Bacterial repellence from polyethylene terephthalate surface modified by acetylene plasma immersion ion implantation–deposition

Surface & Coatings Technology 186 (2004) 299 – 304 www.elsevier.com/locate/surfcoat

Bacterial repellence from polyethylene terephthalate surface modified by acetylene plasma immersion ion implantation–deposition J. Wang a,b, N. Huang b, C.J. Pan b, S.C.H. Kwok a, P. Yang b, Y.X. Leng b, J.Y. Chen b, H. Sun b, G.J. Wan b, Z.Y. Liu c, P.K. Chu a,* a

Department of Physics and Materials Science, City University of Hong Kong, 88 Tat Chee Avenue, Kowloon, Hong Kong b School of Material Science and Engineering, Southwest Jiaotong University, Chengdu 610031, China c Affiliated Hospital of Medical College of Southeast University, Nanjing 210009, China Received 17 September 2003; accepted in revised form 17 February 2004 Available online 18 May 2004

Abstract There is an increasing interest in developing new methods to reduce bacterial adhesion onto polymeric materials used in biomedical implants. The antibacterial adsorption behavior on polyethylene terephthalate (PET) treated by plasma immersion ion implantation – deposition (PIII – D) using acetylene (C2H2) at different working pressures is investigated. The surface structure of the treated PET is determined by Rutherford backscattering spectrometry (RBS), laser Raman spectroscopy, and X-ray photoelectron spectroscopy (XPS). The results show the formation of thin hydrogenated amorphous carbon (a-C:H) films with different structures and chemical bonds on the PET surface. The ability of Staphylococcus aureus (SA) and Staphylococcus epidermidis (SE) to adhere to PET is quantitatively determined by plate counting and Gamma-ray counting of the 125I-labeled bacteria in vitro. The adhesion efficiency of SA on the a-C:H film deposited at 0.5 Pa of working pressure is about 16% of that on the untreated PET surface, and the adhered bacterial concentration of SE on the carbon film deposited at 1.0 Pa is about 1/6 of that of the PET surface. Bacterial adhesion onto a-C:H films is influenced by the structures and chemical bonds of the materials. The reduction in bacterial adhesion can be explained by the free energy of adhesion (DFAdh), which predicts whether microbial adhesion is energetically favorable (DFAdh < 0) or not (DFAdh>0). Our results show that bacterial adhesion is energetically unfavorable on the a-C:H films deposited at 0.5 and 1.0 Pa, and this study suggests one possible method to repel bacteria from polymeric surfaces. D 2004 Elsevier B.V. All rights reserved. Keywords: Plasma immersion ion implantation; Hydrogenated amorphous carbon (a-C:H); Polyethylene terephthalate (PET); Bacterial adhesion

1. Introduction Implantation of artificial organs and biomedical devices has become an important component of modern medical practices. However, problems may emerge from the implanted artificial organs and biomedical devices, and infection is by far one of the major clinical complications. In spite of non-septic conditions during the surgical process and systematic administration of antibiotics, bacterial infection is still prevalent [1,2]. Prevention of device-related infections remains a major dilemma in the delivery of quality medical care, and the problem causes high rates of mortality and * Corresponding author. Tel.: +852-27887724; fax: +852-27887830, +852-27889549. E-mail address: [email protected] (P.K. Chu). 0257-8972/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2004.02.046

morbidity and significant increases in health care costs [3– 5]. In particular, the occurrence of prosthetic valve endocarditis (PVE) is about 2 – 3% in patients with heart valve replacements, and Staphylococcus epidermidis (SE) accounts for about 30% of the overall infections [6]. Bacterial adhesion onto the prothetic and device surface is the first event in a series of host and organism reactions leading to PVE [7,8]. The adhesion is mediated by physicochemical interactions between the bacteria and biomaterials surface. Hence, surface modification of the biomaterials or devices is a relatively straightforward strategy for creating the desirable surfaces which will decrease the susceptibility to bacterial adhesion [9– 11]. Many surface modification techniques have been used to produce devices with anti-infective surfaces. Silver coatings [12 –14], surface-immobilized polyethylene oxide [15], sur-

300

J. Wang et al. / Surface & Coatings Technology 186 (2004) 299–304

Table 1 Instrumental parameters for the synthesis of the a-C:H films Sample Working Bias Pulse Pulse RF Time Thickness pressure voltage frequency width power (min) (nm) (Pa) (kV) (Hz) (As) (W) 1 2 3

0.5 1.0 2.0

5 5 5

100 100 100

20 20 20

300 300 300

40 40 40

102 195 245

face thiocyanation [16], and surfaces modified by various gas plasmas (such as oxygen and H2O) [17,18] have been proposed. Plasma immersion ion implantation – deposition (PIII – D) is an excellent method in this respect. In this technique, the specimens are surrounded by high-density plasma and pulse-biased to a high negative potential relative to the chamber wall. Ions generated in the overlying plasma are implanted into the samples. When conducted under the proper conditions, it is a non-line-of-sight process as opposed to conventional beam-line ion implantation. Most biomedical devices have sophisticated shape and PIII – D is thus a viable technique yielding good conformity and uniformity [19]. Biomedical polyethylene terephthalate (PET; DacronR) is commonly used to weave artificial heart valve sewing cuffs, and so its antibacterial property is very important. Tanaka et al. [20,21] have successfully coated PET films with amorphous carbon or diamond-like carbon (DLC) using acetylene (C2H2) PIII to improve the gas-barrier properties, but the research on amorphous carbon or diamond-like carbon deposited on PET for antibacterial adhesion has been scarcely reported [22]. In this work, amorphous hydrogenated amorphous carbon (a-C:H) films are fabricated on PET using acetylene plasma immersion ion implantation – deposition at different working pressures. The structures and physicochemical properties of films are characterized and the impact on bacterial adhesion is investigated. The interfacial free energies (DFAdh) of various kinds of bacterial cells on different substrates are calculated to explain the results of the bacterial adhesion.

etry (RBS), Raman spectroscopy, and X-ray photoelectron spectroscopy (XPS). The morphology studies were carried out on a Park Scientific Instrument Autoprobe Research System in the non-contact mode. The film thickness was measured with an Alpha-stepR-500 surface profiler. Bacteria were cultured and bacteria cell concentration was adjusted to 1
10 -5 cells ml -1 by dilution with PBS. Quantification of bacteria adhesion on the film is conducted by the plate counting method and Gamma-ray counting of 125 I-labeled bacteria. The contact angle was measured by the sessile drop technique at 25 jC using doubly distilled water, formamide and diiodomethane as the wetting agents. In the case of the bacterial cells, the measurements were performed on the bacterial layers deposited on membrane filters according to the method described by Busscher [23].

3. Results and discussion 3.1. Film structure The chemical composition of the films deposited at three different working pressures determined using RBS reveals that the films are primarily composed of C (Fig. 1). The carbon concentration is quite uniform with depth in all the samples, and the films are thicker at higher working pressure. The Raman spectra acquired from the carbon films are shown in Fig. 2. These spectra are consistent with the spectra of diamond-like carbon films [24]. They can be fitted employing two Gaussian peaks at about 1550 and 1350 cm-1, which correspond to the G and D bands of graphite, and the details are summarized in Table 2. With decreasing working pressure, the G peak position at f 1530 cm-1 mildly shifts toward the low frequencies and the D peak at f 1350 cm-1 becomes more dominant. The positions of the G and D bands and integrated

2. Experimental A 10-Am-thick PET film was laid on stainless-steel substrates attached to an insulated stainless-steel electrode in the center of the vacuum chamber. A negative voltage was then applied to the electrode. Some carbon layers were also deposited onto Si (100) wafers. These wafers were placed on the PET films on the stainless substrate to make the same electrical contact as the PET samples. Acetylene was bled into the chamber and the plasma was sustained by radio frequency (RF). A high negative pulsed voltage was applied to the sample holder. The detailed deposition parameters are listed in Table 1. The chemical composition and structural information were determined using Rutherford backscattering spectrom-

Fig. 1. RBS spectra obtained from the a-C:H films fabricated at different working pressures.

J. Wang et al. / Surface & Coatings Technology 186 (2004) 299–304

301

Fig. 2. Raman spectra of the a-C:H films fabricated at different working pressures.

intensity ratio (ID/IG) can be correlated with the sp2/sp3 bonding ratio [25], graphite cluster size [26], and disorder in the threefold coordinated islands [24]. The higher disorder with increasing working pressure is also reflected in the width of the D band that also increases with working pressure. Increasing in the ID/IG ratio, widening of the D-peak and narrowing of the G-peak are caused by increase of the graphite-like component in the amorphous carbon films [27]. The Raman results indicated that a-C:H films structure becomes more graphite like with decreasing working pressure. Fig. 3 shows the C1s XPS spectra with the curve fittings. As a-C:H is assumed [28] to consist of both sp3 and sp2, the C1s peak of an a-C:H film may consist of two groups of C1s photoelectron, one from carbon atoms in the sp3 configuration and the other from sp2, in addition to any C bonded to O. Hence, three peaks at 284.5, 285.2, and 286.4 eV are used here employing the Lorentzian– Gaussian functions. The peaks include contributions from sp2 carbon (CMC), sp3 carbon (CUC), and CUO. The sp3 hybridized C content in the films increases as a function of the working pressure. The sp3 content has a value of approximately 39.4% at a working pressure 0.5 Pa and rises to a maximum (52.1%) at 2.1 Pa. It is clear that the DLC film deposited at 0.5 Pa is more graphite like. Table 2 Experimental results derived from the Raman spectra Sample

Peak position (cm 1) D band

G band

PET-DLC1# PET-DLC2# PET-DLC3#

1341 1353 1352

1544 1554 1567

ID/IG 1.48 1.11 0.80

Fig. 3. Comparison of C1s XPS spectra of the a-C:H films fabricated at working pressures: (a) 0.5 Pa and (b) 2.1 Pa.

302

J. Wang et al. / Surface & Coatings Technology 186 (2004) 299–304

Fig. 4 shows the surface topography as measured by AFM without any filtering on a 3.5
3.5 Am2 area. The surface of the untreated PET sample exhibits rectangular structures and pinnacle-like structures can be observed in the image of the a-C:H film. Each AFM image is analyzed in terms of surface average roughness. The PET surface modified by acetylene PIII – D treatment shows dramatically reduced average roughness (Ra) from 33.1 to 3.1 nm. This result indicates that the surface morphology of PET film is significantly affected by the C2H2 PIII – D treatment. 3.2. Bacterial adhesion As shown in Fig. 5, Staphylococcus aureus (SA; ATCC6538) and S. epidermidis (ATCC8023) adhere in highest numbers on PET. In contrast, two kinds of bacteria have only minor adherence to amorphous carbon films deposited on PET by PIII – D. In particular, a small amount of bacteria can be observed on PET-DLC1# and PETDLC2#. This demonstrates that the amorphous carbon films can suppress the adhesion of two types of bacteria. The adhesion efficiency of SA on the a-C:H film deposited at 0.5 Pa of working pressure is about 16% of that on the untreated PET surface, and the adhered bacterial concentration of SE on the carbon film deposited at 1.0 Pa is about 1/6 of that of the PET surface. A scanning electron micrograph (SEM) of SA on a filter is shown in Fig. 6. The filter is completely and homogeneously covered with bacteria. The heterogeneity of the deposited bacterial layer is very small according to the water contact angle measured at seven different places on the same sample. The standard deviation of the mean is less than 2j for all the bacterial species investigated.

Fig. 4. AFM micrographs of: (a) PET control and (b) a-C:H film deposited at 2.1 Pa.

Fig. 5. Bacteria adhered to the untreated and the a-C:H films after incubation for 15 h revealed by SEM.

A series of contact angle data on a given surface yields the surface free energy components by fitting the data by the Lifshitz – van der Waals/acid – base approach (LW – AB) [29]. The values of the contact angles with the various liquids and surface energy components for the bacterial cells and substrates are summarized in Table 3. It can be seen that both SE and SA all are predominantly electron-donating, since the component cS is much higher than cS+. The free energy components of the untreated PET are in accordance with the results of Van Oss et al. [30], stating that most polymers have a cSLW c 40 mJ/m2 F 10%. Bacterial colonization on a surface is a complex process. The initial phase is bacterial adhesion to the biomaterials substrate. From a physical – chemical point of view, the adhesion of bacteria cells to a surface is determined by the interplay of electrostatic and hydrophobic/hydrophilic interactions. Because the zeta potential measurement shows that two kinds of bacteria on PET and carbon films have negative potential, bacterial adhesion is not mediated by

Fig. 6. SEM of S. aureus cells deposited on a cellulose triacetate filter.

J. Wang et al. / Surface & Coatings Technology 186 (2004) 299–304

303

Table 3 Contact angles (in degrees) and surface energy components of bacteria cells and supporting materials (in mJ/m2) Materials and bacterials

Contact angle

Surface energy components

Water

Formamide

Diiodomethane

cLW S

c+S

c S

cAB S

cS

SA SE PET PET-DLC1# PET-DLC2# PET-DLC2#

22.0 32.6 83.5 68.5 49.3 60.3

17.1 50.2 46.0 49.8 19.3 24.8

50.9 46.9 30.2 40.1 19.8 28.3

33.8 36.0 44.1 39.6 47.8 44.9

2.7 0.1 1.1 0.3 1.1 2.0

44.7 53.0 0.5 288.8 18.4 9.1

22.0 3.9 1.5 18.6 4.5 4.8

55.8 39.6 45.6 58.2 52.3 54.5

direct electrostatic interactions between the bacteria and substrate. The hydrophobic/hydrophilic interaction becomes the key factor affecting bacterial adhesion. A thermodynamic approach offers a powerful tool to predict bacterial adhesion to solid substrates [31]. On the basis of an interfacial free energy balance, neglecting electrical charge interactions, adhesion may be expected if DFAdh ¼ cSB  cSL  cBL < 0

ð1Þ

where DFAdh is the interfacial free energy of adhesion, cSB is the solid –bacterium interfacial free energy, cSL is the solid – liquid interfacial free energy, and cBL is the bacterium –liquid interfacial free energy. Adhesion is energetically unfavorable if

on PET and PET-DLC3#, but DFAdh is positive for the same bacterial strains on PET-DLC1# and PET-DLC2#. The most negative values are observed for the untreated PET. This result suggests that SA and SE are thermodynamically favorable on untreated PET and PET-DLC3#, whereas thermodynamically unfavorable on PET-DLC1# and PETDLC2#. On the basis of these results, a drastic reduction in bacterial adhesion of SA and SE to amorphous carbon film can be expected. This is confirmed by the results of adhesion tests as shown in Fig. 5. Our data also indicate that bacterial adhesion on the a-C:H film appears to be relative to the structure of film, and with decreasing sp3/sp2, the number of SA and SE adhered on the carbon film diminishes.

ð2Þ

DFAdh > 0:

4. Conclusion The following equation is used to determine the interfacial energy of bacteria adhesion to a solid surface [32].

DFAdh ¼

qffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffi2 qffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffi2 cLW cLW cLW  cLW B L S  S  qffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffi2  cLW cLW B  L qffiffiffiffiffi  pffiffiffiffiffi pffiffiffiffiffi pffiffiffiffiffi  þ2 cþ c c L L S þ cB  qffiffiffiffiffi qffiffiffiffiffi qffiffiffiffiffi pffiffiffiffiffi  þ cL cþ cþ cþ B  L S þ qffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffi þ   cþ c S cB  S cB

ð3Þ

Table 4 shows that the interfacial free energies of adhesion DFAdh are negative for SA and SE bacterial strains

Table 4 Interfacial free energies (DFAdh) of various kinds of bacteria cells on the PET control and PET-DLC solid surface Bacteria

SA SE

DFAdh (mJ/m2) PET

PET-DLC1#

PET-DLC2#

PET-DLC3#

 20.9  28.5

72.9 128.9

4.5 4.7

 6.7  8.5

Hydrogenated amorphous carbon films are fabricated on PET films by acetylene plasma immersion ion implantation – deposition at different working pressures. The RBS, Raman, and XPS results reveal that film graphitization is promoted at lower working pressure. The films deposited at 0.5 and 1.0 Pa working press are shown to significantly mitigate bacterial adhesion. The main physicochemical reason for bacteria repellence from the a-C:H films deposited by acetylene PIII –D on PET is DFAdh < 0. Bacterial adhesion on the a-C:H films is influenced by their structure and chemical bond, and our results suggest that the sp2/sp3 ratio an important role in the bacterial adhesion.

Acknowledgements This research was jointly and financially supported by City University of Hong Kong Strategic Research Grant (SRG) No. 7001447, Hong Kong Research Grants Council (RGC) Competitive Earmarked Research Grant (CERG) No. CityU 1137/03E, Hong Kong RGC/NSFC Joint Research Grant No. N-CityU 101/03, Natural Science Foundation of China (50203011), China Key Basic research (No. G 1999064706), and Southwest Jiaotong University Foundation (2002B02).

304

J. Wang et al. / Surface & Coatings Technology 186 (2004) 299–304

References [1] M. Pearson, Infect. Control Hosp. Epidemiol. 17 (1996) 438. [2] B. Sugarman, Am. J. Med. 81 (1A) (1986) 78. [3] M.A. Matin, M.A. Pfaller, R.P. Wenzel, Ann. Intern. Med. 110 (1989) 9. [4] D. Pittet, D. Tarara, R.P. Wenzel, JAMA 271 (1994) 1598. [5] P.M. Arnow, E.M. Quimosing, M. Brech, Clin. Infect. Dis. 16 (1993) 778. [6] M.G. Threlkeld, C.G. Cobbs, in: G.L. Mandell, R.G. Douglas, J. Bennett (Eds.), Principles and Practice of Infectious Disease, Churchill Livingstone, New York, 1995, p. 783. [7] T.S. Ivert, W.E. Dismukes, C.G. Cobbs, E.H. Blackstone, J.W. Kirklin, L.A. Bergdahl, Circulation 69 (1984) 223. [8] S.B. Calderwood, L.A. Swinski, C.M. Waternaux, A.W. Karchmer, M.J. Buckley, Circulation 72 (1985) 31. [9] J. Davenas, P. The´venard, F. Philippe, M.N. Arnaud, Biomol. Eng. 19 (2002) 263. [10] N.S. Sodhi Rana, V.P. Sahi, M.W. Mittelman, J. Electron Spectrosc. Relat. Phenom. 121 (2001) 249. [11] N.R. James, A. Jayakrishnan, Biomaterials 24 (2003) 2205. [12] J.E. Gray, P.R. Norton, R. Alnouno, C.L. Marolda, M.A. Valvano, K. Griffiths, Biomaterials 24 (2003) 2759. [13] U. Klueh, V. Wagner, S. Kelly, A. Johnson, J.D. Bryers, J. Biomed. Mater. Res. 53 (2000) 621. [14] J. Davenas, P. The´venard, F. Philippe, M.N. Arnaud, Biomol. Eng. 19 (2002) 263. [15] D.P. Ki, S.K. Young, K.H. Dong, H.K. Young, H.B. Lee, E.S. Hwal, S.C. Kyu, Biomaterials 19 (1998) 851.

[16] N.R. James, A. Jayakrishnan, Biomaterials 24 (2003) 2205. [17] K. Triandafillu, D.J. Balazs, B.-O. Aronsson, P. Descouts, P. Tu Quoc, C. van Delden, H.J. Mathieu, H. Harms, Biomaterials 24 (2003) 1507. [18] R.N.S. Sodhi, V.P. Sahi, M.W. Mittelman, J. Electron Spectrosc. Relat. Phenom. 121 (2001) 249. [19] P.K. Chu, J.Y. Chen, L.P. Wang, N. Huang, Mater. Sci. Eng. Rep. 36 (2002) 143. [20] M. Yoshida, S. Watanabe, T. Tanaka, T. Takagi, M. Shinohara, J.W. Lee, Nucl. Instrum. Methods Phys. Res., B Beam Interact. Mater. Atoms 206 (2003) 712. [21] S. Watanabe, M. Shinohara, H. Kodama, T. Tanaka, M. Yoshida, T. Takagi, Thin Solid Films 420 – 421 (2002) 253. [22] V.I. Ivanov-Omskii, L.K. Panina, S.G. Yastrebov, Carbon 38 (2000) 495. [23] H.J. Busscher, A.H. Weerkamp, H.C. Van Der Mei, A.W.J. Van Pelt, H.P. De Jong, Appl. Environ. Microbiol. 48 (5) (1984) 980. [24] J. Robertson, Mater. Sci. Eng., R Rep. 37 (2002) 129. [25] J. Robertson, E.P. O’Reilly, Phys. Rev., B 35 (1987) 2946. [26] J. Schwan, S. Ulrich, V. Batori, H. Ehrhardt, S.R.P. Silva, J. Appl. Phys. 80 (1996) 440. [27] L. Valentini, J.M. Kenny, G. Mariotto, P. Tosi, G. Carlotti, D. Fioretto, L. Lozzi, S. Santucci, Diam. Relat. Mater. 10 (2001) 1088. [28] J. Robertson, Prog. Solid State Chem. 21 (1991) 199. [29] P.K. Sharma, K.H. Rao, Adv. Colloid Interface Sci. 98 (2002) 341. [30] C.J. Van Oss, R.F. Giese, W. Wu, Adhesion 63 (1997) 71. [31] H.J. Busscher, A.H. Weerkamp, H.C. Van Der Mei, A.W.J. Van Pelt, H.P. De Jong, J. Arends, Appl. Environ. Microbiol. 48 (1984) 980. [32] R.P. Schender, J. Colloid Interface Sci. 182 (1996) 204.