The effect of surface composition of titanium films on bacterial adhesion

Y L Jeyachandran¹, Sa K Narayandass¹, D Mangalaraj¹, C Y Bao² and P J Martin³

¹ Department of Physics, Bharathiar University, Coimbatore 641 046, Tamil Nadu, India
² West China College of Stomatology, Sichuan University, Chengdu 610041, People's Republic of China
³ CSIRO Industrial Physics, PO Box 218 Lindfield, NSW 2070, Australia

Email: sakndass@yahoo.com and cybao9933@sina.com

Received 7 September 2005
Published 1 March 2006

Keywords: titanium thin films, surface composition, bacterial adhesion

Introduction

Investigations on bacterial adhesion to surfaces are of great environmental, medical and industrial importance and consequently draw considerable attention in the literature [1]. Bacterial adhesion to material, in particular biomaterial, surfaces is generally determined by two important properties, roughness and chemical stoichiometry. The effect of roughness is the mediation of bacterial establishment due to mechanical retention. The effect of composition is the generation of surface charge states that influence the hydrophobicity and electrostatic interactions resulting in bacterial adhesion and that are restorative through unspecific binding. In order to inhibit the bacterial adhesion and therefore to prevent the bacterial infection, the bioimplant material surfaces were subjected to modifications whereby the charge state and topography are controlled. In the case of titanium (Ti), an established biomedical material, the surface charge state and roughness are interrelated [2] thereby making it difficult to control the surface topography independently of the chemical nature. On the other hand, the surface chemical composition of Ti can be varied independently of the topographical characteristics by controlled oxidation or nitridation. In the present work, the surface chemical composition of the Ti films with similar smooth topography was modified and the bacterial adhesion response was studied.

Few works have reported on the bacterial adhesion characteristics of Ti-based surfaces [3, 4] and the prevailing adhesion mechanism is yet to be established. However, surface composition-related mechanisms such as free electron contribution, electron transfer, resistivity/conductivity and band gap density of states, are said to mediate the bacterial adhesion on conducting surfaces [5-7]. The bacterial adhesion studies on Ti-based materials gain significance for two important reasons. One is that the Ti-based materials are extensively used as bioimplants. The other is that the surface charge states of Ti can be tailored over a wide range by modifying the composition from metallic to nitride to oxide phases whereby the fundamental studies on the charge state interactions between the bacteria and the material surface may be understood to a greater extent. In this communication, we study the surface composition dependence of bacterial adhesion on Ti films and show the large variations in the order of bacterial adhesion on modifying the surface oxide-nitride composition of the films.

Methods and results

Ti thin films of thickness 110 nm (±10 nm variations between different films) were deposited onto cleaned silicon substrates kept at room temperature from a titanium metal target using a conventional dc magnetron sputtering machine. Commercial argon was used as the sputtering gas. The substrate to target distance was 100 mm. Deposition parameters such as the cathode power, sputtering pressure and base pressure were varied to prepare different types of Ti films in terms of surface composition. Accordingly six types of Ti films were prepared. The detailed preparation conditions employed are given in table 1 with the sample codes.

The surface chemical nature of the films was studied using x-ray photoelectron spectroscopy (XPS, SPECS Sage 150). Figure 1 shows the high-resolution Ti2p, O1s and N1s spectra of the Ti films. UNIFITTU software (version 2.1) was used for peak fitting and quantitative chemical analysis. The deconvoluted spectra show the presence of three different chemical species on the surface of the films. The dominant doublet component in the Ti2p peaks corresponding to Ti2p_3/2 and Ti2p_1/2 at binding energies ~459.1 and ~464.7 eV, respectively, was assigned to the titanium dioxide (TiO₂) phase [8]. Additionally, the two components observed at energies ~456.0 and 454.5 eV corresponding to Ti2p_3/2 were attributed to the substoichiometric and titanium nitride (TiN) phases, respectively [8]. The O1s peaks show two components: the titanium oxide component at ~530.5 eV and the hydroxide or hydroxyl (OH ^- ) component at ~532.0 eV [8]. The N1s peak of all the samples featured the component corresponding to the TiN phase at ~397.1 eV [9], whereby the resolution is less significant in the films T1, T2 and T3. In addition, the films T4, T5 and T6 showed a second component at ~399.7 eV which may be attributed to the contributions from the oxynitride phases or adsorbed nitrogen [9]. In conjunction with the observations from N1s peaks, the substoichiometric component observed for the films T1, T2 and T3 in the Ti2p peaks may be assigned to the oxide phase (Ti₂O₃) [8] and that for the films T4, T5 and T6 to the oxynitride phase (TiON_x) [9]. The quantitative details of the chemical components present in the films are given in table 2. Also given in table 2 are the bacterial counts on the films as enumerated from the scanning electron microscopy (SEM, KYKY 2800) images.

The following brief note could be inferred from tables 1 and 2. Oxidation was the predominant reaction in the surface of all Ti films. Under the conditions of low cathode power (75 W) and higher base pressure (>8 × 10 ^- 4 Pa), a significant presence of the nitride component was observed in the films whereas under the other conditions only native nitridation occurred. Oxidation reaction in the Ti films is a natural process because Ti has a high affinity to oxygen. The nitridation in the films may be due to the reaction with the background residual gas components during film preparation [10]. The percentage of OH ^-  component increased in the films with higher nitridation, which may be attributed to the increase in dissociative adsorption of the water molecules, the consequence of the 3 +  charge state of Ti corresponding to the nitride phase [2].

The surface topography of Ti films was studied by atomic force microscopy (AFM, PSIA XE-100 operated in non-contact mode). No significant difference in surface topography between different Ti films was observed. The AFM images revealed uniform and smooth surface of the films. The root mean square surface roughness of the films was around 2.3 nm with a variation of ±0.3 nm between the films. A typical topographical image of the Ti film T2 is shown in figure 2.

Figure 3 is the scanning electron microscope (SEM) images of the Ti films after bacterial culture experiments. The oral bacteria strain Porphyromonas gingivalis (ATCC 33277) was used in the study. The bacterial colony was cultured in the brain heart infusion (BHI) medium (bio Merieux sa) at 37 °C for 48 h under anaerobic conditions^Note1. Then the culture solution was diluted and separated to 1 ml samples in test tubes such that the bacteria concentration is 10⁶ CFU ml ^- 1. In the diluted bacterial solution, the Ti films sterilized by gamma radiation were immersed and cultured under similar conditions as above for 72 h. The cultured Ti film samples were gently washed in a physiological body solution, fixed in gluteraldehyde solution, dehydrated in ethanol solution and cold dried. The results were observed under SEM (KYKY 2800). The experiment was repeated three times to test the consistency of the results. The SEM images shown in figure 3 are one set of the consistent results obtained in the three trials. Enumeration of the attached cells was carried out by counting cells visualized on the scaled images obtained by SEM and the bacterial counts on different Ti films are presented in table 2. This procedure of evaluation was also followed in [11] as an alternative to optical microscopy with conventional staining methods, because the latter was found unsatisfactory due to surface undulations and staining of the background.

The Ti films T1, T2, T3 and T5 with native surface oxidation and nitridation recorded low numbers of adhered bacteria and on the film T1 the bacterial count was almost nil. On the films T4 and T6 with substoichiometric and nitride components in the surface the bacterial adhesion increased whereby T4 recorded relatively higher bacterial counts compared to that on T6.

Discussion and conclusions

For discussion, we focus on the effect of surface composition of the films on bacterial adhesion characteristics assuming the influence of the surface roughness of the films to be less significant. This assumption may be reasonable from the AFM results whereby different Ti films were found to have similar surface topography but showed varying bacterial adhesion characteristics.

On the basis of surface composition, the bacterial adhesion mechanism may be accounted for in terms of charge interactions [6, 12]. The charge characteristics of the film surfaces may be determined by two principle factors, the charge state of Ti and the adsorbed OH ^-  component. The observed variation in the percentage of OH ^-  component that may alter the surface wettability of the films may be considered to have an effect on bacterial adhesion. However, the measure of its influence on bacterial adhesion is uncertain in the present case. This is because the bacterial culture was performed in the solution medium where the surface modification of the films, in terms of OH ^-  content, may happen immediately after immersion of the samples. Alternatively, it may be reasonable to discuss bacterial adhesion in terms of the charge state of Ti on the surface of the films because any modification of the surface in the culture medium will primarily be the manifestation of the initial state [3, 7].

On the strength of the effect of the surface charge state, the chemical stabilization mechanism may have influenced the observed variation in the bacterial adhesion on the films. Generally the chemical stability of Ti is attained through surface oxidation or nitridation thereby making the surface free from unsaturated bonds (i.e. no surface charge). The chemical stabilization reactions may have reduced the bacterial adhesion on the films T1, T2, T3 and T5. It is well known that the TiO₂ phase is highly stable because of the high passivation repassivation rate of the native oxide on the surface. However, the defect states in TiO₂ may be introduced by the Ti substoichiometric charge state impurities as observed in the present case. In the Ti films T1, T2 and T3, the defect charges would have been compensated by oxidation or nitridation that is by the formation of stable Ti₂O₃ or TiN phases thereby making the surface immune to further reactions. In the case of the film T5 the stabilization may have been attained by charge transfer to the adsorbed water molecules [3]. In fact, a clear picture of the characteristics of T5 is yet to be established. It will be discussed in more detail in our future communication on the bacterial adhesion property of TiN films. The increase in bacterial adhesion on the films T4 and T6 would have been mediated by the excess charge carriers induced by the increased nitride and the substoichiometric oxide-nitride phases.

In conclusion, the surface chemistry of Ti films was the key factor that determined bacterial adhesion. Ti films with different surface chemical stoichiometries (oxide and/or nitride) showed varying degrees of bacterial adhesion. The surface stabilizing stoichiometries of typical Ti oxide-nitride blends exhibited minimum bacterial adhesion.

[1] 

Hall-Stoodley L, Costerton J W and Stoodley P 2004 Nat. Rev. Microbiol. 2 95 

[2] 

Guillemot F, Porté M C, Labrugère C and Baquey Ch 2002 J. Colloid Interface Sci. 255 75 

[3] 

Koerner R J, Butterworth L A, Mayer I V, Dasbach R and Busscher H J 2002 Biomaterials 23 2835 

[4] 

Harris L G, Tosatti S, Wieland M, Textor M and Richards R G 2004 Biomaterials 25 4135 

[5] 

Roscoe S G, Fuller K L and Robitaille G 1993 J. Colloid Interface Sci. 160 243 

[6] 

Poortinga A T, Bos R and Busscher H J 1999 J. Microbiol. Methods 38 183 

[7] 

Bolz A and Schaldach M 1993 Med. Biol. Eng. Comput. 31 123 

[8] 

Cai K, Müller M, Bossert J, Rechtenbach A and Jandt K D 2005 Appl. Surf. Sci. 250 252 

[9] 

Zerkout S, Achour S, Mosser A and Tabet N 2003 Thin Solid Films 441 135 

[10] 

Vargheese K D, Mohan Rao G, Balasubramanian T V and Kumar Sanjiv 2001 Mater. Sci. Eng. B 83 242 

[11] 

Tsibouklis J, Stone M, Thorpe A A, Graham P, Nevell T G and Ewen R J 2000 Int. J. Adhes. Adhes. 20 91 

[12] 

Boulange-Peteremann L, Baroux B and Bellon-Fontaine M N 1995 J. Microbiol. Methods 21 83 

Notes

Note1

 To make the condition anaerobic, the incubator was twice evacuated and charged with nitrogen and finally it was vacuumized and charged with the mixture gas (80% nitrogen, 10% carbon dioxide and 10% hydrogen) and maintained in this environment throughout the culture process.