound particles was always transferred to the tape, this change is not detected by ellipsometry. Dopamine
Fig. 2. Photographs showing various polydopamine-coated substrates. A: Glass and Al foil pieces with increasing polydopamine deposition time. Polydopamine has a strong broadband UV/Vis absorption, giving rise to a brown colour increasing in intensity with deposited film thickness. B: Steel and polystyrene pieces with increasing polydopamine deposition time. C: A polydopamine-coated polystyrene sample is subjected to a tape peel test. Comparison with uncoated polystyrene reveals that a brown colour is still present on the peeled section, indicating that most polydopamine is not removed. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
monomer contains polar amine and hydroxyl groups, so strong adhesion to polar surfaces (e.g. silica and alumina) is not surprising [49]. More surprising is the adhesion to non-polar surfaces (PS and LDPE), demonstrating the broad applicability of polydopamine as a surface coating. 3.2. Deposition of ATRP-initiator-modified polydopamine The next step in the synthesis of a polydopamine-based initiator is the introduction of 2-bromoisobutyrate ATRP-initiating groups into the film. Initially, we sought to react the secondary amines and catechol hydroxyl groups of pre-formed polydopamine films with BIBB under basic catalysis. However, after reaction these films showed no capacity to initiate SI-ATRP, and in some cases the films were completely destroyed during the reaction. It is likely that our initial assumption that the polydopamine films would contain much nucleophilic functionality, and react with an electrophilic acid bromide, was incorrect. It has been shown that these films are in fact electrophilic, existing in the quinone form in basic solution, allowing reaction with nucleophilic amines and thiols [22,41]. Indeed, a recently reported alternative approach to ATRP-initiating films relies on the reaction of polydopamime with a thiol-functional ATRP initiator [46]. Nucleophilic hydroxyl groups can be incorporated into polydopamine-like films, but an alternative monomer containing an additional hydroxyl group (norepinephrine) must be used [50]. Since this work was completed, two studies have been reported by Wang and co-workers in which pre-formed polydopamine has been successfully reacted with BIBB, contrary to other reports of the electrophilic nature of polydopamine [47,48]. This could be a consequence of differences in the precise composition of the BIBB esterification mix (e.g. amount of excess base) or sample history (e.g. degree of exposure to oxygen). This suggests that with a thorough investigation, this reaction could be successfully applied. However, in light of reports of the electrophilic reactions of polydopamine, we chose to abandon this route. In our alternative approach, the dopamine monomer is reacted with BIBB under base catalysis before polymerization into polydopamine (Scheme 2). This reaction is carried out in polar aprotic solvent (DMF) to allow the dilution into water for the polymerization, and in the absence of air to prevent premature polymerization. After this reaction was allowed to proceed for 3 h, the mixture was diluted into water with TRIS buffer and the polydopamine polymerization allowed to proceed in the presence of air, as for unmodified dopamine. A greater amount of TRIS was required to increase the pH and initiate polymerization, presumably since
2145
unreacted acid bromide (BIBB) forms carboxylic acid upon addition of water, decreasing the pH. Our route to ATRP-initiating dopamine (pre-deposition dopamine modification) has advantages over those recently reported by others requiring post-deposition modification of polydopamine. Unlike the work of Hu et al. [46] our route does not require the synthesis and purification of a nucleophilic thiol-functional initiator. The work of Wang and co-workers [47,48] requires exposing the substrate to a THF solution of BIBB, obviously making this route incompatible with solvent-sensitive substrates such as polystyrene. Using our route the substrate is only exposed to a solvent of dilute DMF in water, which will be compatible with a much wider range of substrates (including polystyrene, as we show below). It is not known if the BIBB will preferentially react with the hydroxyl or amine groups on the dopamine (our attempts to use 13 C NMR to elucidate the reaction site have so far not been successful). However, in either case it is likely that this modification of dopamine will hinder or completely prevent polymerization. Thus, the mole ratio of BIBB to dopamine was chosen as 0.5, statistically leaving much of the dopamine unmodified so that polymerization could still proceed, but allowing modified dopamine to be incorporated into the polydopamine layer. It appears that the modification of a fraction of the dopamine with BIBB does not greatly affect the polymerization, with film growth proceeding at a similar rate to that for unmodified dopamine (Fig. 1). Substrates coated with polydopamine initiator display the same characteristic brown colour as unmodified polydopamine coatings, increasing in intensity with deposition time (Supporting Information, Fig S2). In order to assess the incorporation of 2-bromoisobutyrate groups into the polydopamine film, XPS was conducted on both unmodified and BIBB-modified films on silicon wafers (Fig. 3). An unmodified film deposited for 24 h showed strong C 1 s, O 1 s and N 1 s signals, consistent with polydopamine. No silicon peaks are observed (e.g. Si 2 s at 155 eV or Si 2p at 103 eV), indicating that the polydopamine film is both continuous (with no defects revealing the underlying wafer) and thicker than the w10 nm sampling depth of XPS, in agreement with ellipsometry (47 nm thickness). From relative peak areas, the surface composition (in atom %) was measured, giving the C:O:N atomic ratio from XPS as 75.4: 17.5: 7.1, close to the calculated ratio for dopamine monomer C:O:N ¼ 8:2:1 ¼ 72.7: 18.2: 9.1. In the BIBB-modified polydopamine film, the presence of a weak Br 3d signal (1.0 atom%) in the XPS spectrum confirms the incorporation of the initiator groups. The Br/N ratio allows us to estimate
Scheme 2. Schematic illustration of the reaction of dopamine with 2-bromoisobutyryl bromide (BIBB) and the subsequent deposition of initiator-functionalised polydopamine onto silicon wafers. PMMA polymer brushes are then grown by surface-initiated ARGET ATRP. The exact structure of BIBB-modified polydopamine is not known; in particular, BIBBfunctionalisation may also occur at the amine position.
2146
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
Fig. 3. XPS spectra for polydopamine and polydopamine initiator coatings with a deposition time of 24 h on silicon wafers. Data for BIBB-modified polydopamine has been vertically offset for clarity. Inset: Expanded 0e200 eV region.
the BIBB/dopamine ratio. For this sample, Br/N ¼ 0.20 compared to a target of 0.5, determined by the reaction stoichiometry. This apparent low yield could be due to incomplete reaction or poor incorporation of BIBB-modified dopamine monomer into the polymer. Ester formation at the catechol hydroxyl group will prevent oxidation into the quinone form, whereas amide formation may hinder cyclisation into the indole; in both cases, polymerization may be hindered. A small amount of silicon is also observed in the XPS spectrum. Since the ellipsometric thickness of this film was 45 nm (greater than the sampling depth of XPS), this may indicate some small defects in the coating, revealing the underlying wafer. XPS narrow-scan spectra (C 1 s, N 1 s, O 1 s and Br 3d) for both unmodified and initiator-modified polydopamine are included in the Supporting Information (Fig S3). Transmission FTIR on free polydopamine and polydopamine initiator powders (precipitated from solution during the growth of the coatings) provides further confirmation of the incorporation of initiator groups (Fig. 4). These powders are polymerised from exactly the same solution as the coatings and so should very similar compositions; the free powders were used to achieve sufficiently strong absorbances. The spectrum of unmodified polydopamine is consistent with the results reported by Fei et al. [40], with peaks at w1250 cm1, w1500 cm1 and w1600 cm1 deriving from the aromatic rings in the polymer and a broad peak at w3300 cm1 from hydroxyls, amines and water absorbed in this hydrophilic material. The FTIR spectrum of the BIBB-modified polydopamine displays two significant differences from unmodified polydopamine (Fig. 4, inset). A weak absorbance at around 1710 cm1 could be indicative of esters formed by reaction of BIBB with catechol hydroxyl groups, whereas a shoulder at w1650 cm1 on the aromatic peak at 1600 cm1 suggests the presence of amides formed by reaction of BIBB with the dopamine primary amine. In both spectra, peaks at w2300 cm1 are from atmospheric carbon dioxide, and peaks at 2800e3000 cm1 are attributed to CeH bonds. The increased intensity of the CeH peak in the initiator sample is consistent with the incorporation of initiator sites (which contain six CeH bonds each).
3.3. Surface-initiated ARGET ATRP from initiator-modified polydopamine on silicon wafers The ability to grow surface-initiated polymer from the BIBBmodified polydopamine provides the most important test of the presence of ATRP-initiating groups. No other functionality in the polydopamine film is known to initiate ATRP. When surface-initiated ARGET ATRP was attempted from unmodified polydopamine on silicon wafers, we found no increase in film thickness after 24 h. Indeed, a 15% decrease in polydopamine thickness was measured, indicating some degradation of the coating by the polymerization solution or desorption of weakly-bonded polydopamine at the surface. XPS measurements support the conclusion that unmodified polydopamine is not capable of initiating polymerization (Supporting Information, Figs S4 & S5). In contrast, BIBB-modified polydopamine films on silicon wafers were used to successfully grow surface-initiated PMMA brushes using ARGET ATRP (Scheme 2). Using methanol as a solvent, 72 nm of PMMA was grown from a 58 nm thick BIBB-modified polydopamine initiator (giving a total thickness of 130 nm) in 24 h (Fig. 5). At 72 h, a PMMA thickness of 239 nm was measured (296 nm total thickness). Such large PMMA layer thicknesses indicate that the grafting density is sufficiently high to be in the brush regime. In unpublished results, we have found that surface-initiated ARGET ATRP of PMMA from more conventional polyelectrolyte macroinitiators [18] can produce similarly thick brush layers, with thicknesses of over 350 nm possible after 24 h1 XPS of PMMA grown from polydopamine initiator for 24 h displays a surface atomic composition consistent with pure PMMA
1 Although ARGET ATRP is significantly less sensitive to oxygen than conventional ATRP [14], we have found that care is still needed for polymerization over long times. Improper sealing or deoxygenation of the reaction vessel or the use of nondeoxygenated solvent leads to a reduction in polymerization rate at long times, although the polymerization still proceeds. This slowing is presumably due to consumption of the reservoir of reducing agent (and so a reduction in Cu(I) concentration) before the end of the polymerization.
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
2147
Fig. 4. FTIR spectra for the polydopamine initiator and polydopamine particles with a polymerization time of 24 h. Data for BIBB-modified polydopamine has been vertically offset for clarity. Inset: Expanded 1400e2000 cm1 region, highlighting differences between the spectra.
(see Supporting Information, Fig S6). In particular, the absence of an N 1s signal from the underlying polydopamine layer suggests the PMMA layer is thick and continuous. Transmission FTIR of the same sample reveals features characteristic of both the polydopamine initiator and the PMMA brushes (see Supporting Information, Fig S7), consistent with the presence of both layers with a similar thickness. Initially, a mole ratio of BIBB : dopamine of 0.5:1 was chosen as a compromise between introducing sufficient initiator sites to produce high grafting density for surface-initiated polymerization, and inhibition of the dopamine polymerisation mechanism by functionalizing too many hydroxyl or amine sites. A subsequent study of the effect of BIBB : dopamine ratio (Supporting Information, Fig S8) indeed revealed that of a number of ratios tested (0.25:1, 0.5:1, 1:1, 2:1), 0.5:1 did indeed produce the thickest films after surface-initiated ARGET ATRP of MMA. For very high BIBB:dopamine (3:1), the dopamine polymerization was completely inhibited. To confirm strong grafting of these PMMA brushes to the surface, a silicon wafer sample consisting of 62 4 nm grown from a 25 nm polydopamine initiator was subjected to Soxhlet extraction with THF (a solvent for PMMA) for 9 h. After this aggressive washing procedure, the measured PMMA thickness was almost unchanged at 58 4 nm, confirming the covalent tethering of the PMMA to the polydopamine initiator, and the strong adhesion of the initiator to the surface.
consistent with initiator deposition, including the presence of bromine atoms from the 2-bromoisobutyrate initiator group. PMMA was then grown by ARGET ATRP for 24 h from steel and Al foil substrates. Although ellipsometry is not possible on these substrates, ATR-FTIR spectra were consistent with a thick PMMA coating, including the strong ester peak at 1720 cm1 confirming the successful surface-initiated polymerization (Fig. 6). Peaks from CeH stretching (2800e3050 cm1), ester C-O stretching (1140 cm1) and CeH bending (1430e1490 cm1) are also consistent with the presence of PMMA. XPS on these substrates was also consistent with a continuous, thick PMMA film (see Supporting Information, Fig S9). PMMA could not be grown from polystyrene, since the monomer solution caused swelling of the substrate. However, a more hydrophilic monomer, 2-hydroxyethyl methacrylate (HEMA), could be used to grow PHEMA polymer films from polystyrene. ATR-FTIR on this sample (Fig. 7) shows peaks originating from both the PHEMA and the underlying PS (notably
3.4. Surface-initiated ARGET ATRP from initiator-modified polydopamine on other substrates Our main motivation for the development of polydopaminebased ATRP initiators is substrate-independence; the same initiator can be applied to a wide range of substrates. To demonstrate this principle, polydopamine initiator was deposited for 24 h on steel, Al foil and polystyrene substrates. On all substrates, XPS analyses were
Fig. 5. Ellipsometric PMMA thickness against growth time for SI-ARGET ATRP of MMA from silicon wafers coated with a 58 5 nm polydopamine initiator layer.
2148
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
Fig. 6. ATR-FTIR spectra for PMMA grown for 24 h from initiator-functionalised polydopamine on steel (top) and Al foil (middle) surfaces. The spectrum of bulk PMMA (bottom) is included for comparison. For clarity, data for steel and Al foil has been vertically offset and the data for bulk PMMA has been vertically scaled.
aromatic CeH stretching at 3000e3100 cm1 and aromatic ring stretching modes at 1600 cm1, 1450 cm1 and 1490 cm1). Growing the PHEMA from the same initiator on steel allows the peaks originating from the PHEMA to be more clearly identified (although the steel can contribute some peaks in this range due to the thin oxide layer and adsorbed water [51], the peaks from the thick polymer film dominate the spectrum), including the strong ester peak at 1720 cm1. Peaks from CeH stretching (2800e3050 cm1), ester CeO stretching (1140 cm1), alcohol CeO stretching (1070 cm1) and OeH stretching (3000e3600 cm1) are also consistent with the presence of PHEMA. This grafted PHEMA spectrum is very similar to that of bulk PHEMA, with the exception of peaks relating to water present in the bulk PHEMA structure (notably 1660 cm1, HeOeH bend). On both steel and polystyrene, XPS analysis is consistent with a continuous, thick PHEMA film (see Supporting Information, Fig S9). XPS narrow-scan analysis of the C 1 s peak for polystyrene substrates confirms the changes at the surface after polydopamine initiator coating and subsequent growth of PHEMA (Fig. 8). Of particular interest is the pronounced sp2 C]O peak from the polydopamine initiator coating. This could be partially attributed to the oxidised quinone form of polydopamine [30], but could also be indicative of ester or amide groups formed during reaction with BIBB. Fig. 8. XPS narrow-scan spectra and fitted peaks for bare polystyrene substrates (top), after coating with polydopamine initiator for 24 h (middle), and after surface-initiated polymerization of PHEMA brushes for 24 h (bottom). Peaks have been assigned based on prior literature [30,52]. To compensate for sample charging, the fitted peak with the lowest binding energy has been corrected to 285 eV.
4. Conclusions
Fig. 7. ATR-FTIR spectra for PHEMA grown for 24 h from initiator-functionalised polydopamine on polystyrene (top) and steel (middle) surfaces.. The spectrum of bulk PHEMA (bottom) is included for comparison. For clarity, data for steel and polystyrene has been vertically offset and the data for bulk PHEMA has been vertically scaled.
In summary, we have demonstrated that polydopamine could be deposited on various surfaces by self-polymerization of dopamine. Bromoester initiating groups for ATRP were incorporated into polydopamine coatings by reacting a fraction of the dopamine monomer with 2-bromoisobutyryl bromide (BIBB) before polymerization. This modification did not appear to significantly impede coating deposition, although by XPS the incorporation of initiator groups into the film appeared to be lower than targeted. PMMA and PHEMA brushes could then be grown from these initiators by surface-initiated ARGET ATRP, with brush thicknesses of up to 239 nm possible after 72 h growth. A range of substrates could be coated, including metals and hydrophobic polystyrene,
B. Zhu, S. Edmondson / Polymer 52 (2011) 2141e2149
showing the versatility of this initiator system. We are currently investigating a broader range of substrates in order to assess this system as a possible “universal initiator” for surface-initiated ATRP. Appendix. Supplementary Data Supplementary data related to this article can be found online at doi:10.1016/j.polymer.2011.03.027. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]
Zhao B. Progr Polymer Sci 2000;25(5):677e710. Edmondson S, Osborne VL, Huck WTS. Chem Soc Rev 2004;33(1):14e22. Tsujii Y, Ohno K, Yamamoto S, Goto A, Fukuda T. Adv Polymer Sci 2006;197:1e45. Lindqvist J, Nyström D, Ostmark E, Antoni P, Carlmark A, Johansson M, et al. Biomacromolecules 2008;9(8):2139e45. Mizutani A, Kikuchi A, Yamato M, Kanazawa H, Okano T. Biomaterials 2008; 29(13):2073e81. Cheng Z, Zhu X, Shi ZL, Neoh KG, Kang ET. Ind Eng Chem Res 2005;44(18): 7098e104. Huang J, Murata H, Koepsel RR, Russell AJ, Matyjaszewski K. Biomacromolecules 2007;8(5):1396e9. Li L, Ke Z, Yan G, Wu J. Polymer Int 2008;57(11):1275e80. Yuan SJ, Pehkonen SO, Ting YP, Neoh KG, Kang ET. ACS Appl Mater Interfaces 2009;1(3):640e52. Yuan SJ, Xu FJ, Pehkonen SO, Ting YP, Neoh KG, Kang ET. Biotechnol Bioeng 2009;103(2):268e81. Chen Y, Liu D, Deng Q, He X, Wang X. J Polymer Sci Polymer Chem 2006; 44(11):3434e43. Bozukova D, Pagnoulle C, De Pauw-Gillet M-C, Ruth N, Jérôme R, Jérôme C. Langmuir 2008;24(13):6649e58. Li L, Yan G, Wu J. J Appl Polymer Sci 2009;111(4):1942e6. Matyjaszewski K, Dong H, Jakubowski W, Pietrasik J, Kusumo A. Langmuir 2007;23(8):4528e31. Hansson S, Ostmark E, Carlmark A, Malmström E. ACS Appl Mater Interfaces 2009;1(11):2651e9. Cao L, Kruk M. Polym Chem 2010;1(1):97e101. Matyjaszewski K, Jakubowski W, Min K, Tang W, Huang J, Braunecker WA, et al. Proc Natl Acad Sci USA 2006;103(42):15309e14. Edmondson S, Armes SP. Polymer Int 2009;58(3):307e16. Edmondson S, Vo C-D, Armes SP, Unali G-F. Macromolecules 2007;40(15):5271e8. Edmondson S, Vo C-D, Armes SP, Unali G-F, Weir MP. Langmuir 2008;24(14): 7208e15. Stöhr T, Rühe J. Macromolecules 2000;33(12):4501e11. Lee H, Dellatore SM, Miller WM, Messersmith PB. Science 2007;318(5849): 426e30.
2149
[23] d’Ischia M, Napolitano A, Pezzella A, Meredith P, Sarna T. Angewandte Chemie International Edition 2009;48(22):3914e21. [24] Lin Q, Gourdon D, Sun C, Holten-Andersen N, Anderson TH, Waite JH, et al. Proc Natl Acad Sci USA 2007;104(10):3782e6. [25] Lee H, Scherer NF, Messersmith PB. Proc Natl Acad Sci USA 2006;103(35): 12999e3003. [26] Holten-Andersen N, Waite JH. J Dent Res 2008;87(8):701e9. [27] Park TJ, Kim J, Kim TK, Park HM, Choi SS. Bull Korean Chem Soc 2008;29(12): 2459e64. [28] Bernsmann F, Ponche A, Ringwald C, Hemmerle c J, Raya J, Bechinger B, et al. J Phys Chem C 2009;113(19):8234e42. [29] Li Y, Liu M, Xiang C, Xie Q, Yao S. Thin Solid Films 2006;497(1e2):270e8. [30] Xi Z-Y, Xu Y-Y, Zhu L-P, Wang Y, Zhu B-K. J Membr Sci 2009;327(1e2): 244e53. [31] Li B, Liu W, Jiang Z, Dong X, Wang B, Zhong Y. Langmuir 2009;25(13): 7368e74. [32] Chen J, Chen X, Yin X, Ma J, Jiang Z. J Membr Sci 2009;344(1e2):136e43. [33] Wang J, Xiao L, Zhao Y, Wu H, Jiang Z, Hou W. J Power Sources 2009;192(2): 336e43. [34] Bernsmann F, Richert L, Senger B, Lavalle P, Voegel J-C, Schaaf P, et al. Soft Matter 2008;4(8):1621e4. [35] Postma A, Yan Y, Wang Y, Zelikin AN, Tjipto E, Caruso F. Chem Mater 2009; 21(14):3042e4. [36] Yu B, Wang DA, Ye Q, Zhou F, Liu W. Chem Commun 2009;44:6789e91. [37] Zhang Y, Wang H, Nie J, Zhou H, Shen G, Yu R. Electrochem Commun 2009; 11(10):1936e9. [38] Liao Y, Wang Y, Feng X, Wang W, Xu F, Zhang L. Mater Chem Phys 2010; 121(3):534e40. [39] Liao Y, Cao B, Wang W-C, Zhang L, Wu D, Jin R. Appl Surf Sci 2009;255(19): 8207e12. [40] Fei B, Qian B, Yang Z, Wang R, Liu WC, Mak CL, et al. Carbon 2008;46(13): 1795e7. [41] Lee H, Rho J, Messersmith PB. Adv Mater 2009;21(4):431e4. [42] Hamming L, Fan X, Messersmith P, Brinson L. Compos Sci Technol 2008;68(9): 2042e8. [43] Fan X, Lin L, Messersmith P. Compos Sci Technol 2006;66(9):1198e204. [44] Dalsin JL, Messersmith PB. Mater Today 2005;8(9):38e46. [45] Fan X, Lin L, Dalsin JL, Messersmith PB. J Am Chem Soc 2005;127(45): 15843e7. [46] Hu H, Yu B, Ye Q, Gu Y, Zhou F. Carbon 2010;48(8):2347e53. [47] Wang WC, Wang J, Liao Y, Zhang L, Cao B, Song G, et al. J Appl Polym Sci 2010; 117(1):534e41. [48] Li CY, Wang WC, Xu FJ, Zhang LQ, Yang WT. J Membr Sci 2011;367(1e2):7e13. [49] Ryu J, Ku SH, Lee H, Park CB. Adv Funct Mater 2010;20(13):2132e9. [50] Kang SM, Rho J, Choi IS, Messersmith PB, Lee H. J Am Chem Soc 2009;131(37): 13224e5. [51] Mueller HJ, Freeman D. Mater Charact 1995;35(2):113e26. [52] Beamson G, Briggs D. High resolution XPS of organic polymers. Chichester: John Wiley & Sons; 1992.