lable at ScienceDirect

Biomaterials 31 (2010) 532–540

Contents lists avai

Biomaterials

journal homepage: www.elsevier .com/locate/biomater ia ls

Tailoring the surface functionalities of titania nanotube arrays

Krasimir Vasilev b,c, Zihan Poh a, Krishna Kant a, Joseph Chan a, Andrew Michelmore b, Dusan Losic a,*

a University of South Australia, Ian Wark Research Institute, Mawson Lakes Campus, Mawson Lakes, Adelaide, SA 5095, Australiab University of South Australia, Mawson Institute, Mawson Lakes Campus, Mawson Lakes, Adelaide, SA 5095, Australiac University of South Australia, School of Advanced Manufacturing and Mechanical Engineering, Mawson Lakes Campus, Mawson Lakes, Adelaide, SA 5095, Australia

a r t i c l e i n f o

Article history:Received 14 July 2009Accepted 21 September 2009Available online 9 October 2009

Keywords:Titanium oxideTitania nanotubesSurface modificationsPlasma polymerizations

* Corresponding author. Tel.: þ61 8 8302 6862; faxE-mail address: dusan.losic@unisa.edu.au (D. Losic

0142-9612/$ – see front matter Crown Copyright � 2doi:10.1016/j.biomaterials.2009.09.074

a b s t r a c t

Nanotubular titanium oxide (TiO2) produced by self-ordering processes using electrochemical anodiza-tion have been extensively explored in recent years as a new biomaterial for implants, drug deliverysystems, cell growth, biosensors, immunoisolations, bioartificial organs and tissue engineering. Chemicalinertness is the main weakness of this material when placed in contact with biological systems andsurface modification is a possible solution of this problem. The aim of this study is to develop a flexibleand facile method for surface modification of TiO2 nanotubes to tailor new interfacial propertiesimportant in many biomedical applications. TiO2 nanotubes were prepared by electrochemical anod-ization of titanium foil using ethylene glycol: NH4F electrolyte (2% water and 0.3% NH4F). Plasma surfacemodification using allylamine (AA) as a precursor has been applied to generate a thin and chemicallyreactive polymer (AAPP) film rich in amine groups on top of the TiO2 nanotube surface. This initialpolymer film was used for further surface functionalization by attachment of desired molecules. Twomodification techniques were used to demonstrate the flexibility for building of new functionalities ontitania nanotube surface: electrostatic adsorption of poly(sodium styrenesulfonate) (PSS) as an exampleof layer-by-layer assembly (LbL), and covalent coupling of poly(ethylene glycol) (PEG) as an example ofcreating a protein-resistant surface. These approaches for tailoring the surface chemistry and wettabilityof TiO2 nanotubes offer considerable prospects for advancing their interfacial properties to improveexisting and develop new functional biomaterials for diverse biomedical applications.

Crown Copyright � 2009 Published by Elsevier Ltd. All rights reserved.

1. Introduction

Electrochemical self-ordering has been recognized as one of themost attractive synthetic approaches for fabrication of highlyordered nanoporous and nanotube materials [1,2]. The attractive-ness of this fabrication strategy arises from its simplicity, low cost,formation of highly-organized and uniform porous/nanotubestructures with controllable dimensions and the ability to beapplied on a variety of metals and semiconductors [2,3]. Poroussilicon, anodic porous alumina (AAO) and nanotubular titaniumoxide (TiO2) are the most popular examples of this technique, beingexploited in many areas such as optoelectronics, solar cells, energystorage, catalysis, sensors, molecular separation, templatesynthesis, immunoisolation, implants and drug delivery [3–8]. Inparticular, nanotubular TiO2 has been recognized as a promisingbiomaterial with proven biocompatibility, thermal stability andcorrosion resistance. In addition, the possibility of creating uniquenanotube architectures with high surface-to-volume ratio and

: 61 8 8302 3683.).

009 Published by Elsevier Ltd. All

controllable dimensions, makes this material desirable for a range ofbiomedical applications [9–12]. Since this new material has beenintroduced by Zwilling et al. [13], it has been the subject ofconsiderable research efforts for a number of applications such asimplants, tissue engineering, implantable drug delivery systems,artificial organs and other medical devices [14–17]. The TiO2

nanotube films have been extensively explored as adhesion andgrowth support platforms for bone and stem cells, for the preven-tion of bacterial adhesion, drug delivery and enhancing bloodclotting for control of haemorrhage [18–21]. Favourable bone cellgrowth, cell differentiation, and apatite-forming abilities wererecently demonstrated on implants with nanotubular titaniasurfaces demonstrating its promise for bone implants [22]. Recentlydevelopment in stem cell research demonstrated that titaniananotubes are able to differentiation of mesenchymal stem cellsinto specific cells (osteoblasts) [23,24]. Furthermore, theoutstanding photocatalytic properties of the nanotube surface wereapplied to photo-induced killing of cancer cells which suggestsa possible application of nanotubes for an anticancer treatment [25].

Among many of these biomedical applications of TiO2 nano-tubes, therapeutic bone and coronary drug eluting implants have

rights reserved.

K. Vasilev et al. / Biomaterials 31 (2010) 532–540 533

been recognized as one of the most promising [15]. However,previous studies clearly indicate that surface modification of TiO2

nanotubes is desirable for further improving their biocompatibilityvia introduction of new functionalities and properties [26–29]. It iswell known that surface modification with a bioactive organiclayer is critical for many biomedical applications such as reducingthe time for osseointegration, enhancing cell-material interaction,increasing cell proliferation, increasing implant life-time, pre-venting infections, increasing biofiltration selectivity, providingimmobilization of biosensing element or providing sustained andcontrolled drug release. While there is considerable research onsurface modification of similar materials such as porous siliconand porous alumina, surprisingly the surface functionalization ofTiO2 nanotubes has not been widely explored [30,31]. The abilityof new functionalities to considerably improve existing propertiesand offer new applications of this material is elegantly demon-strated in a recent report on modification of titania nanotubes bya hydrophobic organic layer facilitating controllable and sustaineddrug release [32].

The aim of our present work is to develop a method for tailoringthe surface functionality of TiO2 nanotubular surfaces to advancetheir surface properties relevant to a wide range of biomedicalapplications. Our surface modification strategy (shown in Fig. 1)involves first deposition of a plasma polymer layer rich in reactiveamine groups (Fig. 1b) which facilitates subsequent surface graftingof various biologically relevant molecules via chemical or physicalmeans as demonstrated by covalent grafting of polyethylene glycol(PEG) and electrostatic deposition of polystyrene sulphonate (PSS).

Fig. 1. Scheme for the tailoring of the surface functionalities on TiO2 nanotube films: a) bareelectrostatic adsorption of poly(sodium styrenesulfonate (PSS) on AAPP (3) and d) covalent

Plasma polymerization is a convenient method for surface modi-fication because ultrathin polymer films of various functionalitiescan be deposited on any type of surfaces without the need ofsurface pre-modification [33–35]. Although plasma polymerizationhas been widely used for surface engineering of a number ofbiomaterials, including titania implants, it has not been applied yetfor surface modifications of TiO2 nanotubes [33]. To generate anamine rich chemically active layer on the nanotube surface (Fig. 1b)we have selected allylamine (AA) as a precursor which facilitatesdeposition of an allylamine plasma polymer (AAPP) film rich inamine functional groups. AAPP confers biocompatibility andenables subsequent grafting of macromolecules with desiredfunctionalities that are important for protein adsorption and celladhesion [36–38]. Two surface grafting strategies have beenapplied to demonstrate the flexibility for providing new function-alities on AAPP modified TiO2 nanotubes by various methods. Thefirst is electrostatic adsorption of poly(sodium styrenesulfonate)(PSS) (Fig. 1c) to demonstrate immobilization of charged moleculesand application of layer-by-layer assembly (LbL) [39]. The secondmodification approach is covalent coupling of poly(ethylene glycol)(PEG) which is known to provide non-fouling surfaces if graftedwith sufficient surface density (Fig. 1d) [38]. The compounds forsurface modification used in this work have been specificallyselected because their biological relevance. AAPP films have beendemonstrated to beneficial for cell adhesion in proliferation[35,36,40]. PSS similar to other sulphonated polymers is known tohave antithrombogenic properties [41,42]. PEG was used because ofits ability to resist protein adsorption if grafted with sufficient

TiO2 nanotube arrays, b) plasma polymerization of n-allylamine polymer (AAPP) (1), c)attachment of polyethylene glycol (PEG) on APPP (3).

K. Vasilev et al. / Biomaterials 31 (2010) 532–540534

surface density [43–45]. Characterization of the morphology,chemical composition, and interfacial properties of functionalizedTiO2 nanotube surface was performed by scanning electronmicroscopy (SEM), X-ray photoelectron spectroscopy (XPS), andcontact angle (CA) measurements.

2. Experimental section

2.1. Materials

Titanium foil (99.6%) thickness 0.25 mm was supplied by Sigma Aldrich(Australia). Ethylene glycol, NH4F, poly(sodium styrenesulfonate) (MW 700 000),PEG-dialdehyde (MW 3400), NaCNBH4, potassium sulphate were obtained fromSigma Aldrich (Australia) and used as supplied. Silicon wafers, p-type boron doped,orientation C100D were obtained from Virginia Semiconductors (USA). High purityultra-pure Milli-Q grade (18.2 MU cm) with a final filtering step through a 0.22 mmfilter was used.

2.2. Fabrication of TiO2 nanotubes

A titanium foil, was mechanically polished and cleaned by sonication in acetonefor 30 min prior to anodization. Two anodization steps were performed usinga specially designed electrochemical cell and computer controlled power supply(Agilent). In the first anodization step a constant voltage of 100 V was applied for 2 hin NH4F/ethyleneglycol electrolyte (3% water and 0.3% NH4F) at room temperature of20 �C [11,46,47]. The resultant layer of anodic TiO2 nanotube film was removed(mechanically followed by sonication) leaving the nanotextured titanium surface forthe second anodization. The second anodization step was performed using the samecondition (100 V) as the first anodization but with different times (20 min to 3 h).The voltage/current, voltage-time, and current-time signals were adjusted andcontinuously recorded during the anodization process by software (Labview,National Instruments) [47]. To prepare the self-supporting TiO2 nanotube layer, theresultant nanotube layer was removed after anodization by sonication in methanolsolution followed by drying with a stream of nitrogen. To achieve in situ poreopening during anodization, and to fabricate TiO2 nanotube membranes withthrough-hole morphology, in the final stages of the anodization process the voltagewas gradually decreased to a low voltage such as 10 V, over 0.5–1 min [46].

2.3. Plasma polymerization

Prior to the plasma deposition, prepared TiO2 nanotubes substrates (self-sup-porting nanotube layer) were repeatedly washed in ethanol followed by Milli-Qwater and dried with a nitrogen jet. Plasma polymerization was carried out ina custom-built reactor described elsewhere [48] using a commercial 13.56 MHzplasma generator (Advanced Energy). The depositions were carried out in anatmosphere of pure allylamine (AA) at a pressure of 0.22 Torr. An input powersetting of 20 W was used in combination with a matching network to minimizereflected power. The deposition of AA under these conditions was carried out for30 s, 60 s, 90 s, 120 s, 200 s, 250 s and 300 s. The deposition rate was about0.5 nm s�1 determined by ellipsometry on polished silicon wafers.

2.4. Functionalization by PSS adsorption

Adsorption of PSS on AAPP modified TiO2 nanotubes and AAPP modified siliconwafers used as control were performed using a concentration of 0.2 monomol/l,where monomol corresponds to the molecular weight of the monomer [49]. The pHof the solution was adjusted to 2 by HCl (0.1 M). The time of adsorption was 1 hfollowed by thorough washing with Milli-Q water.

2.5. Functionalization by PEG grafting

PEG-dialdehyde (Mw 3400) was grafted on the AAPP modified TiO2 nanotubelayer and AAPP modified silicon wafer (control) as previously described [50]. Briefly,PEG-dialdehyde (2 mg/ml) was dissolved in phosphate buffer, pH 7.4 containing0.6 M K2SO4, but without NaCNBH4. Freshly prepared surfaces of amine functionalgroups were immersed in the solution of PEG-dialdehyde. The grafting reaction wascarried out at 60 �C overnight followed by thorough washing with Milli-Q water.

2.6. Scanning electron microscopy

Structural characterization of TiO2 nanotube surfaces before and after AAPPmodification was performed using a field emission scanning electron microscope(SEM) (Philips XL 30). The samples were cut into small pieces, mounted on a holderwith double sided conductive tape and coated with a layer of platinum a few nm thick.Images with a range of scan sizes at normal incidence and at a 30� angle were acquiredfrom the top, the bottom surface and cross-sections.

2.7. X-ray photoelectron spectroscopy

XPS analysis was used to determine the surface composition of AAPP, AAPP-PSSand AAPP-PEG modified TiO2 nanotubes. The XPS spectra were recorded on a SpecsSAGE XPS spectrometer using Mg Ka radiation source (hn¼ 1253.6 eV) operating at10 kV and 20 mA. The elements present in the sample surface were identified froma survey spectrum recorded over the energy range 0–1000 eV at pass energy of100 eV and a resolution of 0.5 eV. The areas under selected photoelectron peaks inthe spectrum were used to calculate the percentage atomic concentrations. High-resolution (0.1 eV) spectra were then recorded for pertinent photoelectron peaks ata pass energy of 20 eV to identify the chemical state of each element. All thebinding energies (BEs) were referenced to the C1s neutral carbon peak at 285 eV,to compensate for the effect of surface charging. The analysis area was circularwith a diameter of 1 mm. The processing and curve-fitting of the high-resolutionspectra was performed using CasaXPS� software. XPS of Si wafer with matchingmodifications was used as control substrate.

2.8. Contact angle measurements

Static (sessile drop) water contact angles (CA) on AAPP, AAPP-PSS and AAPP-PEGfunctionalized TiO2 nanotubes and corresponding control substrates (Si) weredetermined with a custom-built contact angle apparatus. Samples were fixed toa glass slide and a droplet of Milli-Q water was applied to the surface. Images of eachdrop on the surface were recorded using a horizontal digital microscope. CAmeasurements were determined from images of droplets using drop shape analysissoftware (Image J, NIH, USA). The mean value of the contact angles was calculatedfrom at least 5 individual measurements taken at different locations on the exam-ined substrates.

3. Results and discussion

Several forms of TiO2 nanotubes, including a nanotube layer onthe bulk titanium (foil), self-supporting nanotube arrays andnanotube membranes were fabricated using electrochemicalanodization (NH4F/ethyleneglycol electrolyte) to show the versa-tility of this fabrication protocol. The morphology of these struc-tures were characterized by SEM and summarized in Fig. 2. Ananotube layer formed on a titanium foil substrate with a diameterof 1 cm is shown in Fig. 2a (inset). When the fabricated layer is thickenough it can be easily detached from the underlying titanium toform self-supporting nanotube layer shown in Fig. 2a. The cross-sectional SEM image of this layer shows perfectly straight anddensely packed array of TiO2 nanotubes across the whole structure(Fig. 2a). The thickness of the layer was about 20 mm (Fig. 2a), but byselecting the anodization time (from 20 min to 8 h) it is possible toadjust the thickness ranging from <5 mm to >150 mm. SEM imagesof the top nanotube surface (Fig. 2b and d) show porous rather thannanotube structures, with the pore diameters about 140� 20 nm.The pore diameters can be controlled by varying the anodizationpotential, and titania nanotubes with smaller pore diameters(<100 nm) were fabricated by anodization at 60 V and 80 V [46].Higher resolution cross-section images (Fig. 2c, d) show these poresare actually the top of densely packed nanotubes. When thethin top porous layer is removed, a slightly separated nanotubestructure is visible (Fig. 2c). In this work we were focused onmodification of the top of the nanotube surface where poremorphology is more dominant, which was the reason why the term‘‘pore diameter’’ rather than ‘‘nanotube diameter’’ is used. Thebottom surface of the TiO2 nanotube layer after detachment fromthe Ti substrate (Fig. 2f) shows the nanotube cells with irregularhexagonal shape which represents the end of the nanotubesformed at the oxide/metal interface. Fig. 2g shows the bottomsurface of self-supporting TiO2 nanotube membranes withthrough-hole morphology fabricated by a method describedpreviously by our group [46]. The diameter of these pores on thebottom is significantly smaller (60–80 nm) than diameter of thepores on the top, which suggests a slightly conical internal struc-ture. Results presented in Fig. 2 show typical nanotube structuresfrom several forms (at Ti surface, self-supporting layers andmembranes) obtained by anodization from Ti foil. There is no

K. Vasilev et al. / Biomaterials 31 (2010) 532–540 535

limitation to obtain these forms on other titanium substratespotentially important for biomedical applications such as bulk,rods, needles, meshes, wires, thin films etc.

Plasma polymerization is a technique which has been well-established in the last decades as a clean, simple and flexiblemethod for surface modification with the capability to alter surfaceproperties (both chemical and morphological) by providing a rangeof functional groups such as amine, carboxyl, hydroxyl, epoxy andaldehyde groups [33–35]. To demonstrate that plasma polymeri-zation is a suitable method for surface modifications of TiO2

nanotubes, we used allylamine as a common amine containingprecursor which has been previously shown to provide stable andbiocompatible amine functional surface for a number of applica-tions such as membranes, cell adhesion, DNA and protein adsorp-tions, enzyme immobilizations and biosensing [36–38,45]. Fig. 3shows a series of SEM images of TiO2 nanotubes (the top surface)modified by AAPP film using different deposition time (30–300 s).The rationale behind this experiment was to show the capability ofplasma deposition to induce morphological changes to the nano-tubes such as controlled reduction of their pore diameters. Whendeposition times in the range of 30 s and 250 s were used, the pore

Fig. 2. Series of SEM images of TiO2 nanotube structures fabricated by electrochemical ansupporting TiO2 nanotube layer and the entire structure (nanotube film) on the titianiummorphology obtained after removing of thin porous layer, d) the top surface with angle (30�)(30�) showing the nanotube structures detached from underlying titanium substrate, f) the bof self-supporting TiO2 nanotube membrane showing the opened nanotubes.

size was subsequently reduced whilst retaining their initial shapewithout closure of the pore (Fig. 3a–e). Fig. 4 shows the lineardependence of the pore reduction and deposition time which isconsistent with plasma modification of AAO pores previouslydescribed [51]. The complete closing of the pores can be achievedby longer deposition times (>300 s). Fig. 3f shows the stage whenpores are just covered by polymer film with the concave topog-raphy over the pore surfaces.

Typical cross-sectional SEM image of titania nanotube samplecoated with plasma polymer layer (deposition for 300 s) is shownin Supplementary section Fig. S1a which clearly shows the presenceof plasma polymer at the entrance of the nanotubes but the exactdept of plasma polymer deposition inside of nanotube is difficult todetermine because cleavage of the sample through the middle ofthe nanotudes is almost impossible. However, images show frac-tured nanotubes at depth of about 300–500 nm from the top (redarrows Fig. S1a) which are not closed. This is a clear indication ofrestricted deposition inside of nanotubes which is confirmed bycontrol experiment using AAO pores (Fig. S1b). Based on ourprevious work on plasma polymerization on AAO membranes weexpect that that the polymer coating tapers off gradually in

odization in NH4F/ethyleneglycol electrolyte. a) Typical cross-sectional image of self-substrate (inset). b–c) the top of nanotube surface showing porous and nanotubes

showing the nanopores and nanotube structures, e) the bottom surface taken by angleottom surface with more detail of topography of barrier layer and g) the bottom surface

K. Vasilev et al. / Biomaterials 31 (2010) 532–540536

thickness as one descends into the pores from the pore entrance[51]. A schematic representation of the pore reduction (nanotubeentry) from the first stage during plasma polymerization is shownin Fig. 4 (inset left). This corresponds to the formation of a thinconformal polymer film on the top of the nanotube and inside thenanotube entrances as observed in Fig. 3c. The scheme of the finalstage and the closing of pores by AAPP polymerization are pre-sented in Fig. 4 (inset right) and confirmed by corresponding SEMimage in Fig. 3f. The defect on AAPP film sporadically seen on fullycovered pores (Fig. 3f inset) clearly indicates that a thin polymerfilm was formed over the top of pores and nanotubes were notblocked deeply inside. The stability of AAPP films on Si wasconfirmed in our previous studies which make these modificationsreliable to use in aqueous and physiological environments [52].

Fig. 3. SEM images of the pore surface of TiO2 nanotube film a) before and after plasma pd) 90 s; e) 200 s and f) 300 s. Inset in f) with closed pores shows the details of defect in the fi1, and opened pores are marked as 2). The scale bar in insets is 200 nm.

The ability of this surface modification approach to fabricatethin polymer layers on nanotubular surfaces in a controlled fashionand provide fine tuning of pore dimeters with both structural andchemical modifications is a significant step towards extending theapplications of these unique nanomaterials. TiO2 nanotubes in theform of membranes are ideal for development of well-controlled,stable and uniform membranes capable of biomolecular separationof viruses, proteins or peptides and blocking of antibodies andcomplementary molecules from encapsulated xengenenic cells[11]. In addition to their photocatalytic and antibacterial properties,titania nanotubes membranes offer many advantages in compar-ison to existing biofiltration membranes based on polymers [11].TiO2 nanotubes films have been recently studied for drug deliveryapplications and stent implants [14,15,32]. The plasma surface

olymerization of n-allylamine (AAPP) using different deposition times b) 30 s; c) 60 s;lm confirming that nanotubes were not filled with polymer (closed pores are marked as

Fig. 4. The correlation of TiO2 nanotube pore diameters and polymer deposition times (from 60 s to 300 s) obtained during AAPP modification. Scheme of the plasma polymer onthe entry of nanotube structure at initial stage with reduced diameter and final stage (300 s deposition) with covered pores.

K. Vasilev et al. / Biomaterials 31 (2010) 532–540 537

modification demonstrated in this study has the potential to beapplied to drug eluted implant applications and considerablyimprove their biocompatibility and drug delivery performance. Thecovalent attachment of antibacterial molecules on plasma modifiedTiO2 nanotube implants can be easily implemented offering thepermanent defence against infection and implant surgery compli-cations. Furthermore, additional surface modification of nanotubebased implants such as drug eluting stents that contain antibiotics(e.g. gentamicin) will provide controlled and sustained drugrelease. With its considerable advantages of being simple, solventfree and biocompatible, plasma modification is ideal for thedevelopment of implantable drug delivery platforms which iscurrently under our investigation and will be presented separately.

XPS analysis was used to determine the chemical composition ofTiO2 nanotubes surface modified with approximately 50 nm thickfilms of AAPP (film thickness measured on silicon wafer). Thesurvey spectrum (Fig. 5a) reveals the presence of four elements:carbon, oxygen, nitrogen and titanium, with approximately 75.25%C, 11.6% N, 10.04% O and 3.11% Ti. The observed chemical compo-sition of the deposited AAPP was in agreement with a control Si

Fig. 5. a) XPS survey spectrum of allylamine plasma polymer (AAPP) on TiO2 na

modified substrate and previous studies of AA plasma polymers[36–38]. The presence of titanium in the survey spectrum can beattributed to pores in the TiO2 surfaces which remain open for thedeposition conditions used in this experiment. The high-resolutionC1s peak shown in Fig. 5b could be fitted with two componentscorresponding to aliphatic carbon at 285 eV and C–O/C–N compo-nent at 286.5 eV. We have not made an attempt to separate the C–Nand C–O components since they appear at very close bindingenergies. Suffice it to say that both species are probably present.The presence of oxygen is likely due the surface oxidation processesfollowing film exposure to ambient air since the samples weremeasured on the following day. The reactions of atmosphericoxygen with reactive species in the polymer film such as freeradicals resulting in the incorporation of oxygen, has been reportedby others [53]. The underlying TiO2 surface would also contribute tosome of the measured oxygen.

In order to demonstrate that active amine groups are present onthe nanotubes surface we conducted further surface modificationto prove that amine groups are available for both electrostatic andcovalent binding. The AAPP polymer surface is positively charged

notube film (50 nm film thickness) with b) high-resolution C1s spectrum.

K. Vasilev et al. / Biomaterials 31 (2010) 532–540538

after protonation of amine groups in aqueous solution, which canbe used to attract negatively charged molecules. In a first experi-ment PSS was adsorbed on AAPP modified surface for 60 min. TheXPS survey spectrum after adsorption is shown in Fig. 6a. Fourelements were detected on the surface. Quantification via inte-gration over the peak area showed 78.05% C, 6.95% N, 11.46% O and3.54% S. A high-resolution spectrum of the sulphur 2p peak is shownin Fig. 6b. The S2p peak is shifted to a higher binding energy of168.5 eV which is due to the association of sulphur with oxygen asSO3�group present in the PSS. One can notice that in addition to thatpresence of S2p peak in the XPS spectrum the amount of detectednitrogen significantly decreased to about 7% due to depositing anadditional layer of PSS and also the disappearance of the Ti2p peakwhich is probably due to a further decrease of the size of the pores.The LBL assembly of polyelectrolytes has emerged as a powerful andversatile, yet simple strategy to engineer surfaces with specificproperties. The resultant PSS layer can enable adsorption of oppo-sitely charged species (molecules, nanoparticles) and thus, via LbLassembly, generate complex structures on titania nanotubes.

In another example PEG was grafted covalently on AAPP modi-fied TiO2 nanotubes. XPS survey spectrum is presented in Fig. 6c after

Fig. 6. a) XPS survey spectrum of poly(sodium styrenesulfonate (PSS) electrostatically adsosurvey spectrum and high-resolution C 1s spectrum after covalent grafting of PEG on allyla

PEG grafting reveals the presence of four elements: carbon, oxygen,nitrogen and titanium with atomic concentrations of 76.42% C,10.6%N,13.19% O and 0.33% Ti. Upon grafting of PEG to the amine groups ofthe AAPP layer, a decrease in the nitrogen content was measured(11.60–10.06%) accompanied with a rise in the oxygen content(10.04–13.19%), which can be attributed to successful grafting ofa high density PEG layer. In addition the Ti2p almost disappeared asin the case of PEG. The high-resolution C1s spectrum is shown inFig. 6d. Curve fitting of the C1s spectrum showed a substantialincrease in the C–O component and a relative decrease in thealiphatic carbon peak at 285 eV compared with the C1s peak of theAAPP film confirming that PEG was successfully grafted to the AAPP.The intensity of the C–O component relative to the aliphatic carbonshows that PEG was grafted with a significant surface density whichis to be expected upon PEG grafting at cloud point conditions.Previous studies have proven that polyethylene glycol (PEG) modi-fied surfaces are promising for their ability to resist non-specificprotein adsorption, and to prevent colonization by anchorage-dependent cells [45,53,54]. As this is the first demonstration ofmodification of titania nanotubes with PEG, more studies will beconducted in the future for their practical applications.

rbed on AAPP/titania nanotube surface and b) S2p high-resolution spectrum. c–d) XPSmine plasma polymer (AAPP) modified TiO2 nanotube surface (120 s deposition).

Fig. 7. Sessile drop water contact angle measurements on TiO2 nanotube film (bare),TiO2 nanotube film modified by AAPP (120 s deposition), PSS-AAPP and PEG-AAPP andcontrol Si modified by AAPP, PSS-AAPP and PEG-AAPP.

K. Vasilev et al. / Biomaterials 31 (2010) 532–540 539

In biological systems the wettability of the surface plays animportant role in mediation of solute adsorption and cell adhesionand the alteration of wetting behaviour of TiO2 nanotubes is ofgreat importance for number of applications [55]. Hence, additionalbenefits of the surface modifications of TiO2 nanotube films fromthis work are anticipated by the tailoring of their wettability. Thewetting of bare and modified TiO2 nanotube layers was studied bymeasurement of sessile drop contact angle of a water droplet. CAresults are presented on Fig. 7. Contact angles w 0�, on bare TiO2

layer show its superhydrophilic properties in contrast to CA onplanar titania surface (w50�) [55]. The complete spreading of waterdroplets on the surface and inside of nanotubes is observed duringthe CA experiment. This behaviour is consistent with previousreports and in agreement with classical Wenzel and Cassie-Baxtertheory [55]. On the other hand, AAPP modified TiO2 nanotubeshowed a significant change of contact angle (70� 4�), displayingtransformation of their superhydrophilic properties into ratherhydrophobic. CA of AAPP on Si used as the control, showed a lowerCA (61�4�) which is probably due to the impact of the nanotubesurface topography. If plasma modification is performed usingmore hydrophobic molecules than AAPP, such as fluorocarbons, it isexpected to obtain nanotubes with superhydrophobic properties[55]. It was interesting to find that the TiO2 nanotube layers aftersurface modification with PSS and PEG retain their super-hydrophilic properties with contact angles <5�. Comparative CA ofPSS and PEG surfaces on silicon wafers used as control substratesshowed CA values of 43� 6� and 41�3� respectively (Fig. 7).Apparently if surface modification is carried out with ratherhydrophilic molecules (PEG and PSS in this case), this providessufficient wetting for a water droplet to freely spread and go insideof the nanotubes and thus not alter their initial CA. These finalresults prove the other benefits of surface functionalization of TiO2

nanotubes by altering their interfacial properties (hydrophilic tohydrophobic) which are of great importance for many biomedicalapplications.

4. Conclusions

The surface functionalization of TiO2 nanotube substratesfabricated by electrochemical anodization which involved initialgeneration of a chemically active layer by plasma polymerizationand subsequent immobilization of target molecules (in this casePSS and PEG) was successfully implemented. The use plasmapolymerization for modification of TiO2 nanotubes demonstratedfor the first time using allylamine as a precursor. This modification

step results in surfaces rich in amine groups. The versatility ofsurface modification via plasma polymerization for building of newfunctionalities and properties on titania nanotubes were demon-strated using electrostatic adsorption of poly(sodium styrenesul-fonate) and covalent coupling of poly(ethylene glycol). By theprocedures reported in this paper we can adjust the physico-chemical properties (surface reactivity and wettability) of TiO2

nanotubes according to interfacial requirements, via choice ofmonomer and attached functional group. In addition using surfacemodification via plasma polymerization allows control over porediameter at the surface from 140 nm to below 20 nm which isimportant feature for controlled drug release application. Thetailoring of surface functionalities on nanotube surfaces havepotential to significantly improve the properties of this attractivebiomaterial and promote the development of new biomedicaldevices such as drug eluting medical implants with multiplefunctions (orthopaedic implants, dental, coronary stents) providingan elegant route to prevent infection, control clotting, or decreaseinflammation of these implants.

Acknowledgments

The authors acknowledge the financial support of the AustralianResearch Council and the University of South Australia for thiswork.

Appendix

Figures with essential colour discrimination. Figs. 1–7 in thisarticle are difficult to interpret in black and white. The full colourimages can be found in the online version, at doi:10.1016/j.biomaterials.2009.09.074.

Appendix. Supplementary data

Supplementary data associated with this article can be found inthe online version, at doi:10.1016/j.biomaterials.2009.09.074.

References

[1] Chik H, Xu JM. Nanometric superlattices: non-lithographic fabrication, mate-rials, and prospects. Mater Sci Eng R 2004;43:103–38.

[2] Ghicov A, Schmuki P. Self-ordering electrochemistry: a review on growth andfunctionality of TiO2 nanotubes and other self-aligned MOx structures. ChemCommun 2009:2791–808.

[3] Mor GK, Varghese OK, Paulose M, Shankar K, Grimes CA. A review on highlyordered, vertically oriented TiO2 nanotube arrays: fabrication, materialproperties, and solar energy applications. Sol Energy Mater Sol Cells 2006;90:2011–75.

[4] Anglin EJ, Cheng LY, Freeman WR, Sailor MJ. Porous silicon in drug deliverydevices and materials. Adv Drug Deliv Rev 2008;60:1266–77.

[5] Diggle JW, Downie TC, Goulding CW. Anodic oxide films on aluminium. ChemRev 1969;69:365–405.

[6] Schmid G. Materials in nanoporous alumina. J Mater Chem 2002;12:1231–8.[7] Grimes CA. Synthesis and application of highly ordered arrays of TiO2 nano-

tubes. J Mater Chem 2007;17:451–1457.[8] Losic D, Shapter JG, Mitchell JG, Voelcker NH. Fabrication of gold nanorod

arrays by templating from porous alumina. Nanotechnology 2005;10:2275–81.[9] Macak JM, Tsuchiya H, Ghicov A, Yasuda K, Hahn R, Bauer S, et al. TiO2

nanotubes: self-organized electrochemical formation, properties and appli-cations. Curr Opin Solid State Mater Sci 2007;11:3–18.

[10] Popat KC, Leoni L, Grimes CA, Desai TA. Influence of engineered titaniananotubular surfaces on bone cells. Biomaterials 2007;28:3188–97.

[11] Paulose M, Peng L, Popat KC, Varghese OK, LaTempa TJ, Bao NZ, et al. Fabri-cation of mechanically robust, large area, polycrystalline nanotubular/porousTiO2 membranes. J Membr Sci 2008;319:199–205.

[12] Prakasam HE, Shankar K, Paulose M, Varghese OK, Grimes CA. A newbenchmark for TiO2 nanotube array growth by anodization. J Phys Chem C2007;111:7235–41.

[13] Zwilling V, Darque-Ceretti E, Boutry-Forveille A, David D, Perrin MY,Aucouturier M. Structure and physicochemistry of anodic oxide films ontitanium and TA6V alloy. Surf Interface Anal 1999;27:629–37.

K. Vasilev et al. / Biomaterials 31 (2010) 532–540540

[14] Losic D, Simovic S. Self-ordered nanopore and nanotube platforms for drugdelivery applications. Expert Opin Drug Deliv, in press, doi:10.1517/17425240903300857l.

[15] Popat KC, Eltgroth M, La Tempa TJ, Grimes CA, Desai TA. Titania nanotubes:a novel platform for drug-eluting coatings for medical implants. Small2007;3:1878–81.

[16] von Wilmowsky C, Bauer S, Lutz R, Meisel M, Neukam FW, Toyoshima T, et al.In vivo evaluation of anodic TiO2 nanotubes: an experimental study in the pig.J Biomed Mater Res Part B 2009;89B:165–71.

[17] Shrestha NK, Macak JM, Schmidt-Stein F, Hahn R, Mierke CT, Fabry B, et al.Magnetically guided titania nanotubes for site-selective photocatalysis anddrug release. Angew Chem Int Ed Engl 2009;48:969–72.

[18] Peng L, Eltgroth ML, LaTempa TJ, Grimes CA, Desai TA. The effect of TiO2nanotubes on endothelial function and smooth muscle proliferation. Bioma-terials 2009;30:1268–72.

[19] Peng LL, Mendelsohn AD, LaTempa TJ, Yoriya S, Grimes CA, Desai TA. Long-term small molecule and protein elution from TiO2 nanotubes. Nano Lett2009;9:1932–6.

[20] Brammer KS, Oh S, Gallagher JO, Jin S. Enhanced cellular mobility guided byTiO2 nanotube surfaces. Nano Lett 2008;8:786–93.

[21] Popat KC, Eltgroth M, LaTempa TJ, Grimes CA, Desai TA. Decreased Staphylo-coccus epidermis adhesion and increased osteoblast functionality on antibi-otic-loaded titania nanotubes. Biomaterials 2007;28:4880–8.

[22] Oh SH, Finones RR, Daraio C, Chen LH, Jin S. Growth of nanoscale hydroxy-apatite using chemically treated titanium oxide nanotubes. Biomaterials 2005;26:4938–43.

[23] Oh S, Brammer KS, Li YSJ, Teng D, Engler AJ, Chien S, et al. Stem cell fatedictated solely by altered nanotube dimension. PNAS 2009;106:2130–5.

[24] Park J, Bauer S, Schlegel KA, Neukam FW, von der Mark K, Schmuki P. TiO2nanotube surfaces: 15 nm – an optimal length scale of surface topography forcell adhesion and differentiation. Small 2009;5:666–71.

[25] Kalbacova M, Macak JM, Schmidt-Stein F, Mierke CT, Schmuki P. TiO2 nano-tubes: photocatalyst for cancer cell killing. Phys Status Solid RRL 2008;2:194–6.

[26] Nanci A, Wuest JD, Brunet P. Chemical modification of titanium surfacesfor covalent attachment of biological molecules. J Biomed Mater Res 1998;40:324–35.

[27] Huang N, Yang P, Leng YX, Chen JY, Sun H, Wang J, et al. Hemocompatibility oftitanium oxide films. Biomaterials 2003;24:2177–87.

[28] Yoshinari M, Oda Y, Kato T, Okuda K. Influence of surface modifications totitanium on antibacterial activity in vitro. Biomaterials 2001;22:2043–8.

[29] Swan EL, Popat KC, Desai TA. Peptide immobilized nanoporous aluminamembranes for enhanced osteoblast adhesion. Biomaterials 2005;26:1969–76.

[30] Bocking T, Kilian KA, Gaus K, Gooding JJ. Modifying porous silicon with self-assembled monolayers for biomedical applications: the influence of surfacecoverage on stability and biomolecule coupling. Adv Funct Mater 2008;18:3827–33.

[31] Md Jani AM, Anglin EJ, McInnes SJP, Losic D, Shapter JG, Voelcker NH. Fabri-cation of nanoporous anodic alumina membranes with layered surfacechemistry. Chem Commun 2009:3062–4.

[32] Song YY, Schmidt-Stein F, Bauer S, Schmuki P. Amphiphilic TiO2 nanotubearrays: an actively controllable drug delivery system. J Am Chem Soc 2009;131:4230–2.

[33] Siow KS, Britcher L, Kumar S, Griesser HJ. Plasma methods for the generationof chemically reactive surfaces for biomolecule immobilization and cellcolonization – a review. Plasma Process Polym 2006;3:392–418.

[34] Thierry B, Jasieniak M, de Smet L, Vasilev K, Griesser HJ. Reactive epoxy-functionalized thin films by a pulsed plasma polymerization process. Lang-muir 2008;24:10187–95.

[35] Robinson DE, Marson A, Short RD, Buttle DJ, Day AJ, Parry KL, et al. Surfacegradient of functional heparin. Adv Mater 2008;20:1166–70.

[36] Hamerli P, Weigel Th, Groth Th, Paul D. Surface properties of and cell adhesiononto allylamine-plasma-coated polyethylenterephtalates membranes. Bioma-terials 2003;24:3989–99.

[37] Dhayal M, Forder D, Parry K, Short RD, Barton D, Bradley JW. Tailored plasmasfor applications in the surface treatment of materials. Surf Coating Technol2003;162:294–300.

[38] Hook AL, Thissen H, Quinton J, Voelcker NH. Comparison of the binding modeof plasmid DNA to allylamine plasma polymer and poly(ethylene glycol)surfaces. Surf Sci 2008;602:1883–91.

[39] Decher G. Fuzzy nanoassemblies: toward layered polymeric multicomposites.Science 1997;277:1232–7.

[40] Finke B, Luethen F, Schroeder K, Mueller PD, Bergeniann C, Frant M, et al. Theeffect of positively charged plasma polymerization on initial osteoblastic focaladhesion on titanium surfaces. Biomaterials 2007;28:4521–34.

[41] Keogh JR, Wolf MF, Overend ME, Tang LP, Eaton TW. Biocompatibility ofsulphonated polyurethane surfaces. Biomaterials 1996;17:1987–94.

[42] Ito Y, Iguchi Y, Kashiwagi T, Imanishi Y. Synthesis and nonthrombogenicityof polyetherurethaneurea film grafted with poly(sodium vinyl sulfonate).J Biomed Mater Res 1991;25:1347–61.

[43] Ombotz WR, Guanghui W, Horbett TA, Hoffman AS. Protein adsorption topoly(ethylene oxide) surfaces. J Biomed Mater Res 1991;25:1547–62.

[44] Kingshott P, McArthur S, Thissen H, Castner DG, Griesser HJ. Ultrasensitiveprobing of the protein resistance of PEG surfaces by secondary ion massspectrometry. Biomaterials 2002;23:4775–85.

[45] Sharma S, Popat KC, Desai TA. Controlling nonspecific protein interactions insilicon biomicrosystems with nanostructured poly(ethylene glycol) films.Langmuir 2002;18:8728–31.

[46] Kant K, Losic D. A simple approach for synthesis of TiO2 nanotubes withthrough-hole morphology. Phys Status Solidi RRL 2009;3:139–41.

[47] Losic D, Lillo M, Losic Jr D. Porous alumina with shaped pore geometries andcomplex pore architectures fabricated by cyclic anodization. Small 2009;5:1392–7.

[48] Griesser HJ. Small-scale reactor for plasma processing or moving substrate.Vacuum 1989;39:485–8.

[49] Vasilev K, Knoll W, Kreiter M. Fluorescence intensities of chromophores infront of a thin metal film. J Chem Phys 2004;120:3439–45.

[50] Kingshott P, Thissen H, Griesser HJ. Effects of cloud-point grafting, chainlength, and density of PEG layers on competitive adsorption of ocular proteins.Biomaterials 2002;23:334–9.

[51] Losic D, Cole MA, Dollmann B, Vasilev K, Griesser HJ. Surface modification ofnanoporous alumina membranes by plasma polymerization. Nanotechnology2008;19:245704.

[52] Vasilev K, Britcher L, Casanal A, Griesser HJ. Solvent-induced porosity inultrathin amine plasma polymer coatings. J Phys Chem B 2008;112:10915–21.

[53] Muir BW, Tarasova A, Gengenbach TR, Menzies DJ, Meagher L, Rovere F, et al.Characterization of low-fouling ethylene glycol containing plasma polymerfilms. Langmuir 2008;24:3828–35.

[54] Popat KC, Mor G, Grimes C, Desai TA. Poly (ethylene glycol) grafted nano-porous alumina membranes. J Membr Sci 2004;243:97–106.

[55] Balaur E, Macak JM, Taveira L, Schmuki P. Tailoring the wettability of TiO2nanotube layers. Electrochem Commun 2005;7:1066–70.