Influence of thickness and coatings morphology in the antimicrobial performance of zinc oxide coatings

Influence of thickness and coatings morphology in the antimicrobial performance of zinc oxide coatings. P. Carvalho1, P. Sampaio2, J. O. Carneiro1, C.Vaz2, J. P. Espinós3, V. Teixeira1 1 Department of Physics, University of Minho, Campus de Azurém, 4800-058 Guimaraes, Portugal 2 CBMA, University of Minho, Campus de Gualtar, 4700 Braga, Portugal 3 Instituto de Ciencia de Materiales de Sevilla, CSIC-University of Sevilla, Avda. Américo Vespucio 49, 41092 Sevilla, Spain Abstract In this research work, the production of undoped and silver (Ag) doped zinc oxide (ZnO) thin films for food packaging applications was developed. The main goal was to determine the influence of coatings morphology and thickness on the antimicrobial performance of the produced samples. The ZnO based thin films were deposited on PET (Polyethylene terephthalate) by means of DC reactive magnetron sputtering. The thin films were characterized by optical spectroscopy, X-Ray Diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and Scanning Electron Microscopy (SEM). The antimicrobial performance of the undoped and Ag-doped ZnO thin films was also evaluated. The results attained have shown that all the deposited zinc oxide and Ag-doped ZnO coatings present columnar morphology with V-shaped columns. The increase of ZnO coatings thickness until 200nm increases the active surface area of the columns. The thinner samples (50 and 100 nm) present a less pronounced antibacterial activity than the thickest ones (200-600nm). Regarding Ag-doped ZnO thin films, it was verified that increasing the silver content decreases the growth rate of E. coli. and decreases the amount of bacteria cells present at the end of the experiment. Keywords: ZnO thin films, antimicrobial activity, food packaging 1. Introduction More natural, safer and higher quality consumable products have been seen as a market demand of huge importance in what concerns the Food Packaging Industry. Foodborne illnesses and death is rising worldwide, particularly in developed countries[ 1 ]. Indeed, data from the Foodborne Diseases Active Surveillance Network (Food Net) states that comparing 2007 with 2004-2006, the estimated occurrence of infections caused by Campylobacter, Listeria, Salmonella, Shigella, Vibrio, and Yersinia did not diminished significantly, and above all, the incidence of Cryptosporidium infections even increased by 44% [ 2 ]. At the same time, foodborne illness-outbreaks create tremendous social and economic burdens bringing the fear for the re-emergence of infections diseases. Furthermore, the development of antibiotic resistance continues to draw public attention to food safety [ 3 , 4 ]. The combination of the supra-mentioned reasons with the current awareness for environmental conservation and protection has empowered the development of edible coatings and films from biodegradable materials to maintain the quality of both fresh and processed food [ 5 ]. Silver-based thin films [ 6 , 7 , 8 ] and zinc oxide nanoparticles [ 9 , 10 , 11 , 12 , 13 , 14 , 15 ] have emerged as promising candidates for active food packaging systems, particularly due to their antibacterial activity. However, due to the lack of knowledge regarding the interactions of nano-sized materials at the molecular and physiological levels and their potential effects on human body, a major concern related to safety of nanoparticles for consumer’s health is raising [ 16 , 17 , 18 , 19 ]. Titanium dioxide (TiO2) has been used as a selfcleaning and self-sterilizing material to coat different tools, including sanitary ware, food tableware and cooking ware [ 20 , 21 ]. The antimicrobial effects of TiO2 are activated by its photocatalytic behaviour, which is totally dependent on ultraviolet and/or visible light irradiation. However, for food packaging industry this is a huge drawback since the antimicrobial activity will only be effective in irradiated packages. An alternative to overcome this limitation could be the use of Zinc oxide (ZnO). ZnO presents antimicrobial activity, can act as a permeation barrier coating [ 22 ] and is Generally Recognized as Safe material (GRAS) by the U.S. Food and Drug Administration (21CFR182.8991). In this sense, the development of zinc oxide thin films can be considered of great added value to food packaging industry. There are a limited amount of scientific publications available describing the interaction between ZnO nanoparticles and bacterial cells and as far the authors knowledge there is no reported work concerning the antimicrobial activity of zinc oxide coatings deposited by means of reactive magnetron sputtering. In this sense, the main goal of this research work is to evaluate the antimicrobial activity of zinc oxide thin films and study the influence of coating thickness, morphology and silver content in the antimicrobial activity. Its applicability for food packaging systems will also be discussed. To the author’s knowledge, this work is the first to study the influence of thickness and coatings morphology on the antimicrobial activity, contributing to the knowledge of important physical variable in the thin films production. 2. Materials and Methods 2.1. Thin films production and characterization Zinc oxide thin films were deposited on Polyethylene Terephthalate (PET) substrates from Goodfelow and silicon (Si) substrates from Silllicon Materials by reactive DC magnetron sputtering (Advanced Energy Pinnacle Plus, 5K, DC Pulsed). The films deposited on PET were used for the evaluation of the optical properties and the antimicrobial activity and the ones deposited on Si were used for structural, morphological and compositional characterization. It was used a circular zinc target (Φ=75mm) with a purity of 99,9% and a thickness of 4mm acquired from Goodfellow Before deposition the sputtering chamber was pumped down to 10-3 Pa. A gas atmosphere composed by argon (Ar, working gas) and oxygen (O2, reactive gas) was used in the deposition processes. The Ar and O2 flows were kept constant at 70 sccm (standard cubic centimetre per minute) and 18 sccm, respectively. Each deposition was carried out with a working pressure of 0.6 Pa, a current of 0.35 A and a target/substrate distance of 80 mm. Prior to each deposition the zinc target was pre-sputtered during 5 minutes. All other deposition parameters were kept constant and the depositions were performed at room temperature (40 ºC measured inside the chamber). X-ray diffraction (Philips PW 1710 X-ray diffractometer) analysis was used to investigate the crystallographic structure of ZnO thin films. The thickness and morphology of ZnO thin films were obtained by the observation of the cross section of the fractured thin films by SEM (NanoSEM-FEINOVA 200). The optical transmittance was measured by visible spectroscopy (Shimadzu UV-310PC scanning spectrophotometer). The chemical characterization was by using a XPS spectrometer (12 kV, 20 mA) from VG (ESCALAB 210). An unmonocromatized Mg K (1253.6 eV) source was used during the measurements. For all the zinc oxide coatings, the binding energies of the XPS spectra were calibrated by using of Zn 2p3/2 at 1022.2 eV and the spectra were acquired at room temperature. All the coatings were sputtering cleaned with Ar+ until a stationary composition was achieved. At this stage, no carbon resulting from superficial contamination was observed. 2.2. Thin films antimicrobial activity In this research work, the Escherichia coli was the bacteria species chosen and it was obtained from the Centre of Molecular and Environmental Biology (CBMA), Department of Biology, University of Minho. A pre-culture was prepared for each individual batch experiment. One colony of E. coli strain HB101 was picked and loop inoculated into a 125-ml Erlenmeyer flask, containing 20 ml of Luria Bertani (LB) broth (10 g/L tryptone, 10 g/L NaCl, and 5 g/L yeast extract). This pre-culture was incubated at 37°C, for 12 to 15h. On the day after, cells were transferred into different 250-ml Erlenmeyer flasks containing 50 ml of LB broth medium at a starting optical density (OD) of 0.1 measured at a wavelength of 640 nm. A coated PET circular sample (diameter of 6 cm), previously sterilized with 70% ethanol for 1 hour and rinsed in sterile water, was deposited on the bottom of the flask. Flasks were then shaken at 80 rpm in a temperature-controlled incubator at 37 °C, and the OD was monitored every hour. OD measurements were made using a Spectronic 20 instrument at 640 nm and the background (turbidity due to growth medium) was eliminated by taking blank readings. The specific growth rate (μ) was calculated from the exponential phase, according to the following equation:            1 2 12 ln 1 OD OD tt μ (1) where OD1 and OD2 are the optical densities corresponding to time instants t1 and t2, respectively. The generation time (tg) can be calculated according to the equation: μ tg 2ln  (2) 3. Results and discussion 3.1. Thin films structure and morphology The produced ZnO thin films in this research work are highly transparent, presenting a transmittance of about 80% in the visible region of the electromagnetic spectrum. The remainder 20%, is lost by reflection at the air/ZnO interface, by dispersion and by absorption of light in the substrate. In order to study the influence of the coating thickness and morphology in its antimicrobial activity, the thickness was controlled by changing the deposition time. With this methodology it was observed that the film thicknesses ranged from 50 to 600 nm. X-ray diffraction analyses were performed in representative samples and the measurements were done between 20º and 70º () for all thin films. Fig. 1 shows the undoped zinc oxide X-Ray Diffraction patterns of the thin films having 50, 200 and 600 nm thicknesses. Fig. 1 XRD patterns of ZnO thin films with 50, 200 and 600 nm thickness, deposited in PET substrate. In XRD patterns it is observed the presence of the (002) diffraction peak of the ZnO wurtzite structure (JPCDS36-1451[ 23 ]) in all the samples, indicating a preferential orientation to the c-axis perpendicular to the substrate, as previously reported by others authors[ 24 , 25 , 26 ]. Some authors correlate the preferential orientation with the minimization of internal stress and surface energy [ 27 , 28 , 29 ] and others [ 30 ] reported that c-orientation maybe result from the highest atomic density found along (002) plane. In Figure 1 it is also possible to observe that, as expected, the increase of the coatings thickness induces an increase of diffraction peak intensity. The decrease of full width at half maximum also suggests an increase of grain size with the thickness. Fig. 2 shows the SEM micrographs of the undoped ZnO samples cross-sections; the grains extend from the substrate to the top of the film. The typical morphology can be identified as a Zone 1 film of the modified structure zone model [ 31 ] with typically low compactness. Fig. 2. SEM micrographs presenting the thin films cross-sections with different thickness deposited in Si substrate: a) 50nm; b)200nm; c) 300nm and d) 600nm. The thinner coatings, 50 and 100 nm (not showed here) present a more compact morphology in comparison with the thickest ones. It is possible to observe that the increase of coatings thickness induces the formation of less compact coatings, with well define and separated columns (Fig. 2). From this figure it is can also be identified an increase of columns width with the increase of coatings thickness. From the presented a) b) c) d) results it is also possible to conclude that a facet plane can be observed in the upper region for the thickest coatings (300-600 nm). The attained results are in agreement with the ones reported by J. W. Shin et al. [ 32 ] that deposited ZnO coatings at room temperature on Si substrates. These authors explained the formation of facet planes as being a consequence of the lower diffusion rate of the atoms on the ZnO surface during column’s growth the deposition rate. To better understand the morphology of these coatings, top view SEM micrographs of ZnO samples with different thickness were taken and are shown in Fig. 3. Fig. 3. Top-view SEM micrographs of ZnO thin films with different thickness deposited in Si substrate: a) 50nm; b)200nm; c) 300nm and d) 600nm. a) b) c) d) Fig. 7. XPS Zn2p (left) and Ag3d (right) spectra of silver doped ZnO films and undoped ZnO. All coatings were deposited on Si substrate. The Zn2p3/2 core line, (calibrated at 1022.2 eV[ 60 ] of binding energy) exhibits a small asymmetry in the right side, indicating that zinc could present more than one oxidation state. The XPS spectra of Ag3d3/2 peak (Fig. 7) shows binding energies in the range 368.6-369.2eV for the three doped samples. The evolution of the binding energy of Ag3d3/2 with the chemical state of silver is anomalous, since it decreases with the oxidation state of this element. The reported silver compound with the highest binding energy is AlAg2 (368.7 eV), and the one with the lowest is AgF2 (367.7 eV). Binding energies for pure Ag, Ag2O, and AgO bulk samples are 368.22, 367.8, and 367.4 eV, respectively[ 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 ]. It is evident that the peak Ag3d5/2 in Ag doped ZnO samples are located at energies far above bulk metallic silver (+1 eV) and even above that for bulk AlAg2 (+0.5 eV). Unfortunately, since the surface concentration of silver in doped films is very low, it is not possible to detect properly in these samples the AgMVV Auger signals and, consequently, the value of the modified Auger parameter of silver, which is so useful to determine the chemical state of this element, cannot be obtained. We can imagine two different hypothesis that could explain the high binding energy values obtained for Ag3d3/2 peaks in ZnO(Ag) films. One, that silver is forming small metal clusters strongly interacting with ZnO support, since it has been widely discussed in the literature that the binding energy of small metallic particles is usually shifted to lower binding energy [ 69 ]. For instance, a shift of 0.4 eV was observed for silver deposited on graphite[ 70 ], shifts of +0.6 eV [ 71 ] and +0.9 eV [ 72 ] have been reported for silver on Al2O3, while shifts of +0.6 and +1.2 eV have been reported for silver deposited on TiO2 at 300K and 100K, respectively [ 73 ]. These shifts have been attributed to both initial and final state factors affecting the photoemission process. Initial state effects are those affecting the energy of the initial state, as for instance, the metal-substrate interaction, the mean coordination of the atoms in the clusters and their geometry [ 74 , 75 ]. Final state effects are those affecting the relaxation or screening of the photohole, as the Coulombic potential or the dielectric constant of the substrates [ 76 , 77 ]. The binding energy shift observed in our samples, up to +1.0 eV, would be in the upper limit of the range found for small silver clusters on Al2O3 and TiO2. To verify the validity of this hypothesis, we have carried out an experiment in our laboratory where increasing amounts of pure silver are condensed under ultra high vacuum on ZnO films, from a fraction of a monolayer to many monolayers. Note that, regardless of the growth mechanism followed by silver depositing on ZnO (layer by layer or three dimensional island growth), the mean particle size will increases when the deposited amount increases. The result of this experiment is illustrated in Fig. 8, where the evolution of Ag3d spectrum is displayed as a function of the nominal thickness of deposited silver, and in Fig. 9, where the evolution of the Ag3d5/2 peak binding energy is plot versus the same variable. As can be seen, when deposited on ZnO, the binding energy of very small silver aggregates only shift up to +0.6 eV in respect to that one of bulk silver, which is quite lower than the experimental binding energy value found for Ag doped ZnO films. 380 376 372 368 364 360 Ag3d Intensity (a.u.) Binding Energy (eV) 136.0 60.4 30.5 15.2 6.7 4.3 3.2 2.2 1.6 0.7 0.1 Ag/ZnO nominal thickness (A) Fig. 8. Evolution of Ag3d spectrum for growing amounts of metal silver deposited under UHV on ZnO. The nominal thickness of deposited silver, assuming a layer to layer growth mode, is expressed in Angstroms. 020 40 60 80 100 120 140 368.2 368.4 368.6 368.8 369.0 369.2 Ag3d5/2 Binding Energy Ag nominal thickness () Ag/ZnO Fig. 9. Evolution of the Ag3d5/2 binding energy of silver for deposits of silver on ZnO with growing nominal thickness. A second possible hypothesis to explain the high binding energy found for silver in Ag/ZnO, is that metal Ag particles are doped with metal Zn, giving rise to the formation of an intermetallic compound, where the binding energy of Ag3d3/2 signal could be higher than that for pure silver, as reported for AlAg2. Since the percentage of silver doping the ZnO films is very low, the detection of the Zn species linked to the silver atoms is fully hidden by those Zn atoms located in the ZnO phase. Moreover, metal Zn species cannot be distinguish from Zn+2 cations by XPS on the basis of the main primary photoelectron signals (Zn2p), because both chemical species generates just the very same spectrum. Fortunately, the elucidation of the chemical state of Zn can be made effectively by XPS by determining the value of the modified Auger parameter[ 78 ]. For that purpose, both ZnL3M45M45 and Zn2p3/2 photoemission signals must be measured, since the value of the Zn modified Auger parameter, ´, is equal to the sum of the binding energy of the Zn2p3/2 photoelectron peak and the kinetic energy of the ZnL3M45M45 Auger line. In order to further study this second hypothesis, one film of Zn containing a small amount of Ag, was deposited by magnetron sputtering from metallic Zn and Ag targets, and analyzed by XPS. The objective was to obtain a reference sample of Ag doped with metallic Zn, and compare its Ag3d3/2 binding energy with the one for the case of Ag doped ZnO films. In spite of being prepared from pure Zn and Ag targets, the film of Ag doped Zn, when examined by XPS, only showed divalent zinc at the surface, characterized by a modified Auger parameter of ~2010.0 eV, very likely as a consequence of the corrosion of this metal during its exposition to the atmosphere of the laboratory. However, and as expected, the sputtering cleaning of its surface, by bombardment with Ar+ ions, gradually remove the altered layer, and a stationary state is reached in its composition after a treatment of 120 min, treatment long enough to remove around ~12 nm. The evolution of the surface composition of this film by the ion etching treatment is neatly noticeable by the decrease in the O/Zn atomic ratio and by the changes in the shape of ZnL3M45M45 Auger line, where a new peak at 991.8 eV of kinetic energy, due to Zn(0), rises with the sputtering treatment. In Fig. 10, the spectral region for ZnL3M45M45 signal is shown for this sample at the stationary state, where the peaks for Zn(0) and Zn(+2) are labeled and the values of the respective modified Auger parameters calculated. 980 985 990 995 1000 after 120minAr+ ´=2013.83eV Zn0 991.65 Intensity (a.u.) Kinetic Energy (eV) ZnL3M45M45 987.25 Zn+2 ´=2009.43eV Fig. 10- Photoemission spectrum of the ZnL3M45M45 Auger line for the Zn(Ag) sample after surface cleaning by sputtering with Ar+ ions of 2.5 keV for 120 min. As the morphology of this coating is highly columnar, even at the stationary state the collected XPS signal presents the contribution of the corroded lateral surface of the columns. Fig. 11Fig. 11 shows the Zn2p and Ag3d photoemission signals of this silver doped zinc film at the original state (surface corroded) and at the stationary state after sputtering. As can be seen, Zn2p3/2 peak appears in both situations at around 1022.20 eV of binding energy, while Ag3d5/2 peak shifts with the sputtering treatment from 368.64 to 369.47 eV. The intensities of both Zn and Ag signals increase with this treatment, as a result of the removal of surface contaminants and, very likely, the preferential sputtering of oxygen. Obviously, the binding energy of the Ag3d5/2 peak, 369.47 eV, is so high that it cannot be ascribed to pure metallic silver, although consisting of very small particles. Since it increases simultaneously with the detection of metallic zinc, the formation of a Zn-Ag bond is the most probable explanation for this high value of binding energy. In other words, while the presence of very small aggregates of silver (first hypothesis) could only account for a binding energy shift of up to +0.6 eV, the doping of these particles with Zn (second hypothesis) could explain by itself the whole experimental shift and even further. 1028 1024 1020 1016 original after 120minAr+ Intensity (a.u.) Binding Energy (eV) Zn2p1022.20 a) 380 376 372 368 364 360 Intensity (a.u.) Binding Energy (eV) original after 120minAr+ Ag3d5/2 369.5 b) Fig. 11. XPS Zn2p (a)) and Ag3d (b)) spectra of silver doped zinc film at the original state and after 120 min of surface etching with Ar+ ions of 2.5 keV . Fig. 12 shows the top and cross section micrographs of the silver doped ZnO thin films. Fig. 12. Top-view and cross section SEM micrographs of silver doped ZnO thin films; a) and b) ZnO(Ag)-1 sample, c) and d) ZnO(Ag)-3 sample. From the previous figure it is possible to observe that the presence of silver induces a change in the ZnO coatings morphology in comparison with Fig. 2b. For the samples with very low silver content, ZnO(Ag)-1 and 2 (not showed here), the shape of the columns change from circular pillar-like morphology (see Fig. 2) to triangular morphology (see Fig. 12), maintaining a V shape form in the cross section view. It is important to note that these samples do not show a crystalline preferential growth, in contrast with the undoped zinc oxide samples. The sample with the highest silver content presents a columnar morphology with a circular shape, similar to the undoped samples, but with a much less dense morphology, where the space between columns is cleared observed. Once more time, the circular columnar morphology is related to a 200 000x –ZnO(Ag)-1 200 000x –ZnO(Ag)-1 200 000x –ZnO(Ag)-3 200 000x –ZnO(Ag)-3 a) b) c) d) preferential orientation of the c-axis perpendicular to the substrate surface (see Fig. 6). This result suggests that the different crystalline growth related with the presence of silver could induce morphology changes in the coatings. L.N. Wang, et al. [ 79 ] reported that Ag doping could cause morphological changes in the ZnO thin films. The authors observed that the presence of silver induces the formation of a porous morphology. The antibacterial activities of undoped and Ag-doped ZnO thin films against E. coli are depicted in Fig.7. a) b) Fig. 13. Antibacterial activities of undoped and Agdoped ZnO thin films against E. coli. a) Optical density (OD) at 640 nm over time of incubate bacteria cells with different Ag content The control growth was assessed using non coated PET with the same area; b) generation time of E. coli. bacteria. From the previous plot (Fig. 13) it is possible to observe an enhancement of antibacterial activity with the increase of silver content in the zinc oxide based coatings. With the increase of silver content there is a decrease of E. coli. growth rate (see, Fig. 13.a) and 7.b)), a decreased amount of bacteria cells present at the end of the experiment. On the other hand for the sample having the highest Ag/Zn atomic ratio of 0.04 it is observed a complete growth inhibition (see, Fig. 13.a). The antibacterial mechanism of ZnO is still under investigation. Two major mechanisms have been suggested: (1) the release of Zn2+ ions from the coating [ 80 ] and (2) the generation of hydrogen peroxide as well as radical oxygen species from ZnO surface[ 81 , 82 , 83 , 84 , 85 , 86 ]. Sugarman [ 87 ] suggested that zinc ions binds to the membranes of microorganisms, extending the lag phase of the growth cycle and increasing the generation time of the microorganisms [ 88 ]. Others authors [ 89 , 90 , 91 ], observed that Zn2+ ions do not have much antibacterial activity up to a concentration of 7.3x10-5 M, the same order of magnitude of the solubility of ZnO in the solution. This may be consistent with the prediction that E. coli can metabolize Zn2+ as an oligoelement[ 92 ]. This suggests that these ions are not primarily responsible for the antibacterial activity of ZnO. The same authors claimed that the antimicrobial dominant mechanism should result from Radical Oxygen Species (ROS) generation and their subsequent interaction with cell. Regarding the generation of highly reactive species, such as OHradicals, H2O2 and O22−, Nagarajan Padmavathy and co authors[ 93 ] explained as follows: ZnO with defects can be activated by both UV and visible light, which will induce electronhole pairs (e-/h+) generation. Water molecules can spontaneously dissociate into OH- and H3O+ species. Dissolved oxygen molecules are transformed to superoxide radical anions  , that in turn will react with H3O+ to generate (HO2) radicals, which upon subsequent collision with electrons produce hydrogen peroxide anions (HO2-). Then, they then react with H3O+ to produce H2O2 molecules. However, light is needed to produce photocatalytic ROS. Since in the present work the E. coli. A- Bacteria cells were incubated in dark conditions, the generation of (ROS) cannot be explained by a photocatalytic mechanism. By another hand, Ken Hirota et al.[ 94 ] and J. Sawai et al.[ 95 ] observed the generation of OH radicals from ZnO ceramics in dark conditions. Both authors did not explain the formation mechanism of these radicals. 72 K.Lou, T.P.St.Clair, X.Lai and D.W.Goodman. J.Phys.Chem. B, 104 (2000) 3050- 3057. 73 K.Lou, T.P.St.Clair, X.Lai and D.W.Goodman. J.Phys.Chem. B, 104 (2000) 3050- 3057. 74 M.G.Mason, Physical Review B, 27(2) (1983) 748-762. 75 P.S.Bagus, C.R.Brundle, G.Pacchioni, F.Parmegiani, Surf. Sci. Rep. 19 (1993) 265. 76 G.K. Wertheim, S.B.DiCenzo, D.N.E.Buchanan, Physical Review B 33 (1986,)5384 77 W.F.Jr. Egelhoff, G.G.Tibbetts, Physical Review B, 19 (1979) 5028. 78 C.D.Wagner, Faraday Discuss. Chemical Soc., 60 (1975) 291. 79 L. N. Wang, L.Z.Hu, H.Q.Zhang, Y.Qiu,Y.Lang, G.Q.Liu,J.Y.Ji, J.X.Ma,Z.W.Zhao, Materials Science in Semiconductor Processing 14 (3-4) (2011) 274. 80 M. Heinlaan, A. Ivask, I. Blinova, H.C. Dubourguier, A. Kahru, Chemosphere 71 (2008) 1308–16. 81 S.J. Chen, Y.C. Liu, C.L. Shao, R. Mu, Y.M. Lu, J.Y. Zhang, Adv Mater 17 (2005) 586-90. 82 J. Sawai, S. Shoji, H. Igarashi, A. Hashimoto, T. Kokugan, M. Shimizu, J Ferment Bioeng 86 (1998) 521–2. 83 J. Sawai, S. Shoji, H. Igarashi, A. Hashimoto, T. Kokugan, M. Shimizu, H. Kojima, Ferment. Bioeng. 86 (1998) 521–522. 84 R. Brayner, R. Ferrari-Iliou, N. Brivois, S. Djediat, M.F. Benedetti, F. Fievet, Nano Lett 6 (2006) 866–70. 85 T. Xu, CS. Xie, Prog Org Coat 46 (2003) 297–301. 86 J. Sawai, S. Shoji, H. Igarashi, A. Hashimoto, T. Kokugan, M. Shimizu, H. Kojima, J. Ferment. Bioeng. 86 (1998) 521–522. 87 B. Sugarman, Rev. Infect. (1983) 138-147. 88 L.L. Radke, B.L. Hahn, D.K. Wagner, P.G. Sohnle, Clinical Immun. And Immuno. Pathol. 73(3) (1994) 344-349. 89 J. Sawai, Journal of Microbiological Methods 54 (2003) 177– 182. 90 Lingling Zhang, Yunhong Jiang, Yulong Ding, Nikolaos Daskalakis, Lars Jeuken, Malcolm Povey, Alex J. O’Neill, David W. York, J Nanopart Res 12 (2010) 1625– 1636. 91 Guy Applerot, Anat Lipovsky, Rachel Dror, Nina Perkas, Yeshayahu Nitzan, Rachel Lubart, Aharon Gedanken, Adv. Funct. Mater. 19 (2009) 842–852. 92 M. Roselli, A. Finamore, I. Garaguso, M.S. Britti, E. Mengheri J. Nutr. 133 (2003) 4077-4082. 93 Nagarajan Padmavathy, Rajagopalan Vijayaraghavan, Sci. Technol. Adv. Mater. 9 (2008) 035004. 94 Ken Hirota, Maiko Sugimoto, Masaki Kato, Kazuhiko Tsukagoshi, Tooru Tanigawa, Hiroshi Sugimoto, Ceramics International 36 (2010) 497–506. 95 J. Sawai, E.Kawada, F.Kanou, H. Igarashi., A. Hashimoto, , T. Kbkugan, , M. Shimizu, J. Chem. Eng. Japan, 29 (1996) 627-633. 96 J. M. Schierholz, L. J. Lucas, A. Rump, G. Pulverer, journal of Hospital Infection 40 (I 998) 257-262. 97 J. M. Schierholz, L. J. Lucasj, A. Rump, G. Pulverer, journal of Hospital Infection 40 (I 998): 257. 98 K. Modak, C. Fox, Biochem Pharm 22 (1973) 2392-2404. 99 J. M. Schierholz, L. J. Lucas, A. Rump, G. Pulverer, journal of Hospital Infection 40 (I 998) 257-262. 100 J.A. Spadaro, T.J. Berger, S.D. Barranco, S.E. Chapin, R.O. Becker, Microb. Agents Chemother. 6 (1974) 637. 101 Q.L. Feng, J. Wu, G.Q. Chen, F.Z. Cui, T.M. Kim, J.O. Kim, J. Biomed. Mater. Res. 52 (2000) 662.