lable at ScienceDirect

Vacuum 122 (2015) 360e368

Contents lists avai

Vacuum

journal homepage: www.elsevier .com/locate/vacuum

Biological behaviour of thin films consisting of Au nanoparticlesdispersed in a TiO2 dielectric matrix

J. Borges a, b, c, D. Costa a, d, E. Antunes a, C. Lopes a, e, M.S. Rodrigues a, e, M. Apreutesei f,E. Alves g, N.P. Barradas h, P. Pedrosa a, b, i, C. Moura a, L. Cunha a, T. Polcar c, F. Vaz a, b, *,P. Sampaio d

a Centro de Física, Universidade do Minho, Campus de Gualtar, 4710 e 057 Braga, Portugalb SEG-CEMUC, Mechanical Engineering Department, University of Coimbra, 3030-788 Coimbra, Portugalc Department of Control Engineering, Faculty of Electrical Engineering, Czech Technical University in Prague, Technick�a 2, Prague 6, Czech Republicd Centro de Biologia Molecular e Ambiental (CBMA), Universidade do Minho, Campus de Gualtar, 4710 e 057 Braga, Portugale Instituto Pedro Nunes, Laborat�orio de Ensaios, Desgaste e Materiais, Rua Pedro Nunes, 3030-199 Coimbra, Portugalf MATEIS Laboratory-INSA de Lyon, 21 Avenue Jean Capelle, 69621 Villeurbanne Cedex, Franceg Instituto de Plasmas e Fus~ao Nuclear, Instituto Superior T�ecnico, Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisboa, Portugalh Centro de Ciencias e Tecnologias Nucleares, Instituto Superior T�ecnico, Universidade de, Lisboa, E.N. 10 (km 139,7), 2695-066 Bobadela LRS, Portugali Universidade do Porto, Faculdade de Engenharia, Departamento de Engenharia Metalúrgica e de Materiais, Rua Dr. Roberto Frias, s/n, 4200-465 Porto,Portugal

a r t i c l e i n f o

Article history:Received 30 October 2014Received in revised form27 March 2015Accepted 31 March 2015Available online 11 April 2015

Keywords:Thin filmsTitanium dioxideEmbedded nanoparticlesBiological applicationsProtein surface interactionMicrobial cell viability

* Corresponding author. Centro de Física, UniversGualtar, 4710 e 057 Braga, Portugal. Tel.: þ351 253 6

E-mail address: fvaz@fisica.uminho.pt (F. Vaz).

http://dx.doi.org/10.1016/j.vacuum.2015.03.0360042-207X/© 2015 Elsevier Ltd. All rights reserved.

a b s t r a c t

In this work it was studied the possible use of thin films, composed of Au nanoparticles (NPs) embeddedin a TiO2 matrix, in biological applications, by evaluating their interaction with a well-known protein,Bovine Serum Albumin (BSA), as well as with microbial cells (Candida albicans). The films were producedby one-step reactive DC magnetron sputtering followed by heat-treatment. The samples revealed acomposition of 8.3 at.% of Au and a stoichiometric TiO2 matrix. The annealing promoted grain size in-crease of the Au NPs from 3 nm (at 300 �C) to 7 nm (at 500 �C) and a progressive crystallization of theTiO2 matrix to anatase. A broad localized surface plasmon resonance (LSPR) absorption band (l ¼ 580e720 nm) was clearly observed in the sample annealed at 500 �C, being less intense at 300 �C. Thebiological tests indicated that the BSA adhesion is dependent on surface nanostructure morphology,which in turn depends on the annealing temperature that changed the roughness and wettability of thefilms. The Au:TiO2 thin films also induced a significant change of the microbial cell membrane integrity,and ultimately the cell viability, which in turn affected the adhesion on its surface. The microstructuralchanges (structure, grain size and surface morphology) of the Au:TiO2 films promoted by heat-treatmentshaped the amount of BSA adhered and affected cell viability.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

Nanocomposite thin films containing noble metal nanoparticles(NPs), embedded in a host dielectric matrix, have attracted theinterest of scientists in both basic scientific research and in a widerange of technological applications. Beyond those of purely deco-rative purposes, as observed in the windows of the medieval

idade do Minho, Campus de01 118.

cathedrals and in Roman glasses (such as the well-known LycurgusCup from the 4th century [1,2]), plasmonic materials can be foundin many applications. Some examples are in the field of photovol-taics, pollutant-degradation materials, gas sensors, Surface-Enhanced Raman Scattering (SERS), photochromism, chemicalsurface activation, optical tweezers, among several others [2e12].Systems composed of noble metals (e.g. Au, Ag) and oxides (e.g.TiO2) have been also extensively studied due to their functionalproperties for antibacterial applications [13] and in the study ofbiological processes in live cells [14]. Moreover, the knowledgeacquired about plasmonic materials allowed building chemical andbiological sensors [15], which are based on the detection of changes

J. Borges et al. / Vacuum 122 (2015) 360e368 361

of their optical response due to presence of molecules and bio-layers. Nowadays, the area of optical biosensors, namely surfaceplasmon resonance (SPR) biosensors, is very successful in a widerange of fields, from fundamental biological studies to the detectionof chemical and biological species in different areas such as foodquality and safety analysis (e.g. pathogens, toxins), medicaldiagnostics (e.g. drugs) and environmental monitoring (e.g. pesti-cides, heavy elements) [16e18].

Recently, it has been realised that the sensor transductionmechanism of the localised (L)SPR-based nanosensors is analogousto that of SPR biosensors, which rely on gold-based sensors [19]. InLSPR sensors the optical response strongly depends on theclustering tendency of the nanoparticles, namely the clusters size,shape and distribution, the interaction between them, as well as onthe host matrix dielectric function itself [11,20e22].

In this work, Au:TiO2 thin films were produced by one-stepmagnetron sputtering deposition and post-deposition annealing(in order to promote the Au nanoparticles formation). Themicrostructure of the films (characterized in terms of crystallinestructure, grain size, phase composition and surface morphology)was firstly analysed and correlatedwith the optical responses of thefilms, as a function of the annealing temperature. Afterwards, andtaking into account envisaged applications using these plasmonicnanocomposite materials (for biosensing devices and anti-microbial surfaces), the interaction of the films with a well-known protein, Bovine Serum Albumin (BSA), as well as withmicrobial cells (Candida albicans) were analysed. On the one hand,the BSA was selected since protein adsorption is the first of acomplex series of events that regulates many phenomena at thenano-biointerface [23] and thus the understanding of how thenanocomposite (micros)structure influences protein adsorption isfundamental for many processes such as those related with affinitybiosensors [17]. On the other hand, the films were also tested interms of antimicrobial activity. Since TiO2 thin films do not presentantimicrobial activity against C. albicans without UV irradiation orwithout doping the surface with an antimicrobial agent (e.g. Ag[24]), the influence of the embedded Au nanoparticles on the TiO2host matrix was also evaluated.

2. Experimental details

2.1. Au:TiO2 thin films preparation and characterization

The Au:TiO2 thin films were deposited on glass lamellae ISO8037 (for the optical characterization, surface morphology analysisand biological-related tests) and silicon <100> (for chemical,structural and morphological analysis) substrates using reactive DCmagnetron sputtering. One-step deposition has been used to de-posit the host matrix (TiO2) with embedded gold atoms/clusters.The cathodewas composed of a titanium target (99.6% purity), withthree pellets of Au (each one with 15 mm2 of surface area), placedon the preferential erosion zone of the target. A DC current of 2 Awas applied. The substrates were placed on a rotating sampleholder, positioned at 70 mm from the target, with a constantrotation speed of 9 rpm, heated at 100 �C. A mixture of argon andoxygenwith a constant flow (60 sccm and 6 sccm, respectively) wasinjected into the deposition chamber and theworking pressurewaskept approximately constant for the deposition process at about0.5 Pa. After 90 min deposition, the samples were subjected to anin-air annealing procedure at several temperatures, ranging from200 to 800 �C. The heat-treatment experiments were conducted ina furnace at atmospheric pressure. The annealed temperatureschosen were reached rising the temperature at a rate of 5 �C/minuntil the desired temperature was attained, fixing this temperature

for 60 min (isothermal period). The samples cooled down freelybefore their removal from the furnace.

The chemical composition of the films and uniformity acrossthickness was investigated by Rutherford Backscattering Spec-trometry (RBS). RBS measurements were carried out with 2 MeV4He at an angle of incidence 0� in the small (RBS) chamber. Therewere three detectors in the chamber: standard at 140�, and twopin-diode detectors located symmetrical each other, both at 165�

(detector 3 on same side as standard detector 2). The RBS datawereanalysed with the IBA DataFurnace NDF v9.6a [25]. The crystallinestructure of the Au:TiO2 thin films was studied by in-situ X-rayDiffraction (XRD) during annealing, using a q-q Brucker D8 AdvanceSystem diffractometer, with a Cu-Ka radiation, in a Bragg-Brentanoconfiguration, equipped with a furnace and applying the sameannealing procedure as described above. The Au nanoparticles sizewas determined by fitting the Au diffraction peaks with a PearsonVII function, using the winfit software [26]. The morphology andthickness of the samples annealed at different annealing temper-atures were studied using a FEC-SEM Nova 200 NanoSEM scanningmicroscopy, in cross-sections of the Au:TiO2 films. The film surfacetopography was analyzed by Atomic Force Microscopy, using aMultiMode SPM microscope, controlled by a NanoScope IIIa fromDigital Instruments, in order to characterize the effect of theannealing in the surface roughness. Further analysis of the surfaceof the Au:TiO2 films was conducted by measuring its contact anglewith 3 mL of water, using a OCA 20 angle meter from Dataphysics.The optical response of the films (transmittance) were character-ized in the spectral range between 250 and 1800 nm using a Shi-madzu UV-3101 PC UVeviseNIR spectrophotometer.

2.2. Adhesion of proteins, Bovine Serum Albumin (BSA), to theAu:TiO2 films

The adhesion properties of Au:TiO2 thin films annealed atdifferent temperatures were evaluated by incubating with a solu-tion of Bovine Serum Albumin (BSA). After washing the films, withethanol and deionised water, they were incubated with a 3 mMaqueous solution of BSA for 24 h at a constant stirring of 80 rpm.After the incubation, a part of the BSA solution was storedat �20 �C, for posterior analysis, and the thin film was used foroptical analysis. The protein that remained in the solution wasquantified by using Bradford's protein assay [27], and the integrityof the proteins analysed by a 12% SDS-polyacrylamide gel electro-phoresis (SDS-PAGE) [28]. After the electrophoresis, the gel waswashed and stained with Brilliant Comassie Blue.

2.3. Adhesion of microorganisms (Candica albicans), to the Au:TiO2

films

To further study the adhesion behaviour of the samples, theAu:TiO2 nanocomposite thin films were incubated with yeast cells.The yeast C. albicans (strain SC5314) was obtained from the Centreof Molecular and Environmental Biology (CBMA), Department ofBiology, University of Minho. A pre-culture was prepared for eachindividual batch experiment. One colony of C. albicans strain waspicked and loop inoculated into a 125 ml Erlenmeyer flask, con-taining 20mL of Yeast Peptone Dextrose growthmedium (1% bacto-peptone (w/v), 1% yeast extract (w/v), 2% Glucose (w/v)), andincubated at 30 �C, for 12e15 h. On the next day, the cells weretransferred into different 250 ml Erlenmeyer flasks containing50 ml of YPD broth medium at an initial optical density (OD) of 0.1measured at a wavelength of 600 nm. The thin films (a square of26 � 26 mm2 deposited on glass), previously sterilized with 70%ethanol for 1 h and rinsed in sterile water, were placed on thebottom of the flask. Flasks were then shaken at 80 rpm at 30 �C, and

Fig. 1. Concentration profile of the different elements obtained by RBS.

J. Borges et al. / Vacuum 122 (2015) 360e368362

the OD monitored. OD measurements were made using a Spec-tronic 20 instrument at 600 nm and the background (turbidity dueto growth medium) was eliminated by blanking the instrumentwith medium alone. At the time points 12 h and 24 h after incu-bation, the number of viable cells was determined by flux cytom-etry after incubation with 1% (w/v) in propidium iodide (PI) (SigmaAldrich). At least 17,000 cells per condition were analysed by fluxcytometry. After incubation with the cells, the thin films werewashedwithwater, and surface images of their surfaces were takenwith an opticalmicroscope, before and after incubation. Once again,a severe washing step, vortexing the film with water during 5 min,was applied to evaluate if the cells were firmly attached, and againphotographs were taken using an optical microscope.

3. Results and discussion

3.1. Samples characterization

The composition profile analysis of the as-deposited sample wasstudied by RBS, showing that the thin film is homogeneous acrossits thickness, Fig. 1. A Au concentration (CAu) of about 8.3 at% and astoichiometric-like TiO2 (CO/CTi atomic ratio ~2) matrix weremeasured.

In order to study the structural modifications caused by theannealing process, the diffraction profiles, for the as-deposited and

Fig. 2. (a) XRD profiles of the Au:TiO2 films obtained in-

annealed samples, were analysed. The XRD profiles were obtainedin-situ during annealing (in-situ XRD) and are presented in Fig. 2.

For the as-deposited sample it is not possible to observe any Auor TiO2 crystallization peaks, revealing an amorphous-like struc-ture. However, it is possible to detect, firstly, the Au crystallizationat 200 �C and secondly, the crystallization of TiO2 in its anatasephase, starting at an annealing temperature of 300 �C. Thisbehaviour was already expected taking into account previousstudies about this system [4e6,20,29]. In fact, the set of processingconditions (deposition parameters and annealing conditions) usedto produce the samples has been selected based on previous works,where they were optimized in order to promote the formation of ananocomposite LSPR material [4,20,30]. The annealing processpromotes the crystallization of Au in the typical face centered cubic(fcc) structure [ICDD card No.04-0784] [4]. The presence of Au isevidenced by the detection of the (111) diffraction peak, located at2q ¼ 38.2� (T � 200 �C), as well as the (200) peak at 2q ¼ 44.4�

(T � 400 �C). The diffraction peak corresponding to the (111)orientation starts as a broad and low intensity peak at 200 �C and,as the annealing temperature increases, becomes sharper and withhigher intensity. The X-ray diffraction in the (200) crystalline planeof Au is only detected at 400 and 500 �C and, at both temperatures,it appears as a low intensity and broad peak. Regarding the trans-formations of the amorphous TiO2 matrix in its anatase phase, thepeak (101), located at 2q ¼ 25.3� [ICDD card No.78-1508], starts toappear at annealing temperatures of 300 �C. The peak intensityincreases with the increase of the annealing temperature, givingevidence for an enhanced crystallinity. By fitting the Pearson VIIfunction to the peak (111) and based on the integral breadth anal-ysis of the XRD patterns, it was possible to determine the Au NPsgrain size using the winfit software [26], Fig. 2(b), for annealingtemperatures higher than 200 �C. At this temperature, the peakfitting was quite difficult to carry out with coherent and reliableresults due to the low intensity signal of the diffraction pattern.With the heat-treatment the Au atoms start to aggregate in nano-particles with average grain size ranging from 3 nm (at 300 �C) to7 nm (at 500 �C).

According to the above set of results and the different structuralbehaviour that were observed, three samples were selected for thebiological tests with BSA and C. albicans; the as-deposited sample(reference) and the samples annealed at 300 �C and 500 �C. Theseare the samples where themajor differences were found in terms ofcrystallinity and phase composition. Therefore, a deeper analysis ofthe selected samples was performed in terms of morphology andsurface behaviour.

situ during annealing and (b) Au NP size evolution.

J. Borges et al. / Vacuum 122 (2015) 360e368 363

The cross-sectional images of the films were taken using SEM,Fig. 3(aec), which allowed the study of the morphological changesinduced by the heat-treatment. The results showed an average filmthickness of about 0.5 mm, corresponding to a deposition rate forthe as-deposited sample close to 5.5 nm.min�1. The as-depositedfilm revealed dense columnar-like growth but, as the annealingtemperature was increased, the microstructure became morevoided and the columns were increasingly difficult to distinguish.Particularly, the morphology of the sample annealed at 500 �C,Fig. 3(c), appears to be granular-voided, most likely due to the TiO2crystallization in its anatase phase.

Furthermore, the analysis of the composition and (micro)structure of the films strongly suggest the formation of a nano-composite material for annealing temperatures above 200 �C,

Fig. 3. SEM micrographs of the films cross-section.

where the nanoparticles of Au are dispersed in a TiO2 host matrix.Therefore, an absorption band in the visible range is expected as aresult of the formation of Au NPs [20]. Fig. 4 shows the opticaltransmittance of the as-deposited sample and samples annealed at300 and 500 �C.

The transmittance decreases with the increase of the annealingtemperature. For the as-deposited sample, as expected, aninterference-like behaviour with clear interference fringes wasfound, since the incident wavelength is of the same order ofmagnitude of the film thickness, and no LSPR absorption bandcould be found for this sample. This result is in agreement with theXRD analysis, which showed an amorphous structure with nodetectable Au nanoclusters. At 300 �C, the transmittance profileshifts to lower values and the interference fringes start to disappeardue to absorption by the Au NPs. This result is, again, in agreementwith the XRD analysis of the films that detected the presence ofcrystalline Au, although a low intensity peak could be observed.When the annealing temperature was further increased to 500 �C,the transmittance drops to zero between l ¼ 580 and 720 nm,corresponding to an absorption band characteristic of Au:TiO2 thinfilms due to LSPR [22]. This effect is also in straight correlationwiththe results of the XRD analysis, which evidenced the increase of AuNPs size, induced by the higher annealing temperature. Further-more, the broadening of the absorption band can be attributed tothe aggregation of the Au NPs into clusters, possibly of fractaldimension [22]. This result can be important in chemical sensingand single molecule detection, since the spectrally broad LSPR bandmight become necessary to fulfil the condition of resonantmatching to electronic transition in detected species [22].

Another important surface physical property of the films thatmust be evaluated is their roughness, since it plays an importantrole in regulating the interaction with biological systems, influ-encing, for example, the adsorption of proteins or the cell adhesionand growth [23,31]. The roughness of the films was studied usingatomic force microscopy and the wettability was evaluated bycontact angle measurements with water, Fig. 5. The roughnessvalues were estimated based on AFM analysis in various regions ofthe sample.

The heat-treatment affected the surface of the films, changingfrom a very smooth surface, in the case of the as-deposited andannealed at 300 �C samples, to a more rough surface in the sampleannealed at 500 �C. Fig. 5(aec) shows typical surface topographiesfor each one of the samples analysed. It was found an averageroughness change from 1.4 nm (as-deposited sample and annealedat 300 �C) towards 4.9 nm for the sample annealed at highertemperature (500 �C). This means that the heat treatment at 300 �Cdid not promoted significant morphological changes that couldaffect the roughness of the sample. Although some crystallization

Fig. 4. Transmittance spectra of the Au:TiO2 films and of the glass used as substrate.

Fig. 5. Topographic images of the Au:TiO2 films obtained by AFM (left) and corre-sponding water contact angles (right).

J. Borges et al. / Vacuum 122 (2015) 360e368364

of Au and anatase-TiO2 could be reported (Fig. 2) at 300 �C, themorphology of the film is not significantly affected, as also sug-gested by the cross-sectional SEM images (Fig. 3). The microstruc-tural changes induced by heating and the surface roughnessincrease, especially at 500 �C, also affected the surface hydropho-bicity. This surface characteristic was evaluated by measuring thecontact angle of water drops in the top of the film, Fig. 5. Thehydrophobicity changes from more hydrophobic surfaces, for thesamples as-deposited (contact angle, q ¼ 99.5�) and annealed at

Fig. 6. a) Percentage of BSA adsorbed at the surface of the Au:TiO2 thin films and b) SDS-elethin film. The initial solution before incubation was used as the control.

300 �C (q¼ 87.0�), to a less hydrophobic one in the film annealed at500 �C (q ¼ 57.5�).

It has been reported that increasing the annealing temperature(from 200 to 400 �C) of nanostructured titanium oxide (ns-TiOx)surfaces switches the surface from hydrophobic to hydrophilic [32].The same trend was observed in this study (Fig. 5); increasing theannealing temperature increased the wettability of the Au:TiO2thin films surface. In addition, the presence of Au nanoparticles didnot change this trend. Correlating the wettability with the rough-ness of the Au:TiO2 thin films it is possible to observe that the poorhydrophilicity of the as-deposited sample (AD) is reverted as theannealing temperature increases, but the roughness only increasesin the film annealed at 500 �C. According to the literature the in-crease inwettability without a significant increase in the roughnessof the nanostructured titanium oxide surface could be due to theremoval of organic contaminants and the increase of surfacehydroxyl groups (OH) bonded to undercoordinated Ti atoms [33].This could be the case of the film annealed at 300 �C where it wasobserved a wettability increase without a significant variation of itsroughness.

3.2. Adhesion of proteins and microorganisms to the Au:TiO2 thinfilms

3.2.1. Adhesion of Bovine Serum Albumin (BSA)It has been reported that ns-TiOx films interact with proteins via

non-specific interaction (electrostatic, hydrophobic and Van derWaals interactions) and via specific covalent bond between proteinacidic side chains and undercoordinated Ti atoms on the surface)[34]. The adsorption of proteins depends significantly on surfacenanostructure morphology, increasing with roughness [23]. How-ever, the same study has also demonstrated that the increase ofsurface roughness causes a decrease of protein binding to the filmand the higher adsorption observed in the rougher sample resultsfrom high surface protein density due to the formation of proteinmultilayers. Additionally, it has been referred that thermal treat-ment also influences the amount of adhered proteins due to therelease of atmospheric organic contaminants, freeing sites forproteins [34].

In order to evaluate protein adhesion to the Au:TiO2 films, thesamples were firstly incubated with BSA. After incubation with theselected thin films (as-deposited, 300 �C and 500 �C), the amount ofprotein remaining in the supernadants was quantified by theBradford's protein assay and the protein pattern analysed by SDS-PAGE. As shown in Fig. 6, only a small percentage of BSA was

ctrophoresis pattern of the BSA after 24h of incubation of the protein with the Au:TiO2

Fig. 7. Viability of C. albicans yeast cells during incubation with Au:TiO2 thin films; a) yeast cells were incubated with Au:TiO2 thin films after different annealing treatments andgrowth monitored at an OD of 600 nm over time. The control growth was assessed monitoring the growth of the yeast cells in the absence of a thin film. b) Flow cytometry analysisof the yeast cells membrane permeability after staining with PI. The percentage of PI positive cell in each condition is indicated.

J. Borges et al. / Vacuum 122 (2015) 360e368 365

adsorbed at the surface of the Au:TiO2 films. This low proteinadsorption observed in all samples may be due to i) the hydro-phobic character of the films surface, since BSA is a hydrophilicprotein; and ii) because the roughness is low, since it was estimatedas 1.4 nm, for the samples as-deposited and annealed at 300 �C, andas 4.9 nm for the sample annealed at 500 �C (Fig. 5). The latterconclusion is based on previous studies that have shown a greatprotein adsorption rate obtained with TiOx thin films with higherroughness values (between 15 and 30 nm) [23].

Comparing the three samples incubated with BSA, it wasobserved a higher amount of protein adhered to the film annealedat 300 �C. This results may be due to the creation of new free sitesfor proteins interaction due to thermal treatment, as previouslyreported [34]. The lower adhesion observed with Au:TiO2 thin filmannealed at 500 �Cmay be due to an increase in roughness that willincrease protein to protein interaction and formation of proteinscomplexes that can dissociate from the film surface, since the in-cubations were performed at constant shaking.

The SDS-PAGE electrophoresis of the supernadants of incuba-tion allowed the evaluation of the BSA band patterns and deter-mine if the interaction with the Au:TiO2 thin films had causedstructure modifications in the protein. The electrophoresis gelpattern, displayed in Fig. 6(b)), shows a strong band relative to theBSA and two other bands ascribed to dimmers and trimmers of theprotein [35]. However, patterns of the bands observed after incu-bation are very similar for all the samples, indicating no relevantstructure modifications after contact of the protein with thedifferent Au:TiO2 thin films. Furthermore, comparing the trans-mittance spectra before (Fig. 4) and after incubation (data notshown), no significant optical changes were found in the LSPR ab-sorption band.

3.2.2. Adhesion of microorganisms: C. albicansTiO2 surfaces have been described as antimicrobial due to its

photocatalytic activity after irradiation [13]. However, the antimi-crobial activity of TiO2 thin films without irradiation or without

Fig. 8. Microscopic photographs of the cells adhered to the different thin films after 12 and 24 h of incubation, before and after a vigorous washing step.

J. Borges et al. / Vacuum 122 (2015) 360e368366

doping the surface with of a known antimicrobial agent such assilver is low against C. albicans [24].

In this study, the adhesion ability of Au:TiO2 films was furtherevaluated studying the interaction between C. albicans cells and thethin films. The chosen yeast species was C. albicans due to theirability to adhere firmly to biotic and abiotic surfaces and to formbiofilms. Since the optical density (OD) is proportional to the num-ber of microorganisms in the culture medium, higher OD valuesindicate higher number of yeast cells in the culture. In Fig. 7a) it ispossible to observe that in the first 8 h of incubation the ODincreased exponentially, reaching a stationary phase. After 10 h ofincubation it decreased slightly, indicating that themicroorganismsare starting to die due to nutrient limitations. However, no relevantdifferences between the “control” (cells incubated in the absence ofa thinfilm) and thepresenceof thedifferent thinfilmswasobserved,indicating that the interaction of the cells with the Au:TiO2 filmsdidn't promote any alteration in the growth of the microbial cells.

In order to further ascertain if the interaction with the Au:TiO2samples affected the permeability of the yeast cell membrane, fluxcytometry analysis of the yeast cells was performed after incuba-tionwith propidium iodide (PI), a fluorochrome that only stain cellswith a permeable membrane [36,37]. The analyses of the yeast cellsstained with PI after 12 h of incubation showed that all cells (morethan 99%) were PI negative and thus with no problem in thepermeability of its membrane, being considered live cells. No sig-nificant differences were observed comparing the incubation withthe different thin films. However, the analysis of PI stained cellsafter 24 h of incubation, revealed significant different scenario,more than 80% of the cells incubated with the thin films were PIpositive and thus considered dead cells. Additionally, the percent-age of dead cells increases with the increase of the annealingtemperature. This result was a complete surprise since the evalu-ation of cell growth by OD (Fig. 7a) did not show a significant dif-ference in the number of the cells in each condition.

J. Borges et al. / Vacuum 122 (2015) 360e368 367

The discrepancy between the two methods, OD measurementsand flow cytometry, is due to the fact that bymeasuring cell growthOD, cells that might be dead are also counted since they representparticles in the solution, while by flow cytometry one can reallydistinguish the cells that have their membrane integrity compro-mised, considered dead or in a death process, and assess theirpercentage. These results highlight the need to evaluate the cellviability by more than one technique.

In order to observe yeast cells adhesion to the films surface,photographs were taken before and after the washing step (Fig. 8).A significant cell adhesion to the films was observed after 12 h ofincubation, with an apparent higher cell adhesion at the surface ofthe thin film annealed at 500 �C. After the washing step of the filmsa significant amount of cells were removed, but at the surface of thefilm annealed at 500 �C an apparent higher number of cells stillremained. After 24 h of incubation, a significant lower cell adhesionwas observed at the surface of the film annealed at 500 �C, whichwas particularly evident before the washing step. After washing, ahigh number of cells was removed from the surface of the as-deposited film and annealed at 300 �C. Adhesion at the surface ofthe film annealed at 500 �C was even lower after the washing step.Considering the adhesion of the cells at the surface of the as-deposited film it is possible to observe that the profile is similarto the adhesion observed at 12 h of incubation. However, at thesurface of the films annealed at 300 �C and 500 �C a significantlower adhesion was observed after washing. The cell adhesion atthe surface of the Au:TiO2 thin films seems to be inversely corre-lated with the annealing temperature, with a lower number of cellsremaining at the surface of the film annealed at 500 �C (Fig. 8). Thelower adhesion at the surface of the Au:TiO2 thin films seems to berelated with the viability of the cells, in which an increase in mi-crobial dead cells was observed at the surface of the film annealedat 500 �C.

Altogether, these results indicate that Au:TiO2 induce a signifi-cant alteration of the cell membrane integrity, and ultimately thecell viability, which in turn affects the adhesion on its surface. Thealteration induced on the film surface due annealing temperatureseems to slightly increase this deleterious effect on cell viability.

4. Conclusions

Surface physical properties have a relevant role in regulating theinteraction between thin films surface and biological systems and,in particular, surface nanoscale morphology profoundly influencescell adhesion and interaction. It has been reported in the literaturethat TiO2 films shows a high biocompatibility with different celllines, a good efficacy in protein immobilization, and the possibilityof tailoring their physical and chemical properties by controllingthe cluster assembling parameters. It is also known that biomaterialsurfaces in biological environments are rapidly coated by proteinsthat mediate the interaction between the biomaterial and cells.

In this work, thin films of Au:TiO2 (gold dispersed in a TiO2matrix) were deposited by one-step DC magnetron sputtering fol-lowed by a heat-treatment process in order to promote clusteringof Au nanoparticles. The major concern was to investigate the in-fluence of the microstructure (characterized in terms of grain size,crystallographic orientations, phase composition and surfacemorphology) of the films on the interaction with a well-knownprotein, Bovine Serum Albumin (BSA), as well as with microbialcells (C. albicans).

The composition analysis of the films revealed the formation ofa stoichiometric TiO2 matrix and 8.3 at% of gold (dispersed in thematrix). The characterization results revealed the formation ofsmall Au nanoparticles for annealing temperatures of 300 �C withgrain size of about 3 nm, increasing to about 7 nm at 500 �C. The Au

clustering process is also accompanied by the crystallization of theTiO2 matrix in its anatase phase, starting at 300 �C. The annealingtemperature promoted microstructural changes in what concernsthe type of growth of the films, changing from columnar (typical forsputtered films), in the case of the as-deposited sample, towards agranular-voided microstructure as the temperature increases. Thesurface morphology of the films was also affected by the heat-treatment, namely the roughness and wettability, increasing both.It was also demonstrated that the structural and morphologicalchanges induced by heat-treatment affects the optical response ofthe films and that the LSPR absorption band can be tuned bychoosing the appropriate annealing temperature. The broadeningof the LSPR absorption band, observed for the samples annealed at500 �C, can be useful for some application such as those involvingbiosensor devices however more research is needed in thisparticular point.

Although a more detailed analysis of the protein adhesion toAu:TiO2 must be performed, it was shown that a modification of thesurface characteristics of the thin films, by applying annealingtreatment, alterations on protein adhesion could be observed. Theseresults may be important in several biological applications such asbiosensors development inwhich several layers are mounted at thesurface of the thin film. Additionally, it was also detected that theantimicrobial potential of these films directly affected the celladhesion, and the observed interaction between the films and cells,which certainly involves adhesion proteins, was correlated with thecell integrity. This may be important in the development ofbiomedical implants and smart prosthetics by coating/altering thesurface in order to modulate and control the interaction withdifferent types of cells, microbial vs. mammalian cells.

Acknowledgements

This research is sponsored by FEDER funds through the programCOMPETE e Programa Operacional Factores de Competitividade e

and by national funds through FCT e Fundaç~ao para a Ciencia e aTecnologia e, under the projects PEST-C/FIS/UI607/2013, PEst-C/EME/UI0285/2013 and PEst-OE/BIA/UI4050/2014. J. Borges also ac-knowledges the support by the European social fund within theframeworkof realizing theproject “Supportof inter-sectoralmobilityand quality enhancement of research teams at Czech Technical Uni-versity in Prague”, CZ.1.07/2.3.00/30.0034. The authors alsoacknowledge Catarina Carneiro for helpingwith the FlowCytometer.

References

[1] Barber DJ, Freestone IC. An investigation of the origin of the colour of thelycurgus cup by analytical transmission electron microscopy. Archaeometry1990;32:33e45.

[2] Stockman MI. Nanoplasmonics: the physics behind the applications. PhysToday 2011;64:39e44.

[3] Pedrueza E, Sancho-Parramon J, Bosch S, Vald�es JL, Martinez-Pastor JP. Plas-monic layers based on Au-nanoparticle-doped TiO2 for optoelectronics:structural and optical properties. Nanotechnology 2013;24:065202.

[4] Torrell M, Machado P, Cunha L, Figueiredo NM, Oliveira JC, Louro C, et al.Development of new decorative coatings based on gold nanoparticlesdispersed in an amorphous TiO2 dielectric matrix. Surf Coat Technol2010;204:1569e75.

[5] Torrell M, Cunha L, Cavaleiro A, Alves E, Barradas NP, Vaz F. Functional andoptical properties of Au:TiO2 nanocomposite films: the influence of thermalannealing. Appl Surf Sci 2010;256:6536e42.

[6] Torrell M, Cunha L, Kabir MR, Cavaleiro A, Vasilevskiy MI, Vaz F. Nanoscalecolor control of TiO2 films with embedded Au nanoparticles. Mater Lett2010;64:2624e6.

[7] Atwater HA, Polman A. Plasmonics for improved photovoltaic devices. NatMater 2010;9:205e13.

[8] Sancho-Parramon J, Janicki V, Zorc H. On the dielectric function tuning ofrandom metal-dielectric nanocomposites for metamaterial applications. OptExpress 2010;18:26915e28.

J. Borges et al. / Vacuum 122 (2015) 360e368368

[9] Alivisatos P. The use of nanocrystals in biological detection. Nat Biotechnol2004;22:47e52.

[10] Siozios A, Koutsogeorgis DC, Lidorikis E, Dimitrakopulos GP, Kehagias T,Zoubos H, et al. Optical encoding by plasmon-based patterning: hard andinorganic materials become photosensitive. Nano Lett 2011;12:259e63.

[11] Daniel MC, Astruc D. Gold nanoparticles: assembly, supramolecular chemistry,quantum-size-related properties, and applications toward biology, catalysis,and nanotechnology. Chem Rev 2004;104:293e346.

[12] Ooms MD, Bajin L, Sinton D. Culturing photosynthetic bacteria through sur-face plasmon resonance. Appl Phys Lett 2012;101:253701.

[13] Carvalho P, Sampaio P, Azevedo S, Vaz C, Espin�os JP, Teixeira V, et al. Influenceof thickness and coatings morphology in the antimicrobial performance ofzinc oxide coatings. Appl Surf Sci 2014;307:548e57.

[14] Kim BYS, Jiang W, Oreopoulos J, Yip CM, Rutka JT, Chan WCW. Biodegradablequantum dot nanocomposites enable live cell labeling and imaging of cyto-plasmic targets. Nano Lett 2008;8:3887e92.

[15] Homola J, Yee SS, Gauglitz G. Surface plasmon resonance sensors: review.Sensors Actuators B Chem 1999;54:3e15.

[16] Hoa XD, Kirk AG, Tabrizian M. Towards integrated and sensitive surfaceplasmon resonance biosensors: a review of recent progress. Biosens Bio-electron 2007;23:151e60.

[17] Homola J. Surface plasmon resonance sensors for detection of chemical andbiological species. Chem Rev 2008;108:462e93.

[18] Guo XW. Surface plasmon resonance based biosensor technique: a review.J Biophot 2012;5:483e501.

[19] Homola J, Dostalek J, Chen SF, Rasooly A, Jiang SY, Yee SS. Spectral surfaceplasmon resonance biosensor for detection of staphylococcal enterotoxin B inmilk. Int J Food Microbiol 2002;75:61e9.

[20] Torrell M, Kabir R, Cunha L, Vasilevskiy MI, Vaz F, Cavaleiro A, et al. Tuning ofthe surface plasmon resonance in TiO2/Au thin films grown by magnetronsputtering: the effect of thermal annealing. J Appl Phys 2011;109:074310.

[21] Toudert J, Simonot L, Camelio S, Babonneau D. Advanced optical effectivemedium modeling for a single layer of polydisperse ellipsoidal nanoparticlesembedded in a homogeneous dielectric medium: surface plasmon resonances.Phys Rev B 2012;86:045415.

[22] Pereira RMS, Borges J, Pereira PAS, Smirnov GV, Vaz F, Cavaleiro A, et al.Optical response of fractal aggregates of polarizable particles. In:Costa MFPCM, Nogueira RN, editors. Second International Conference on ap-plications of optics and photonics. Aveiro: SPIE; 2014. 92865Me8M.

[23] Scopelliti PE, Borgonovo A, Indrieri M, Giorgetti L, Bongiorno G, Carbone R,et al. The effect of surface Nanometre-Scale morphology on protein adsorp-tion. PLoS One 2010;5.

[24] Lopes C, Fonseca P, Matam�a T, Gomes A, Louro C, Paiva S, et al. Protective Ag:TiO2 thin films for pressure sensors in orthopedic prosthesis: the importanceof composition, structural and morphological features on the biologicalresponse of the coatings. J Mater Sci Mater Med 2014;25:2069e81.

[25] Barradas NP, Jeynes C. Advanced physics and algorithms in the IBA Data-Furnace. Nucl Instrum Methods Phys Res Sect B Beam Interact Mater Atoms2008;266:1875e9.

[26] Ruhm A, Topeverg BP, Dosch H. Supermatrix approach to polarized neutronreflectivity from arbitrary spin structures. Phys Rev B 1999;60:16073.

[27] Bradford MM. A rapid and sensitive method for the quantitation of microgramquantities of protein utilizing the principle of protein-dye binding. Anal Bio-chem 1976;72:248e54.

[28] Gallagher SR. One-dimensional SDS gel electrophoresis of proteins. In:Ausubel Frederick M, et al., editors. Current protocols in molecular biology;2006 [Chapter 10], Unit 10.12A.

[29] Macedo F, Vaz F, Torrell M, Faria RT, Cavaleiro A, Barradas NP, et al. TiO2coatings with Au nanoparticles analysed by photothermal IR radiometry.J Phys D Appl Phys 2012;45.

[30] Borges J, Buljan M, Sancho-Parramon J, Bogdanovic-Radovic I, Siketic Z,Scherer T, et al. Evolution of the surface plasmon resonance of Au:TiO2nanocomposite thin films with annealing temperature. J Nanopart Res2014;16:1e14.

[31] Singh AV, Vyas V, Patil R, Sharma V, Scopelliti PE, Bongiorno G, et al. Quan-titative characterization of the influence of the nanoscale morphology ofnanostructured surfaces on bacterial adhesion and biofilm formation. PLoSOne 2011;6.

[32] Podest�a A, Bongiorno G, Scopelliti PE, Bovio S, Milani P, Semprebon C, et al.Cluster-assembled nanostructured titanium oxide films with tailored wetta-bility. J Phys Chem C 2009;113:18264e9.

[33] Takeda S, Fukawa M, Hayashi Y, Matsumoto K. Surface OH group governingadsorption properties of metal oxide films. Thin Solid Films 1999;339:220e4.

[34] Giorgetti L, Bongiorno G, Podest�a A, Berlanda G, Scopelliti PE, Carbone R, et al.Adsorption and stability of streptavidin on cluster-assembled nanostructuredTiOx films. Langmuir 2008;24:11637e44.

[35] Brahma A, Mandal C, Bhattacharyya D. Characterization of a dimeric unfoldingintermediate of bovine serum albumin under mildly acidic condition. BiochimBiophys Acta (BBA) e Proteins Proteomics 2005;1751:159e69.

[36] BjerknesR.Flowcytometric assay forcombinedmeasurementofphagocytosis andintracellular killing of candida-albicans. J Immunol Methods 1984;72:229e41.

[37] Ramani R, Ramani A, Wong SJ. Rapid flow cytometric susceptibility testing ofCandida albicans. J Clin Microbiol 1997;35:2320e4.