Accelerated Escherichia coli inactivation in the dark on uniform copper flexible surfaces Sami Rtimi, Rosendo Sanjines, Michël Bensimon, César Pulgarin, and John Kiwi Citation: Biointerphases 9, 029012 (2014); doi: 10.1116/1.4870596 View online: http://dx.doi.org/10.1116/1.4870596 View Table of Contents: http://scitation.aip.org/content/avs/journal/bip/9/2?ver=pdfcov Published by the AVS: Science & Technology of Materials, Interfaces, and Processing Articles you may be interested in Cost-effective and highly sensitive cholesterol microsensors with fast response based on the enzyme-induced conductivity change of polyaniline Appl. Phys. Lett. 105, 113304 (2014); 10.1063/1.4896289 Antibacterial properties and cytocompatibility of tantalum oxide coatings with different silver content J. Vac. Sci. Technol. A 32, 02B117 (2014); 10.1116/1.4862543 Antibacterial efficacy of silver nanoparticles against Escherichia coli AIP Conf. Proc. 1512, 372 (2013); 10.1063/1.4791066 Sonochemical coating of textile fabrics with antibacterial nanoparticles AIP Conf. Proc. 1433, 400 (2012); 10.1063/1.3703213 Controlled surface adsorption of fd filamentous phage by tuning of the pH and the functionalization of the surface J. Appl. Phys. 109, 064701 (2011); 10.1063/1.3549113
Accelerated Escherichia coli inactivation in the dark on uniform copper flexible surfaces Sami Rtimi Ecole Polytechnique F ed erale de Lausanne, EPFL-SB-ISIC-GPAO, Station 6, CH-1015 Lausanne, Switzerland
Rosendo Sanjines Ecole Polytechnique F ed erale de Lausanne, EPFL-SB-IPMC-LNNME, Bat PH, Station 3, CH-1015 Lausanne, Switzerland
€l Bensimon Miche Ecole Polytechnique F ed erale de Lausanne, EPFL-ENAC-IIEGR-CEL, Bat GC, Station 18, CH-1015 Lausanne, Switzerland
sar Pulgarinb) Ce Ecole Polytechnique F ed erale de Lausanne, EPFL-SB-ISIC-GPAO, Station 6, CH-1015 Lausanne, Switzerland
John Kiwia) Ecole Polytechnique F ed erale de Lausanne, EPFL-SB-ISIC-LPI, Bat Chimie, Station 6, CH-1015, Lausanne, Switzerland
(Received 14 January 2014; accepted 26 March 2014; published 9 April 2014) The bacterial inactivation of Escherichia coli on Cu/CuO-polyester surfaces prepared by direct current magnetron sputtering was investigated in the dark and under actinic light (360 nm k 720 nm; 4.1 mW/cm2) as used commonly in hospital facilities. In the dark, complete bacterial inactivation (6log10 reduction) was observed within 150 min and under actinic light within 45 min. Sputtered samples led to nanoparticulate uniform Cu/CuO films 70 nm thick. The deposition rate used was 2.21015 atoms/cm2 s as determined by profilometry. X-ray fluorescence was used to determine the sample Cu-content and transmission electron microscopy determined Cu-particles 20 6 5 nm in size. The film optical absorption was observed to increase with Cu-content of the sample by diffuse reflection spectroscopy. The bacterial inactivation involved redox processes between Cu/CuO-polyester and the bacteria as observed by x-ray photoelectron spectroscopy. During sample recycling, the amount of Cu-release was determined by inductively coupled plasma-mass spectroscopy. The values required for E. coli inactivation were below the cytotoxicity level threshold allowed for mammalian cells. The E. coli inactivation by Cu/CuO-polyester seems to involve an oligodynamic effect since bacterial C 2014 American Vacuum Society. inactivation was achieved at very low Cu-concentrations. V [http://dx.doi.org/10.1116/1.4870596]
I. INTRODUCTION The manufacturing of high value added products such as bactericide textiles has increased rapidly in the last few years.1,2 Cu-particles are used for their catalytic and bactericide properties in the dark. The bactericide properties of Cutextiles have been reported recently by Borkow and Gabbay3–5 for textiles presenting antiviral, algicide, and fungicide properties and able to sustain an appropriate number of washing cycles loaded with Cu/CuO 1%. These textiles have been reported to be effective against Escherichia coli, Staphylococcus aureus, Pseudomonas aeruginosa, and Candida albicans, not causing skin irritation when externally applied. Cu has been known for a long time to have effective bactericide action.6,7 Cu-ions have been reported to be biocidal by binding to specific sites in the DNA.8 These ions enter a)
Electronic mail: [email protected] Electronic mail: [email protected]
b)
029012-1 Biointerphases 9(2), June 2014
the bacterial cell wall and disrupt the cytoplasm.9 Cu has been shown also to produce reactive oxygen species (highly oxidative radicals), leading to the damage of iron–sulfur enzymes. The mechanism of bacterial inactivation has not been completely worked out. Current research in the field of antimicrobial surfaces focuses on the incorporation of Ag, Cu, Zn, and TiO2 on medical devices, solid surfaces, textiles, and thin flexible polymers membranes. These antimicrobial agents inactivate bacteria, fungi, viruses, and algae due to the metal/oxide generating highly oxidative radicals on the cell wall.10 The ambient contamination by biofilms spreading bacteria for long times in hospitals, schools, and other public places require the preparation of adhesive antibacterial/antifungal thin films with an acceptable kinetics. Conventional methods such as precipitation or ion exchange result in a broad size distribution when depositing Ag, Cu on a variety of substrates. Up to date sputtering and PVD techniques are used to deposit thin film uniform nanoparticles with a narrow size distribution and significant bactericide action.11,12 To increase and stabilize the antibacterial
1934-8630/2014/9(2)/029012/8/$30.00
C 2014 American Vacuum Society V
029012-1
029012-2 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
activity of Cu-surfaces is important in order to decrease or eliminate completely the costly nosocomial hospital-acquired infection (HAI).13,14 The microbes causing HAI have two properties: many of them are toxic/resistant to antibiotics and they are able to survive on hospital surfaces for long times. These nosocomial infections due to antibiotic resistant bacteria are becoming more frequent during the last decade and contribute to the increasing of hospital care costs. The level of contamination of public hospitals in the United Kingdom and Switzerland has been found to be higher than the allowed level set for hospital rooms. For example, the contamination of 105 colony forming unit (CFU)/cm2 was observed in a diabetic wound dressing. But in the vicinity of the patient, a microbial density of 102 CFU/cm2 was found. Disinfecting surfaces were applied and strongly decreased the microbial density, since bacterial concentration was not high and regrowth of bacteria was not observed.15 This observation warrants the investigation of antibacterial surfaces/films. Cu-surfaces reported by our laboratory for Cu sputtered on cotton showing significant antibacterial activity.16 The present study addresses: (1) the optimization of Cu deposition by DC-magnetron sputtering on polyester as a follow up of textiles showing bactericide properties,17 (2) the investigation of the Cu-layers on polyester inducing E. coli inactivation in the dark as recently shown by Cu-textiles inactivating Methicillin-resistant Staphylococcus aureus,18 (3) the effect of actinic light on Cu/CuO–polyester when inactivating E. coli, (4) the stereomicroscopy in the dark and under light of during E. coli inactivation, (5) the XPS deconvolution of the Cu-oxidation states intervening in bacterial inactivation, and finally, (6) the Cu released during bacterial inactivation during the inactivation as monitored by inductively coupled plasma-mass spectroscopy (ICP-MS). Haenle et al.,19 have shown that Cu possess the most favorable ratio of bacterial activity/toxicity at a cytotoxic biocompatible level.
II. EXPERIMENT
029012-2
The 2 in. Cu-diameter was obtained from Lesker (Hastings, East Sussex, UK). The vacuum in the chamber was of 0.1 Pa, and the distance between the Cu-target and the polyester was 10 cm.16–18 The nominal film thickness was determined with a profilometer (Alphastep500, TENCOR) on silica wafers. Figure 1 shows the results of the Cu-sputtering for two different intensities. The samples prepared at 280 mA were found to provide the faster bacterial inactivation and the sample sputtered for 160 s attained a thickness of 70 nm. One atomic layer/cm2 has about 1015 atoms. Sputtering for 160 s led to a coating 70 nm thick, it follows that 2.2 layers (0.2 nm layer thickness) were deposited per second on the polyester leading to a deposition rate of 2.2 1015 atoms/cm2 s. The polyester used corresponds to the EMPA test cloth sample No. 407. This is a polyester Dacron polyethyleneterephthalate; type 54 spun, plain weave ISO 105-F04 used for color fastness determinations. The thermal stability of Dacron polyethylene terephthalate was 115 C and the thickness of the polyester was 6130 lm 610%. The Cu-content on the polyester was evaluated by x-ray fluorescence (XRF) in a PANalytical PW2400 spectrometer. The Cu films on polyester were resisting to friction with paper or cloth and did not smear Cu. B. Diffuse reflectance spectroscopy and x-ray diffraction
Diffuse reflectance spectroscopy was carried out in a Perkin Elmer Lambda 900 UV–VIS–NIR spectrometer within the wavelength range of 200–800 nm. The rough UV–Vis reflectance data cannot be used directly to assess the absorption of the Cu/CuO–polyester samples because of the large scattering contribution to the reflectance spectra. Normally, a weak dependence is assumed for the scattering coefficient S on the wavelength. The absorption of the samples was plotted in Kubelka–Munk units. The crystalline structure of the Cufilm was investigated by x-ray diffraction (XRD) INEL Model XRG instrument, power 3.5 kW and coupled with a detector CPS120 INEL to register peaks from h from 2 to 120 .
A. Sputtering of polyester, determination of Cu-coating thickness, and loadings
C. High-resolution transmission electron microscopy and fluorescence stereomicroscopy
The positive Ar-ions were accelerated toward the target by applying a voltage of 400 V and currents 100 mA–280 mA.
A Philips HRTEM CM 300 (field emission gun, 300 kV, 0.17 nm resolution) microscope and a Philips EM 430
FIG. 1. (Color online) Thickness calibration of Cu-coating sputtered on polyester: (1) DC-sputtered at 280 mA and (2) DC-sputtered at 100 mA. Biointerphases, Vol. 9, No. 2, June 2014
029012-3 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
(300 kV, LaB6, and 0.23 nm resolution) were used to measure the size of the Cu-nanoparticles. The textiles were embedded in epoxy resin (Embed 812) and the fabrics were cross-sectioned with an ultramicrotome (Ultracut E) up to a thin section of 70 nm. A magnification from about 6800 to 41 000 was used to report the results shown in Fig. 4. The fluorescence stereomicroscopy of polyester and Cusputtered polyester was carried out on samples inoculated with 108 CFU of E. coli and incubated in a humidified chamber to adhere the bacteria to the sample surface. This method can differentiate the live or dead bacteria using a simple fluorochrome-based staining method as described below. Then, the samples were washed with 10 ml milli-Q sterile water to remove the nonadherent bacteria. The inoculated polyester fabrics were stained using FilmtracerTM LIVE=DEADV Biofilm Viability Kit (Molecular Probes, Invitrogen). The kit contains a combination of the SYTOV 9 green fluorescent nucleic acid stain and propidium iodide fluorochromes for the staining of live and dead cells, respectively. The sample fluorescence was monitored with a fluorescence stereomicroscope (Leica MZ16 FA, Leica Microsystems GmbH Wetzlar, Germany), and the images were processed using the LAS v.1.7.0 build 1240 software from Leica Microsystems CMS GmbH. Adhesion of bacteria to the sputtered polyester was allowed for 2 min before washing the sample with sterile Milli-Q water to remove nonadherent bacteria. R
R
D. X-ray photoelectron spectroscopy
An AXIS NOVA photoelectron-spectrometer (Kratos Analytical, Manchester, UK) equipped with monochromatic AlKa (h ¼ 1486.6 eV) anode was used during the study. The carbon C1s line with position at 284.6 eV was used as a reference to correct the charging effects. The surface atomic concentration of some elements was determined from the peak areas using known sensitivity factors.20,21 Spectrum background was subtracted using the Shirley subtraction GL(30) program of the Kratos unit.22 The XPS Cu-species were analyzed by means of spectra deconvolution software (CasaXPS-Vision 2, Kratos Analytical UK). E. Bacterial counting, irradiation sources and ICP-MS and irradiation sources
The samples of E. coli (K12) was obtained from the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ) ATCC23716, Braunschweig, Germany, to test the antibacterial activity of the Cu–polyester sputtered fabrics. The polyester fabrics were sterilized by autoclaving at 121 C for 2 h. The 20 ll culture aliquots with an initial concentration of 106 CFU ml1 in NaCl/KCl (pH 7) were placed on coated and uncoated (control) polyester fabric. The initial concentration used in this study was high (106 CFU/ml), resembling to biofilm formation on surfaces in hospitals, schools, and public devices. The polyester is a microporous substrate and distributes the inoculum on the Cu-films without needing an adsorption stage. A Biointerphases, Vol. 9, No. 2, June 2014
029012-3
nonheterogeneous contact was established between the 20-ll E. coli solution and the Cu sample at 25–28 C. The samples were then placed on Petri dishes provided with a lid to prevent evaporation. After each determination, the fabric was transferred into a sterile 2 ml Eppendorf tube containing 1 ml autoclaved NaCl/KCl saline solution. This solution was subsequently mixed thoroughly using a Vortex for 3 min. Serial dilutions were made in NaCl/KCl solution. A 100-ll aliquot was pipetted onto a nutrient agar plate and then spread over the surface of the plate using standard plate method. Agar plates were incubated lid down, at 37 C for 24 h before colonies were counted. Three independent assays were done for each sputtered sample. Statistical analysis of the results was performed for the CFU data to determine the standard deviation values. The average values obtained were compared by one-way analysis of variance (ANOVA) and with the value of statistical significance. The one-way ANOVA was used to compare the mean of the samples using the Fisher distribution. The response was approximated for samples obtained from the photocatalytic inactivation of test samples presenting the same distribution within the same sputtering time. To verify that no regrowth of E. coli occurs after the first bacterial inactivation cycle, the Cu-film was incubated for 24 h at 37 C. Then, the bacterial suspension of 100 ll is deposited on three Petri dishes to obtain replicates. Thee samples were incubated at 37 C for 24 h. No bacterial regrowth was observed. Irradiation of the Cu/CuO–polyester was carried in a cavity by Osram Lumilux 18 W/827 actinic lamps widely used in hospital facilities with emission between 360 and 720 [see Fig. 1(b)] presenting an integrated light output of 4.1 mW/cm2. These lamps are commonly used in hospital indoor illumination. They present an efficient compromise of energy consumption and intensity of the emitted light. A FinniganTM ICPS-MS unit was used to detect the Cureleased during bacterial inactivation. This instrument was equipped with a double focusing reverse geometry mass spectrometer and presented an extremely low background signal and high ion-transmission coefficient. The detection limit was 0.2 ppb. III. RESULTS AND DISCUSSION A. XRF determination of the Cu-content, XRD, and diffuse reflectance spectra of Cu/CuO-polyester
By x-ray fluorescence the content of the Cu–polyester sputtered for 60 s, 160 s, and 300 s was determined as 0.05%, 0.11%, and 0.17% wt. Cu/wt. polyester. The diffuse reflectance spectra for the samples in Fig. 2 shows a wide spectral range for Cu/Cu(I,II)O species extending up to 740 nm. This corresponds to a band gap of 1.70 eV. Figure 2 shows a higher optical absorption for the Cu/CuO-polyester samples sputtered for longer times up to 300 s. The optical density increased for Cu loadings from 0.05% up to 0.17%. The optical absorption between 500 and 600 nm is due to the interband transition of Cu(I) species.
029012-4 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
FIG. 2. (Color online) Spectral distribution of the Osram Lumilux 18 W/827 light used during the course of this study.
The absorption between 600 and 740 nm is attributed to the exciton band and the Cu (II) d-d transition.23,24 Figure 3 presents the x-ray diffraction of polyester DC sputtered for 60 and 160 s. A higher amount of Cu was deposited after 160 s, leading to a higher signal as seen in trace (1), Fig. 3. The Cu/CuO clusters do not show sharp peaks in Fig. 3, since these clusters were not well crystallized due to: (1) the small size of the Cu/CuO nanoparticles (see Fig. 4) and (2) to the very low loading of the Cu/CuO of 0.05% for the 60 s sputtered samples and 0.11% for the nanoparticles sputtered for 160 s. The Miller indices (h, k, l) indexing for the three peaks shown in Fig. 3 were investigated. The high intense peak of the FCC Cu-oxide (111) reflection is noted in Fig. 3. B. Transmission electron microscopy (TEM) and x-ray photoelectron spectroscopy
Figure 4 presents the TEM results for polyester samples. In Fig. 4 for Cu/CuO-polyester, the epoxide and polyester present the same contrast. The denser Cu-particles coating show Cu/CuO nanoparticles with particle sizes of 15–25 nm.18 Figure 5 presents the XPS Cu2p doublet and the shakeup satellites deconvolution of Cu/CuO-polyester sputtered for
160 s at time zero. Figure 5 shows a high Cu2O (Cuþ) peak at 932.3 eV.21 Table I and Fig. 5 show a reduction of the Cu2O content on the polyester surface from 80% to 68% after 45 min irradiation. Concomitantly, the CuO (Cu2þ) peak at 934.32 eV increases from 13.67 to 31.76% after bacterial inactivation and the Cu0 at 931.5 eV of 6.67% initially present vanishes. This is the evidence for the charge transfer between the Cu/CuO-polyester and the bacteria due to redox reactions occurring during bacterial inactivation. Table II shows the variation of C, O, Cu, S, and N on polyester and Cu/CuO-polyester surfaces at different times during the bacterial inactivation. Addition of the bacteria at time zero increases the C-loading on the Cu/CuO-polyester. The C-content decreases after 30, 60 min, as shown in the first column in Table II, due to the bacterial inactivation process evolving CO2. In the third column of Table II, the Cu-content is seen to decrease from time zero as shown by the C-loading on the polyester during the disinfection processes. The levels of S, N due to the bacterial cell-wall destruction remain stable 1% during the bacterial disinfection process. This is an evidence for the rapid destruction of N, S residues on the Cu/CuO–polyester. Figure 6 shows the deconvolution of the O1s doublet showing changes for the O1s signal during bacterial inactivation. Figure 6(a) shows the predominant signal for Cu-O at 530.6 eV and the minor Cu peak at 528.5 eV. Figure 6(b) shows the CuO line shifted to 529.4 eV and the associated C¼O and C-O peaks due to bacterial oxidation at 530.8 eV and 532.2 eV. Figure 6(b) shows that hydroxylation during bacterial inactivation of the Cu/CuO-polyester occurs and give rise to the observed -OH peak at 532.2 eV [Fig. 6(b)].25 We suggest due to the XPS data (Tables I and II) that reactive oxygen species (ROS) on E. coli lead to bacterial inactivation and involved redox reactions as Cuþ þ O2• $ Cu0 þ O2 ;
(1)
Cu2þ þ e cb $ Cuþ ;
(2)
and possibly a two electron transfer from Cu2þ leading to Cu0 atoms Cu2þ þ 2e $ Cu0 :
FIG. 3. (Color online) Diffuse reflectance spectroscopy of Cu sputtered on polyester for (1) 300 s, (2) 160 s, and (3) 60 s. Biointerphases, Vol. 9, No. 2, June 2014
029012-4
(3)
The Cu-atoms then coalesce to Cu0 nanoparticles settling in the Cu-network of the polyester with Eredox ¼ 0.34 V versus normal hydrogen electrode.26 Figure 7(a) shows the bacterial inactivation on Cu/CuO–polyester under the light and in the dark as a function of the Cu-loading on polyester. Figure 7(a), trace 1, shows the complete CFU photo-killing within 45 min on Cu/CuO-polyester sputtered for 160 s with a loading of 0.11% wt. Cu/wt. polyester present. The 160 s sputtered sample present the highest amount of Cu-sites held in exposed positions interacting on the polyester surface with E. coli. Figure 7(a), trace 2, presents a slower bacterial inactivation kinetics for Cu/CuO–polyester samples sputtered for 300 s.
029012-5 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
029012-5
FIG. 4. XRD of Cu sputtered on polyester for (1) 60 s and (2) for 160 s.
It seems that an inward diffusion mechanism into the bulk at a thicker Cu coating takes place decreasing the surface free charges generated under light.27 The surface charges generated in the Cu/CuO-polyester samples are responsible for the electrostatic interaction of the sample with the bacteria.28 After sputtering Cu for 60 s, not enough Cu was sputtered on the polyester sample. Lower Cu-loadings led to longer bacterial inactivation times under light in Fig. 7(a), trace 3. The flexible, low cost, stable Cu/CuO-polyester kills bacteria in the dark. This is a finding presenting a wide application potential. Figure 7(a), trace 4, shows that the redox potential Cu-nanoparticles induce bacterial inactivation within reasonable times.7,10,16 Finally, Fig. 7(a), trace 5, shows no bacterial inactivation on the polyester surface under actinic light. The actinic Osram Lumilux 18 W/827 light for the runs reported in Fig. 7(a) with a dose of 4.1 mW/cm2 has been shown previously in Fig. 1(b). CuO under light irradiation is
a p-type semiconductor with a band gap of 1.7 eV, a flat-band potential of 0.3 V versus saturated calomel electrode (SCE) (pH 7) and a valence band of þ1.4 V SCE. Figure 7(b) shows the recycling of the Cu/CuO-polyester up to the fourth cycling under the actinic Osram Lumilux 827/18 W lamp. The data reported in Fig. 7(b) shows the stability up to the third cycling. Longer bacterial inactivation times were observed after the fourth cycle. The same pattern is shown below in Fig. 7(b) for runs in the dark. Figure 7(c) shows the release of bactericide Cunanoparticules (NPs) inactivating E. coli as a function of the sample up to the fourth cycle induced by Osram Lumilux 18 W/827 light. Figure 7(c) shows that Cu thicker layers sputtered for 300 s release Cu-ions at a higher level compared to materials sputtered for 160 s. The small amounts of toxic Cu NPs below 25 ppb/cm2 released by the Cu-polyester allow for a higher cytocompatibility compared to a similar Ag-concentration as reported for mammalian cells. Cu is a metabolizable agent compared to Ag remaining in the body after ingestion increasing the Ag-serum levels.29 Cu is considered not cytotoxic in ppb concentrations interacting with the bacteria through an oligodynamic effect.30 The Cu-ions have been reported to bind S, N, and COOelectron donor negative groups of the cell wall or entering the bacteria cytoplasma.31 Copper in the blood exists in two forms: bound to ceruloplasmin (85–95%), and the rest "free," loosely bound to albumin and small molecules. Free copper causes toxicity, as it generates ROS such as superoxide, hydrogen peroxide, and hydroxyl radicals. These oxidants damage proteins, lipids, and DNA.32 The copper toxicity toward mammalian cells have been reported with a median L(E)C50 of 25 mg/L for mammalian cells.30,32 TABLE I. Cu surface atomic percentage concentration detected by XPS for Cu/CuO–polester irradiated by actinic light and sputtered for 160 s.
FIG. 5. (Color online) Transmission electron microscopy (TEM) of bare polyester and Cu sputtered on polyester for 160 s. Biointerphases, Vol. 9, No. 2, June 2014
Before bacterial inactivation After bacterial inactivation
Cu(0)
Cu2O
CuO
6.67 0
80 68
13 31.76
029012-6 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
029012-6
TABLE II. XPS-surface atomic percentage concentration of elements on the Cu–polyester sputtered for 160 s as a function of time. Sample Polyester at time zero Cu–polyester (160 s) at time zero Cu–polyester þ bacteria at time zero Cu–polyester þ bacteria at time 30 min Cu–polyester þ bacteria at time 45 min
C
O
80.02 55.23 60.88 58.20 57.30
22.17 25.67 28.17 30.09 28.41
Cu
S
N
– – – 19.01 – – 11.00 0.21 0.19 10.74 0.98 1.06 12.08 1.02 1.30
C. Cell viability by fluorescence stereomicroscopy in the dark and under light for polyester and Cu/CuOpolyester
Figure 8(a) shows by fluorescence stereomicroscopy the images of the E. coli green living cells at zero and 150 min. The stable nature in the dark of E. coli up to 150 min on the surface of nonsputtered polyester was seen by the conservation of the green cells.18,33,34 By fluorescence stereomicroscopy, it is shown that bacteria entered into the polyester fabric and dispersing well on the fabric surface. This observation is consistent with the E. coli viability determination reported within this period by bacterial agar-plate CFU counting as shown in Fig. 7(a), trace (6). Figure 8(b) shows the green living cells (green dots) inactivation on Cu–polyester sputtered for 160 s in the dark. After 60 min, a growing number of dead cells (red dots) were observed along the green living cells (green dots). After 150 min, the initially inoculated bacteria (green dots) disappeared >98%. At this time, dead cells (red dots) were observed on the agar-plate as shown previously in Fig. 7(a), trace (4).
FIG. 6. (Color online) XPS Cu2p doublet and the shakeup satellites of Cu–polyester sputtered for 160 s: (a) before bacterial inactivation and (b) after bacterial inactivation. Biointerphases, Vol. 9, No. 2, June 2014
FIG. 7. (Color online) (a) XPS deconvolution of O1s envelope at time zero on Cu/CuO-polyester samples sputtered for 160 s: (a) at time zero and (b) after 45 min bacterial inactivation under actinic light irradiation. (b) E. coli inactivation on Cu sputtered on polyester for: (1) 160 s, (2) 300 s, (3) 60 s and irradiated under Osram Lumilux 18 W/827 light. (4) Cu sputtered for 160 s tested in the dark (5) nonsputtered polyester in the dark (6) nonsputtered polyester under light. (c) Recycling of Cu sputtered on polyester for 160 s inactivating bacteria for four cycles: (a) under low intensity light Osram Lumilux 18 W/827 irradiation and (b) in the dark. (d) Cu-release from Cu polyester sputtered for: (1) 300 s and (2) 160 s during the recycling cycles.
029012-7 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
029012-7
FIG. 8. (Color online) (a) Stereomicroscope images of E. coli cells at time zero (green dots) and after 150 min on uncoated polyester. (b) Stereomicroscope images of E. coli cells at time zero (green dots) on Cu/CuO-polyester in the dark at time zero and at times of 60 min and 150 min showing the progressive growth of dead bacteria (red dots). (c) Stereomicroscope images of E. coli cells at times of 15 min and 45 min when irradiating with Osram Lumilux 18 W/827 on Cu/CuO-polyester. The progressive transition of the living cells (green dots) to dead bacteria (red dots) is shown within this period.
Figure 8(c) shows the bacterial loss of viability on Cu/CuO polyester sputtered for 160 s under the Osram Lumilux 18 W/827 light irradiation. After 15 min, a loss of bacterial viability of 60% was observed. After 45 min, >98% of the initially inoculated bacteria appeared as dead cells (red dots). This last result is consistent with the E. coli agar-plate counting reported in Fig. 7(a), trace (1). The distribution and viability of the E. coli is presented in Figs. 8(a)–8(c) on the polyester and providing additional evidence for the significant bactericidal effect of the Cu/CuO–polyester. IV. CONCLUSIONS This study presents the first evidence for Cu/CuO–polyester adhesive and uniform films showing E. coli presenting evidence by stereomicroscopy for the bacterial loss of viability in the dark. Evidence is presented by XPS for redox processes on Cu/CuO–polyester surface and the variation of the Cu-oxidation states and Cu-surface availability within the time of bacterial inactivation. The bacterial inactivation in the dark seems to proceed due to the high oxidative redox potentials of the Cu-nanoparticles in contact with the bacteria outer cell wall. Under light irradiation, the CuO semiconductor accelerates the bacterial inactivation by a factor of 3 with respect to dark runs. Evidence is presented by ICP-MS for the slow amounts of Cu required for bacterial inactivation. The Cu-concentration required during the disinfection is cytocompatible since it proceeds at a level below 25 ppb/cm2. ACKNOWLEDGMENTS The authors thank the EPFL and Swiss National Science Foundation (SNF) Project (200021-143283/1) for financial Biointerphases, Vol. 9, No. 2, June 2014
support, the Interdisciplinary Center for Electron Microscopy (CIME) at the EPFL for their help with the electron microscopy. The authors also thank the COST Actions MP1101 and MP 1106 for interactive discussions during the course of this study.
1
T. Yuranova, A. Rincon, A. Bozzi, S. Parra, C. Pulgarin, and J. Kiwi, J Photochem. Photobiol. A: Chem. 161, 27 (2003). J. Kiwi and C. Pulgarin, Catal. Today 151, 2 (2010). 3 G. Borkow and J. Gabbay, Med. Hyp. 70, 990 (2008). 4 S. Malnick, R. Bardenstein, M. Huszar, J. Gabbay, and G. Borkow, J. Hosp. Infect. 70, 89 (2008). 5 G. Borkow and J. Gabbay, J. FASEB 188, 1728 (2004). 6 K. Kikuchi, K. Sunada, T. Yoda, K. Hashimoto, and A. Fujishima, J. Photochem. Photobiol. A 106, 51 (1997). 7 T. Watanabe and K. Hashimoto, Environ. Sci. Technol. 37, 4785 (2003). 8 T. Matsunaga, R. Tomoda, Y. Nakajima, N. Nakamura, and T. Komine, Appl. Environ. Microb. 54, 1330 (1988), available at http://aem.asm.org/ content/54/6/1330. 9 Y. Ohsumi, K. Kitamoto, and Y. Anraku, J. Bacteriol. 170, 2676 (1988), available at http://jb.asm.org/content/170/6/2676. 10 H. Foster. I. Ditta, and S. A. Varghese Appl. Microb. Biotechnol. 90, 1847 (2011). 11 L. Zhang, R. Dillert, D. Bahnemann, and M. Vormoor, Energy Environ Sci. 5, 7491 (2012). 12 P. Kelly and R. Arnell, Vacuum 56, 159 (2000). 13 A. Kramer, I. Schwebke, and G. Kampf BMC Infect. Diseases 6, 130 (2006). 14 K. Taylor, R. Roberts, and J. Roberts, “The challenge of hospital acquired infections (HAI),” Nat. Audit Office (2002). 15 K. Page, M. Wilson, and P. I. Parkin, J. Mater. Chem. 12, 3819 (2009). 16 C. Castro, R. Sanjines, C. Pulgarin, P. Osorio, A. Giraldo, and J. Kiwi, J. Photochem. Photobiol. A 216, 295 (2010). 17 O. Baghriche, A. Ehiasarian, E. Kusiak-Nejman, A. Morawski, C. Pulgarin, and R. Sanjines, J. Kiwi, J. Photochem. Photobiol. A 227, 11 (2012). 18 L. Rio, E. Kusiak, J. Kiwi, C. Pulgarin, A. Trampuz, and A. Bizzini, Appl. Env. Microb. 78, 8176 (2012). 2
029012-8 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces 19
C. Zietz, A. Fritsche, B. Fine, and M. Haenle, Bioinorg Chem Appl. 2012, 850390 (2012). 20 S. Rtimi, O. Baghriche, C. Pulgarin, J.-C. Lavanchy, and J. Kiwi, Surf. & Coat. Technol. 232, 804 (2013). 21 C. Wagner, W. Riggs, L. Davis, and G. Mullenberg, Handbook of X-ray Photoelectron Spectroscopy (Perkin-Elmer Corporation Physical Electronics Division, MN, 1979). 22 D. A. Shirley, Phys. Rev. B 5, 4709 (1982). 23 K. Hardee, A. Bard, and X. Electrodes, J. Electrochem. Soc. 124, 215 (1977). 24 See: http://en.wikipedia.org/wiki/Powder_diffraction 25 M. Dhananjeyan, E. Mielczarski, K. Thampi, Ph. Buffat, M. Bensimon, A. Kulik, J. Mielczarski, and J. Kiwi. J. Phys. Chem. B 105, 12046 (2001)
Biointerphases, Vol. 9, No. 2, June 2014
26
029012-8
M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions (Nace International, Cebelcor, Brussels, 1974). 27 J. Mathews, “Nucleation of thin films,” Epitaxial Growth, Part B, edited by W. Mathews (Academic, New York, 1975), Chap. 4, pp. 382–486. 28 A. Fujishima, X. Zhang, and D. Tryck, Surf. Sci. Rep. 63, 515 (2008). 29 F. Heidenau. W. Mittelmeier, R. Detsch, M. Haenle, F. Stenzel, G. Ziegler, and H. A. Gollwitzer, J. Mater. Sci. Mater. Med. 16, 883 (2005). 30 O. Bondarenko and J. Katre, Arch. Toxicol. 87, 1181 (2013). 31 J. Brewer, Clin. Neurophysiol. 121, 459 (2010). 32 See: http://www.opsi.gov.uk/si/si2000/20003184.htm#30. 33 M. Radzig, V. Nadtochenko, O. Koksharova, V. Lipasova, J. Kiwi, and I. Khmel, Colloids Surf. B 102, 300 (2013). 34 M. Arsianti, M. Lim, P. Marquis, and R. Amal, Langmuir 26, 7314 (2012).
Accelerated Escherichia coli inactivation in the dark on uniform copper flexible surfaces Sami Rtimi Ecole Polytechnique F ed erale de Lausanne, EPFL-SB-ISIC-GPAO, Station 6, CH-1015 Lausanne, Switzerland
Rosendo Sanjines Ecole Polytechnique F ed erale de Lausanne, EPFL-SB-IPMC-LNNME, Bat PH, Station 3, CH-1015 Lausanne, Switzerland
€l Bensimon Miche Ecole Polytechnique F ed erale de Lausanne, EPFL-ENAC-IIEGR-CEL, Bat GC, Station 18, CH-1015 Lausanne, Switzerland
sar Pulgarinb) Ce Ecole Polytechnique F ed erale de Lausanne, EPFL-SB-ISIC-GPAO, Station 6, CH-1015 Lausanne, Switzerland
John Kiwia) Ecole Polytechnique F ed erale de Lausanne, EPFL-SB-ISIC-LPI, Bat Chimie, Station 6, CH-1015, Lausanne, Switzerland
(Received 14 January 2014; accepted 26 March 2014; published 9 April 2014) The bacterial inactivation of Escherichia coli on Cu/CuO-polyester surfaces prepared by direct current magnetron sputtering was investigated in the dark and under actinic light (360 nm k 720 nm; 4.1 mW/cm2) as used commonly in hospital facilities. In the dark, complete bacterial inactivation (6log10 reduction) was observed within 150 min and under actinic light within 45 min. Sputtered samples led to nanoparticulate uniform Cu/CuO films 70 nm thick. The deposition rate used was 2.21015 atoms/cm2 s as determined by profilometry. X-ray fluorescence was used to determine the sample Cu-content and transmission electron microscopy determined Cu-particles 20 6 5 nm in size. The film optical absorption was observed to increase with Cu-content of the sample by diffuse reflection spectroscopy. The bacterial inactivation involved redox processes between Cu/CuO-polyester and the bacteria as observed by x-ray photoelectron spectroscopy. During sample recycling, the amount of Cu-release was determined by inductively coupled plasma-mass spectroscopy. The values required for E. coli inactivation were below the cytotoxicity level threshold allowed for mammalian cells. The E. coli inactivation by Cu/CuO-polyester seems to involve an oligodynamic effect since bacterial C 2014 American Vacuum Society. inactivation was achieved at very low Cu-concentrations. V [http://dx.doi.org/10.1116/1.4870596]
I. INTRODUCTION The manufacturing of high value added products such as bactericide textiles has increased rapidly in the last few years.1,2 Cu-particles are used for their catalytic and bactericide properties in the dark. The bactericide properties of Cutextiles have been reported recently by Borkow and Gabbay3–5 for textiles presenting antiviral, algicide, and fungicide properties and able to sustain an appropriate number of washing cycles loaded with Cu/CuO 1%. These textiles have been reported to be effective against Escherichia coli, Staphylococcus aureus, Pseudomonas aeruginosa, and Candida albicans, not causing skin irritation when externally applied. Cu has been known for a long time to have effective bactericide action.6,7 Cu-ions have been reported to be biocidal by binding to specific sites in the DNA.8 These ions enter a)
Electronic mail: [email protected] Electronic mail: [email protected]
b)
029012-1 Biointerphases 9(2), June 2014
the bacterial cell wall and disrupt the cytoplasm.9 Cu has been shown also to produce reactive oxygen species (highly oxidative radicals), leading to the damage of iron–sulfur enzymes. The mechanism of bacterial inactivation has not been completely worked out. Current research in the field of antimicrobial surfaces focuses on the incorporation of Ag, Cu, Zn, and TiO2 on medical devices, solid surfaces, textiles, and thin flexible polymers membranes. These antimicrobial agents inactivate bacteria, fungi, viruses, and algae due to the metal/oxide generating highly oxidative radicals on the cell wall.10 The ambient contamination by biofilms spreading bacteria for long times in hospitals, schools, and other public places require the preparation of adhesive antibacterial/antifungal thin films with an acceptable kinetics. Conventional methods such as precipitation or ion exchange result in a broad size distribution when depositing Ag, Cu on a variety of substrates. Up to date sputtering and PVD techniques are used to deposit thin film uniform nanoparticles with a narrow size distribution and significant bactericide action.11,12 To increase and stabilize the antibacterial
1934-8630/2014/9(2)/029012/8/$30.00
C 2014 American Vacuum Society V
029012-1
029012-2 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
activity of Cu-surfaces is important in order to decrease or eliminate completely the costly nosocomial hospital-acquired infection (HAI).13,14 The microbes causing HAI have two properties: many of them are toxic/resistant to antibiotics and they are able to survive on hospital surfaces for long times. These nosocomial infections due to antibiotic resistant bacteria are becoming more frequent during the last decade and contribute to the increasing of hospital care costs. The level of contamination of public hospitals in the United Kingdom and Switzerland has been found to be higher than the allowed level set for hospital rooms. For example, the contamination of 105 colony forming unit (CFU)/cm2 was observed in a diabetic wound dressing. But in the vicinity of the patient, a microbial density of 102 CFU/cm2 was found. Disinfecting surfaces were applied and strongly decreased the microbial density, since bacterial concentration was not high and regrowth of bacteria was not observed.15 This observation warrants the investigation of antibacterial surfaces/films. Cu-surfaces reported by our laboratory for Cu sputtered on cotton showing significant antibacterial activity.16 The present study addresses: (1) the optimization of Cu deposition by DC-magnetron sputtering on polyester as a follow up of textiles showing bactericide properties,17 (2) the investigation of the Cu-layers on polyester inducing E. coli inactivation in the dark as recently shown by Cu-textiles inactivating Methicillin-resistant Staphylococcus aureus,18 (3) the effect of actinic light on Cu/CuO–polyester when inactivating E. coli, (4) the stereomicroscopy in the dark and under light of during E. coli inactivation, (5) the XPS deconvolution of the Cu-oxidation states intervening in bacterial inactivation, and finally, (6) the Cu released during bacterial inactivation during the inactivation as monitored by inductively coupled plasma-mass spectroscopy (ICP-MS). Haenle et al.,19 have shown that Cu possess the most favorable ratio of bacterial activity/toxicity at a cytotoxic biocompatible level.
II. EXPERIMENT
029012-2
The 2 in. Cu-diameter was obtained from Lesker (Hastings, East Sussex, UK). The vacuum in the chamber was of 0.1 Pa, and the distance between the Cu-target and the polyester was 10 cm.16–18 The nominal film thickness was determined with a profilometer (Alphastep500, TENCOR) on silica wafers. Figure 1 shows the results of the Cu-sputtering for two different intensities. The samples prepared at 280 mA were found to provide the faster bacterial inactivation and the sample sputtered for 160 s attained a thickness of 70 nm. One atomic layer/cm2 has about 1015 atoms. Sputtering for 160 s led to a coating 70 nm thick, it follows that 2.2 layers (0.2 nm layer thickness) were deposited per second on the polyester leading to a deposition rate of 2.2 1015 atoms/cm2 s. The polyester used corresponds to the EMPA test cloth sample No. 407. This is a polyester Dacron polyethyleneterephthalate; type 54 spun, plain weave ISO 105-F04 used for color fastness determinations. The thermal stability of Dacron polyethylene terephthalate was 115 C and the thickness of the polyester was 6130 lm 610%. The Cu-content on the polyester was evaluated by x-ray fluorescence (XRF) in a PANalytical PW2400 spectrometer. The Cu films on polyester were resisting to friction with paper or cloth and did not smear Cu. B. Diffuse reflectance spectroscopy and x-ray diffraction
Diffuse reflectance spectroscopy was carried out in a Perkin Elmer Lambda 900 UV–VIS–NIR spectrometer within the wavelength range of 200–800 nm. The rough UV–Vis reflectance data cannot be used directly to assess the absorption of the Cu/CuO–polyester samples because of the large scattering contribution to the reflectance spectra. Normally, a weak dependence is assumed for the scattering coefficient S on the wavelength. The absorption of the samples was plotted in Kubelka–Munk units. The crystalline structure of the Cufilm was investigated by x-ray diffraction (XRD) INEL Model XRG instrument, power 3.5 kW and coupled with a detector CPS120 INEL to register peaks from h from 2 to 120 .
A. Sputtering of polyester, determination of Cu-coating thickness, and loadings
C. High-resolution transmission electron microscopy and fluorescence stereomicroscopy
The positive Ar-ions were accelerated toward the target by applying a voltage of 400 V and currents 100 mA–280 mA.
A Philips HRTEM CM 300 (field emission gun, 300 kV, 0.17 nm resolution) microscope and a Philips EM 430
FIG. 1. (Color online) Thickness calibration of Cu-coating sputtered on polyester: (1) DC-sputtered at 280 mA and (2) DC-sputtered at 100 mA. Biointerphases, Vol. 9, No. 2, June 2014
029012-3 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
(300 kV, LaB6, and 0.23 nm resolution) were used to measure the size of the Cu-nanoparticles. The textiles were embedded in epoxy resin (Embed 812) and the fabrics were cross-sectioned with an ultramicrotome (Ultracut E) up to a thin section of 70 nm. A magnification from about 6800 to 41 000 was used to report the results shown in Fig. 4. The fluorescence stereomicroscopy of polyester and Cusputtered polyester was carried out on samples inoculated with 108 CFU of E. coli and incubated in a humidified chamber to adhere the bacteria to the sample surface. This method can differentiate the live or dead bacteria using a simple fluorochrome-based staining method as described below. Then, the samples were washed with 10 ml milli-Q sterile water to remove the nonadherent bacteria. The inoculated polyester fabrics were stained using FilmtracerTM LIVE=DEADV Biofilm Viability Kit (Molecular Probes, Invitrogen). The kit contains a combination of the SYTOV 9 green fluorescent nucleic acid stain and propidium iodide fluorochromes for the staining of live and dead cells, respectively. The sample fluorescence was monitored with a fluorescence stereomicroscope (Leica MZ16 FA, Leica Microsystems GmbH Wetzlar, Germany), and the images were processed using the LAS v.1.7.0 build 1240 software from Leica Microsystems CMS GmbH. Adhesion of bacteria to the sputtered polyester was allowed for 2 min before washing the sample with sterile Milli-Q water to remove nonadherent bacteria. R
R
D. X-ray photoelectron spectroscopy
An AXIS NOVA photoelectron-spectrometer (Kratos Analytical, Manchester, UK) equipped with monochromatic AlKa (h ¼ 1486.6 eV) anode was used during the study. The carbon C1s line with position at 284.6 eV was used as a reference to correct the charging effects. The surface atomic concentration of some elements was determined from the peak areas using known sensitivity factors.20,21 Spectrum background was subtracted using the Shirley subtraction GL(30) program of the Kratos unit.22 The XPS Cu-species were analyzed by means of spectra deconvolution software (CasaXPS-Vision 2, Kratos Analytical UK). E. Bacterial counting, irradiation sources and ICP-MS and irradiation sources
The samples of E. coli (K12) was obtained from the Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH (DSMZ) ATCC23716, Braunschweig, Germany, to test the antibacterial activity of the Cu–polyester sputtered fabrics. The polyester fabrics were sterilized by autoclaving at 121 C for 2 h. The 20 ll culture aliquots with an initial concentration of 106 CFU ml1 in NaCl/KCl (pH 7) were placed on coated and uncoated (control) polyester fabric. The initial concentration used in this study was high (106 CFU/ml), resembling to biofilm formation on surfaces in hospitals, schools, and public devices. The polyester is a microporous substrate and distributes the inoculum on the Cu-films without needing an adsorption stage. A Biointerphases, Vol. 9, No. 2, June 2014
029012-3
nonheterogeneous contact was established between the 20-ll E. coli solution and the Cu sample at 25–28 C. The samples were then placed on Petri dishes provided with a lid to prevent evaporation. After each determination, the fabric was transferred into a sterile 2 ml Eppendorf tube containing 1 ml autoclaved NaCl/KCl saline solution. This solution was subsequently mixed thoroughly using a Vortex for 3 min. Serial dilutions were made in NaCl/KCl solution. A 100-ll aliquot was pipetted onto a nutrient agar plate and then spread over the surface of the plate using standard plate method. Agar plates were incubated lid down, at 37 C for 24 h before colonies were counted. Three independent assays were done for each sputtered sample. Statistical analysis of the results was performed for the CFU data to determine the standard deviation values. The average values obtained were compared by one-way analysis of variance (ANOVA) and with the value of statistical significance. The one-way ANOVA was used to compare the mean of the samples using the Fisher distribution. The response was approximated for samples obtained from the photocatalytic inactivation of test samples presenting the same distribution within the same sputtering time. To verify that no regrowth of E. coli occurs after the first bacterial inactivation cycle, the Cu-film was incubated for 24 h at 37 C. Then, the bacterial suspension of 100 ll is deposited on three Petri dishes to obtain replicates. Thee samples were incubated at 37 C for 24 h. No bacterial regrowth was observed. Irradiation of the Cu/CuO–polyester was carried in a cavity by Osram Lumilux 18 W/827 actinic lamps widely used in hospital facilities with emission between 360 and 720 [see Fig. 1(b)] presenting an integrated light output of 4.1 mW/cm2. These lamps are commonly used in hospital indoor illumination. They present an efficient compromise of energy consumption and intensity of the emitted light. A FinniganTM ICPS-MS unit was used to detect the Cureleased during bacterial inactivation. This instrument was equipped with a double focusing reverse geometry mass spectrometer and presented an extremely low background signal and high ion-transmission coefficient. The detection limit was 0.2 ppb. III. RESULTS AND DISCUSSION A. XRF determination of the Cu-content, XRD, and diffuse reflectance spectra of Cu/CuO-polyester
By x-ray fluorescence the content of the Cu–polyester sputtered for 60 s, 160 s, and 300 s was determined as 0.05%, 0.11%, and 0.17% wt. Cu/wt. polyester. The diffuse reflectance spectra for the samples in Fig. 2 shows a wide spectral range for Cu/Cu(I,II)O species extending up to 740 nm. This corresponds to a band gap of 1.70 eV. Figure 2 shows a higher optical absorption for the Cu/CuO-polyester samples sputtered for longer times up to 300 s. The optical density increased for Cu loadings from 0.05% up to 0.17%. The optical absorption between 500 and 600 nm is due to the interband transition of Cu(I) species.
029012-4 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
FIG. 2. (Color online) Spectral distribution of the Osram Lumilux 18 W/827 light used during the course of this study.
The absorption between 600 and 740 nm is attributed to the exciton band and the Cu (II) d-d transition.23,24 Figure 3 presents the x-ray diffraction of polyester DC sputtered for 60 and 160 s. A higher amount of Cu was deposited after 160 s, leading to a higher signal as seen in trace (1), Fig. 3. The Cu/CuO clusters do not show sharp peaks in Fig. 3, since these clusters were not well crystallized due to: (1) the small size of the Cu/CuO nanoparticles (see Fig. 4) and (2) to the very low loading of the Cu/CuO of 0.05% for the 60 s sputtered samples and 0.11% for the nanoparticles sputtered for 160 s. The Miller indices (h, k, l) indexing for the three peaks shown in Fig. 3 were investigated. The high intense peak of the FCC Cu-oxide (111) reflection is noted in Fig. 3. B. Transmission electron microscopy (TEM) and x-ray photoelectron spectroscopy
Figure 4 presents the TEM results for polyester samples. In Fig. 4 for Cu/CuO-polyester, the epoxide and polyester present the same contrast. The denser Cu-particles coating show Cu/CuO nanoparticles with particle sizes of 15–25 nm.18 Figure 5 presents the XPS Cu2p doublet and the shakeup satellites deconvolution of Cu/CuO-polyester sputtered for
160 s at time zero. Figure 5 shows a high Cu2O (Cuþ) peak at 932.3 eV.21 Table I and Fig. 5 show a reduction of the Cu2O content on the polyester surface from 80% to 68% after 45 min irradiation. Concomitantly, the CuO (Cu2þ) peak at 934.32 eV increases from 13.67 to 31.76% after bacterial inactivation and the Cu0 at 931.5 eV of 6.67% initially present vanishes. This is the evidence for the charge transfer between the Cu/CuO-polyester and the bacteria due to redox reactions occurring during bacterial inactivation. Table II shows the variation of C, O, Cu, S, and N on polyester and Cu/CuO-polyester surfaces at different times during the bacterial inactivation. Addition of the bacteria at time zero increases the C-loading on the Cu/CuO-polyester. The C-content decreases after 30, 60 min, as shown in the first column in Table II, due to the bacterial inactivation process evolving CO2. In the third column of Table II, the Cu-content is seen to decrease from time zero as shown by the C-loading on the polyester during the disinfection processes. The levels of S, N due to the bacterial cell-wall destruction remain stable 1% during the bacterial disinfection process. This is an evidence for the rapid destruction of N, S residues on the Cu/CuO–polyester. Figure 6 shows the deconvolution of the O1s doublet showing changes for the O1s signal during bacterial inactivation. Figure 6(a) shows the predominant signal for Cu-O at 530.6 eV and the minor Cu peak at 528.5 eV. Figure 6(b) shows the CuO line shifted to 529.4 eV and the associated C¼O and C-O peaks due to bacterial oxidation at 530.8 eV and 532.2 eV. Figure 6(b) shows that hydroxylation during bacterial inactivation of the Cu/CuO-polyester occurs and give rise to the observed -OH peak at 532.2 eV [Fig. 6(b)].25 We suggest due to the XPS data (Tables I and II) that reactive oxygen species (ROS) on E. coli lead to bacterial inactivation and involved redox reactions as Cuþ þ O2• $ Cu0 þ O2 ;
(1)
Cu2þ þ e cb $ Cuþ ;
(2)
and possibly a two electron transfer from Cu2þ leading to Cu0 atoms Cu2þ þ 2e $ Cu0 :
FIG. 3. (Color online) Diffuse reflectance spectroscopy of Cu sputtered on polyester for (1) 300 s, (2) 160 s, and (3) 60 s. Biointerphases, Vol. 9, No. 2, June 2014
029012-4
(3)
The Cu-atoms then coalesce to Cu0 nanoparticles settling in the Cu-network of the polyester with Eredox ¼ 0.34 V versus normal hydrogen electrode.26 Figure 7(a) shows the bacterial inactivation on Cu/CuO–polyester under the light and in the dark as a function of the Cu-loading on polyester. Figure 7(a), trace 1, shows the complete CFU photo-killing within 45 min on Cu/CuO-polyester sputtered for 160 s with a loading of 0.11% wt. Cu/wt. polyester present. The 160 s sputtered sample present the highest amount of Cu-sites held in exposed positions interacting on the polyester surface with E. coli. Figure 7(a), trace 2, presents a slower bacterial inactivation kinetics for Cu/CuO–polyester samples sputtered for 300 s.
029012-5 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
029012-5
FIG. 4. XRD of Cu sputtered on polyester for (1) 60 s and (2) for 160 s.
It seems that an inward diffusion mechanism into the bulk at a thicker Cu coating takes place decreasing the surface free charges generated under light.27 The surface charges generated in the Cu/CuO-polyester samples are responsible for the electrostatic interaction of the sample with the bacteria.28 After sputtering Cu for 60 s, not enough Cu was sputtered on the polyester sample. Lower Cu-loadings led to longer bacterial inactivation times under light in Fig. 7(a), trace 3. The flexible, low cost, stable Cu/CuO-polyester kills bacteria in the dark. This is a finding presenting a wide application potential. Figure 7(a), trace 4, shows that the redox potential Cu-nanoparticles induce bacterial inactivation within reasonable times.7,10,16 Finally, Fig. 7(a), trace 5, shows no bacterial inactivation on the polyester surface under actinic light. The actinic Osram Lumilux 18 W/827 light for the runs reported in Fig. 7(a) with a dose of 4.1 mW/cm2 has been shown previously in Fig. 1(b). CuO under light irradiation is
a p-type semiconductor with a band gap of 1.7 eV, a flat-band potential of 0.3 V versus saturated calomel electrode (SCE) (pH 7) and a valence band of þ1.4 V SCE. Figure 7(b) shows the recycling of the Cu/CuO-polyester up to the fourth cycling under the actinic Osram Lumilux 827/18 W lamp. The data reported in Fig. 7(b) shows the stability up to the third cycling. Longer bacterial inactivation times were observed after the fourth cycle. The same pattern is shown below in Fig. 7(b) for runs in the dark. Figure 7(c) shows the release of bactericide Cunanoparticules (NPs) inactivating E. coli as a function of the sample up to the fourth cycle induced by Osram Lumilux 18 W/827 light. Figure 7(c) shows that Cu thicker layers sputtered for 300 s release Cu-ions at a higher level compared to materials sputtered for 160 s. The small amounts of toxic Cu NPs below 25 ppb/cm2 released by the Cu-polyester allow for a higher cytocompatibility compared to a similar Ag-concentration as reported for mammalian cells. Cu is a metabolizable agent compared to Ag remaining in the body after ingestion increasing the Ag-serum levels.29 Cu is considered not cytotoxic in ppb concentrations interacting with the bacteria through an oligodynamic effect.30 The Cu-ions have been reported to bind S, N, and COOelectron donor negative groups of the cell wall or entering the bacteria cytoplasma.31 Copper in the blood exists in two forms: bound to ceruloplasmin (85–95%), and the rest "free," loosely bound to albumin and small molecules. Free copper causes toxicity, as it generates ROS such as superoxide, hydrogen peroxide, and hydroxyl radicals. These oxidants damage proteins, lipids, and DNA.32 The copper toxicity toward mammalian cells have been reported with a median L(E)C50 of 25 mg/L for mammalian cells.30,32 TABLE I. Cu surface atomic percentage concentration detected by XPS for Cu/CuO–polester irradiated by actinic light and sputtered for 160 s.
FIG. 5. (Color online) Transmission electron microscopy (TEM) of bare polyester and Cu sputtered on polyester for 160 s. Biointerphases, Vol. 9, No. 2, June 2014
Before bacterial inactivation After bacterial inactivation
Cu(0)
Cu2O
CuO
6.67 0
80 68
13 31.76
029012-6 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
029012-6
TABLE II. XPS-surface atomic percentage concentration of elements on the Cu–polyester sputtered for 160 s as a function of time. Sample Polyester at time zero Cu–polyester (160 s) at time zero Cu–polyester þ bacteria at time zero Cu–polyester þ bacteria at time 30 min Cu–polyester þ bacteria at time 45 min
C
O
80.02 55.23 60.88 58.20 57.30
22.17 25.67 28.17 30.09 28.41
Cu
S
N
– – – 19.01 – – 11.00 0.21 0.19 10.74 0.98 1.06 12.08 1.02 1.30
C. Cell viability by fluorescence stereomicroscopy in the dark and under light for polyester and Cu/CuOpolyester
Figure 8(a) shows by fluorescence stereomicroscopy the images of the E. coli green living cells at zero and 150 min. The stable nature in the dark of E. coli up to 150 min on the surface of nonsputtered polyester was seen by the conservation of the green cells.18,33,34 By fluorescence stereomicroscopy, it is shown that bacteria entered into the polyester fabric and dispersing well on the fabric surface. This observation is consistent with the E. coli viability determination reported within this period by bacterial agar-plate CFU counting as shown in Fig. 7(a), trace (6). Figure 8(b) shows the green living cells (green dots) inactivation on Cu–polyester sputtered for 160 s in the dark. After 60 min, a growing number of dead cells (red dots) were observed along the green living cells (green dots). After 150 min, the initially inoculated bacteria (green dots) disappeared >98%. At this time, dead cells (red dots) were observed on the agar-plate as shown previously in Fig. 7(a), trace (4).
FIG. 6. (Color online) XPS Cu2p doublet and the shakeup satellites of Cu–polyester sputtered for 160 s: (a) before bacterial inactivation and (b) after bacterial inactivation. Biointerphases, Vol. 9, No. 2, June 2014
FIG. 7. (Color online) (a) XPS deconvolution of O1s envelope at time zero on Cu/CuO-polyester samples sputtered for 160 s: (a) at time zero and (b) after 45 min bacterial inactivation under actinic light irradiation. (b) E. coli inactivation on Cu sputtered on polyester for: (1) 160 s, (2) 300 s, (3) 60 s and irradiated under Osram Lumilux 18 W/827 light. (4) Cu sputtered for 160 s tested in the dark (5) nonsputtered polyester in the dark (6) nonsputtered polyester under light. (c) Recycling of Cu sputtered on polyester for 160 s inactivating bacteria for four cycles: (a) under low intensity light Osram Lumilux 18 W/827 irradiation and (b) in the dark. (d) Cu-release from Cu polyester sputtered for: (1) 300 s and (2) 160 s during the recycling cycles.
029012-7 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces
029012-7
FIG. 8. (Color online) (a) Stereomicroscope images of E. coli cells at time zero (green dots) and after 150 min on uncoated polyester. (b) Stereomicroscope images of E. coli cells at time zero (green dots) on Cu/CuO-polyester in the dark at time zero and at times of 60 min and 150 min showing the progressive growth of dead bacteria (red dots). (c) Stereomicroscope images of E. coli cells at times of 15 min and 45 min when irradiating with Osram Lumilux 18 W/827 on Cu/CuO-polyester. The progressive transition of the living cells (green dots) to dead bacteria (red dots) is shown within this period.
Figure 8(c) shows the bacterial loss of viability on Cu/CuO polyester sputtered for 160 s under the Osram Lumilux 18 W/827 light irradiation. After 15 min, a loss of bacterial viability of 60% was observed. After 45 min, >98% of the initially inoculated bacteria appeared as dead cells (red dots). This last result is consistent with the E. coli agar-plate counting reported in Fig. 7(a), trace (1). The distribution and viability of the E. coli is presented in Figs. 8(a)–8(c) on the polyester and providing additional evidence for the significant bactericidal effect of the Cu/CuO–polyester. IV. CONCLUSIONS This study presents the first evidence for Cu/CuO–polyester adhesive and uniform films showing E. coli presenting evidence by stereomicroscopy for the bacterial loss of viability in the dark. Evidence is presented by XPS for redox processes on Cu/CuO–polyester surface and the variation of the Cu-oxidation states and Cu-surface availability within the time of bacterial inactivation. The bacterial inactivation in the dark seems to proceed due to the high oxidative redox potentials of the Cu-nanoparticles in contact with the bacteria outer cell wall. Under light irradiation, the CuO semiconductor accelerates the bacterial inactivation by a factor of 3 with respect to dark runs. Evidence is presented by ICP-MS for the slow amounts of Cu required for bacterial inactivation. The Cu-concentration required during the disinfection is cytocompatible since it proceeds at a level below 25 ppb/cm2. ACKNOWLEDGMENTS The authors thank the EPFL and Swiss National Science Foundation (SNF) Project (200021-143283/1) for financial Biointerphases, Vol. 9, No. 2, June 2014
support, the Interdisciplinary Center for Electron Microscopy (CIME) at the EPFL for their help with the electron microscopy. The authors also thank the COST Actions MP1101 and MP 1106 for interactive discussions during the course of this study.
1
T. Yuranova, A. Rincon, A. Bozzi, S. Parra, C. Pulgarin, and J. Kiwi, J Photochem. Photobiol. A: Chem. 161, 27 (2003). J. Kiwi and C. Pulgarin, Catal. Today 151, 2 (2010). 3 G. Borkow and J. Gabbay, Med. Hyp. 70, 990 (2008). 4 S. Malnick, R. Bardenstein, M. Huszar, J. Gabbay, and G. Borkow, J. Hosp. Infect. 70, 89 (2008). 5 G. Borkow and J. Gabbay, J. FASEB 188, 1728 (2004). 6 K. Kikuchi, K. Sunada, T. Yoda, K. Hashimoto, and A. Fujishima, J. Photochem. Photobiol. A 106, 51 (1997). 7 T. Watanabe and K. Hashimoto, Environ. Sci. Technol. 37, 4785 (2003). 8 T. Matsunaga, R. Tomoda, Y. Nakajima, N. Nakamura, and T. Komine, Appl. Environ. Microb. 54, 1330 (1988), available at http://aem.asm.org/ content/54/6/1330. 9 Y. Ohsumi, K. Kitamoto, and Y. Anraku, J. Bacteriol. 170, 2676 (1988), available at http://jb.asm.org/content/170/6/2676. 10 H. Foster. I. Ditta, and S. A. Varghese Appl. Microb. Biotechnol. 90, 1847 (2011). 11 L. Zhang, R. Dillert, D. Bahnemann, and M. Vormoor, Energy Environ Sci. 5, 7491 (2012). 12 P. Kelly and R. Arnell, Vacuum 56, 159 (2000). 13 A. Kramer, I. Schwebke, and G. Kampf BMC Infect. Diseases 6, 130 (2006). 14 K. Taylor, R. Roberts, and J. Roberts, “The challenge of hospital acquired infections (HAI),” Nat. Audit Office (2002). 15 K. Page, M. Wilson, and P. I. Parkin, J. Mater. Chem. 12, 3819 (2009). 16 C. Castro, R. Sanjines, C. Pulgarin, P. Osorio, A. Giraldo, and J. Kiwi, J. Photochem. Photobiol. A 216, 295 (2010). 17 O. Baghriche, A. Ehiasarian, E. Kusiak-Nejman, A. Morawski, C. Pulgarin, and R. Sanjines, J. Kiwi, J. Photochem. Photobiol. A 227, 11 (2012). 18 L. Rio, E. Kusiak, J. Kiwi, C. Pulgarin, A. Trampuz, and A. Bizzini, Appl. Env. Microb. 78, 8176 (2012). 2
029012-8 Rtimi et al.: Accelerated E. coli inactivation in the dark on uniform copper flexible surfaces 19
C. Zietz, A. Fritsche, B. Fine, and M. Haenle, Bioinorg Chem Appl. 2012, 850390 (2012). 20 S. Rtimi, O. Baghriche, C. Pulgarin, J.-C. Lavanchy, and J. Kiwi, Surf. & Coat. Technol. 232, 804 (2013). 21 C. Wagner, W. Riggs, L. Davis, and G. Mullenberg, Handbook of X-ray Photoelectron Spectroscopy (Perkin-Elmer Corporation Physical Electronics Division, MN, 1979). 22 D. A. Shirley, Phys. Rev. B 5, 4709 (1982). 23 K. Hardee, A. Bard, and X. Electrodes, J. Electrochem. Soc. 124, 215 (1977). 24 See: http://en.wikipedia.org/wiki/Powder_diffraction 25 M. Dhananjeyan, E. Mielczarski, K. Thampi, Ph. Buffat, M. Bensimon, A. Kulik, J. Mielczarski, and J. Kiwi. J. Phys. Chem. B 105, 12046 (2001)
Biointerphases, Vol. 9, No. 2, June 2014
26
029012-8
M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions (Nace International, Cebelcor, Brussels, 1974). 27 J. Mathews, “Nucleation of thin films,” Epitaxial Growth, Part B, edited by W. Mathews (Academic, New York, 1975), Chap. 4, pp. 382–486. 28 A. Fujishima, X. Zhang, and D. Tryck, Surf. Sci. Rep. 63, 515 (2008). 29 F. Heidenau. W. Mittelmeier, R. Detsch, M. Haenle, F. Stenzel, G. Ziegler, and H. A. Gollwitzer, J. Mater. Sci. Mater. Med. 16, 883 (2005). 30 O. Bondarenko and J. Katre, Arch. Toxicol. 87, 1181 (2013). 31 J. Brewer, Clin. Neurophysiol. 121, 459 (2010). 32 See: http://www.opsi.gov.uk/si/si2000/20003184.htm#30. 33 M. Radzig, V. Nadtochenko, O. Koksharova, V. Lipasova, J. Kiwi, and I. Khmel, Colloids Surf. B 102, 300 (2013). 34 M. Arsianti, M. Lim, P. Marquis, and R. Amal, Langmuir 26, 7314 (2012).
