Surface & Coatings Technology 186 (2004) 295 – 298 www.elsevier.com/locate/surfcoat
Surface modification of polyethylene terephthalate by plasma immersion ion implantation M. Ueda a,*, K.G. Kostov a,b, A.F. Beloto c, N.F. Leite c, K.G. Grigorov d a
Associated Plasma Laboratory, National Institute for Space Research INPE, P.O. Box 515, 12201-970 Sa˜o Jose´ dos Campos, SP, Brazil b Department of General Physics, Sofia University, 5 J. Bourchier Blvd., Sofia 1164, Bulgaria c Associated Laboratory of Material and Sensors, INPE, P.O. Box 515, 12201-970 Sa˜o Jose´ dos Campos, SP, Brazil d Institute of Electronics, Bulgarian Academy of Science BAS, 72 Tzarigradsko Chaussee Blvd., Sofia 1784, Bulgaria Available online
Abstract Plasma immersion ion implantation (PIII) of nitrogen has been successfully employed to form an amorphous carbon layer on the surface of 0.25-mm-thick polyethylene terephthalate (PET) sheet used for manufacturing plastic bottles. A DC glow discharge source with controlled floating plasma potential was used to create nitrogen plasma in a 100-l PIII system. The polymer specimens were pulsed (through a metallic grid or sample holder) at repetition rate of 300 Hz with high negative voltage pulse of 10 kV magnitude and 80 As duration. Formation of carbon film on the PET surface as a result of nitrogen ion implantation was investigated using Raman spectroscopy, optical and atomic force microscopy (AFM). The obtained Raman spectra reveal that the amorphous carbon layer has diamond-like characteristics. AFM micrographs demonstrate that after PIII treatment, the PET surface became much smoother and no cracks were found on it. D 2004 Elsevier B.V. All rights reserved. Keywords: Plasma immersion ion implantation; PET; Diamond-like carbon; RAMAN
1. Introduction Polyethylene terephthalate (PET) has been widely used as one of the important polymers for fibers, films, and food packaging materials because of its high melting temperature, good mechanical properties and recyclability. However, its thermal performance and gas barrier characteristics are not good enough to meet requirements for some particular applications such as hot-fill packaging and also for beer and some soft drink containers. The use of other polymer, such as polyethylene naphthalate (PEN), is effective to some extent for resolving the gas barrier problem, however the cost is higher and also the material is not suitable for recycling. Improvement of the oxygen and carbon dioxide barrier effectiveness of PET has recently attracted much attention and numerous techniques have been employed to fabricate plastic films with good gas-barrier properties. There are two general approaches to improve the oxygen barrier: chemical
* Corresponding author. Tel.: +55-12-3945-6676; fax: +55-12-39456710. E-mail address: [email protected]
(M. Ueda). 0257-8972/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.surfcoat.2004.03.035
coatings, which involves applying appropriate polymer materials and combination of layers , and physical – chemical methods, by which thin films (AlOx, TiOx, etc.) are formed by chemical or physical vapor deposition . However, the true effectiveness of such thin films is typically limited by the prevalence of defects and a tendency to crack easily when the plastic material is deformed. Diamond-like carbon (DLC) films have been the subject of intense research and development over the last 25 years, because of their excellent properties of high hardness; good wear resistance, very low friction coefficient, and good chemical inertness. Moreover, DLC application to gas barrier coating on plastic materials is expected to prevent the passing of oxygen and water vapor. Some authors have studied surface coating of PET with diamond-like carbon films in order to enhance the gas barrier characteristics. DLC films can be deposited by various methods, in particularly, by plasma source ion deposition/implantation using some hydrocarbon gases (C2H4, C2H2, CH4, etc.) [3– 8]. The oxygen transmission rate (OTR) of PET films covered with DLC was reduced significantly in comparison with the OTR of untreated PET samples [4,5,8]. However, coating technology presents a problem of DLC-deposit peeling.
M. Ueda et al. / Surface & Coatings Technology 186 (2004) 295–298
surface modification of polymers. Plasma enhanced chemical vapor deposition (PECVD) [4,5] and plasma-source ion implantation (PSII) of carbon [3,6 –8] have been successfully applied for PET coating surface with DLC film. Nitrogen plasma-based ion implantation  was used to convert the well-finished, 0.1-mm-thick PET sheet into DLC film. In this study, we have investigated the surface modification of industrial PET, utilized for manufacturing plastic bottles, into DLC by plasma immersion implantation of nitrogen.
2. Experimental setup
Fig. 1. AFM micrographs of (a) untreated and (b) PIII-treated PET.
Therefore, another approach for modification of PET surface layer into DLC by plasma immersion ion implantation of nitrogen was proposed . This surface modification process is a promising technique for improving the oxygen barrier, because the modified layer remains flexible and as such is not susceptive to cracking. Therefore, as-treated polymer sheets will not lose their gas barrier effectiveness even if bent or struck. Plasma immersion ion implantation (PIII) sometimes also called plasma-based ion implantation (PBII) or plasmasource ion implantation (PSII) is a rapidly advancing surface modification technology, recently developed, for improvement of surface properties of diverse materials, including semiconductors, metals or dielectrics. In PIII, the ions to be implanted into the near surface of the target are extracted directly from the plasma in which the samples to be processed are immersed without the need of extraction or acceleration grids. The three-dimensional ion implantation is achieved by applying a high negative voltage pulse (typically 10– 100 kV, 10– 50 As, 10– 3000 Hz repetition rate) to the sample holder or the component itself. In this configuration, ions from the plasma are accelerated and implanted into all surfaces of the sample without preference with regard to the sample orientation or complexity. However, until very recently, little attention has been paid to PIII
Nitrogen ion implantation into PET was carried out with the PIII experimental equipment reported elsewhere . The grounded vacuum vessel was evacuated to a base pressure of 6.5 10 3 Pa by using a turbomolecular pump. After filling the chamber with nitrogen, the working gas pressure was held at 0.2 Pa during the ion implantation. A DC glow discharge source with controlled floating plasma potential was used to create plasma in a 100-l PIII system. The 0.25-mm-thick PET sample, used for manufacturing plastic bottles, was set on an isolated stainless-steel holder in the center of the vacuum chamber. The entire area of the PET layer was fixed in contact with metal electrode. It was pulsed at repetition rate of 300 Hz with high negative voltage pulses with 10 kV amplitude and 80 As duration. For these experimental parameters, the estimated incident ion dose was on the order of 5.1017 ions/cm2. Ion implantation into insulating materials results in a charge accumulation on the surface, which in some cases can significantly reduce the energy of the implanted ions. To resolve this problem, in some experiments, a metal grid with about 60% transparency was placed over the PET surface and connected to the high voltage pulser. The chemical bonding characteristics of the PET substrates were examined by optical micro-Raman spectrosco-
Fig. 2. Raman spectra of implanted and unimplanted PET sample after subtracting the luminescence background.
M. Ueda et al. / Surface & Coatings Technology 186 (2004) 295–298
Fig. 3. Typical deconvoluted Raman spectrum using Gaussian fitting.
py (Renishaw 2000) using 514.5 nm Ar+ laser light. Surface morphology of the samples was analyzed by atomic force microscopy (AFM) operating the Shimadzu SPM-9500J3 nanoindenter in dynamic mode.
Raman spectra can greatly vary depending on DLC formation conditions, graphite cluster size, and the content of sp3/sp2 fractions in the sample [11,13]. Nevertheless, the precise sp3/sp2 ratio in hydrogen containing carbon films cannot be derived directly from the Raman spectra but some quantitative information still can be extracted. Independent measurements of sp3 fraction in hydrogenated carbon films performed by EELS and Raman spectra analysis [11,12] demonstrated that the G-peak position and the ratio of integral intensity I(D)/I(G) are related with the of sp3 bonding content . Relations between the carbon film sp3 bonding content and G-peak position and I(D)/I(G) ratio are plotted in Ref. . By using these empirical relations with the intensity ratio I(D)/I(G ) = 0.6 together and the G-peak position of 1550 cm 1, obtained from the fitting of Raman spectrum in Fig. 3, we can evaluate that the sp3 bond content of the modified PET sample is about 40%. The downshift of the D line to the extent we observe in our Raman spectra ( f 1340 cm 1) can be explained by bond-angle disorder in the sample . These observations confirm the modification of PET surface to amorphous DLC by the PIII process.
3. Results and discussion 4. Conclusions After treating the target by the PIII process, the PET layer changed its color from transparent to dark brownish. The three-dimensional AFM surface morphology of the PET samples before and after the implantation is shown in Fig. 1. The surface of the untreated PET sample is rough with many tiny peaks. After the implantation, the polymer surface became smoother and uniform, except for some blisters scattered on the surface. These larger features probably appear as results of the sample heating during the implantation or arcing. Fig. 2 shows typical Raman shift spectra of implanted and non-implanted PET with flat baseline. Sharp peaks at about 1300, 1600 and 1700 cm 1 are the characteristic peaks of PET itself. Along with a meaningful reduction of the characteristic peak intensity, the Raman spectrum of the implanted PET sample exhibits a broad shoulder from about 1300 cm 1 to approximately 1600 cm 1, which indicates that some structural modification of the PET surface occurred as a result of the ion bombardment. After removing the PET characteristic peaks for clarity, the resulting spectrum of PIII-treated PET is presented in Fig. 3. It exhibits characteristics very similar to those of the DLC films grown on PET substrates by plasma-based carbon deposition [3,6 – 8] and plasma immersion ion implantation . The spectrum was fitted with two Gaussian functions corresponding to a graphite-like peak (Gbond) at 1550 cm 1 and a disorder peak (D-bond) at 1340 cm 1, typical for DLC. Based on the fitting parameters, the peaks position, width and ratio of the integrated areas under the G and D peaks I(D)/I(G) were determined. The position, intensity, and the width of G and D peaks in
The plasma immersion implantation process has been applied to change the molecular bonding characteristics of PET using the impact of nitrogen ions. The polymer surface has been successfully converted into amorphous carbon layer as a result of the treatment. The modified layer was observed to be smooth and without cracks by AFM. The Raman shift spectra of the PIII-treated PET samples exhibit DLC characteristics with G-bond peak at 1550 cm 1 and disorder D-bond peak at 1340 cm 1. The content of sp3 fraction in the modified layer, which characterizes the DLC physical properties, was estimated by Raman spectrum analysis using the D and G peak area ratio and the G-peak position.
Acknowledgements This work is partially supported by Fundacßa˜o de Amparo a` Pesquisa do Estado de Sa˜o Paulo (FAPESP). One of the authors, K.G. Kostov, wishes to thank FAPESP for a visiting researcher scholarship, grant 02/08486-5.
References  F. Johansson, A. Leaufven, Packag. Technol. Sci. 7 (1994) 275.  T. Ohya, S. Suzuki, M. Yoshimoto, K. Iseki, S. Yokoyama, H. Ishihara, Society of Vacuum Coaters 43rd Annual Technical Conference Proc., Society of Vacuum Coaters, Japan, 2000, pp. 368.  T. Tanaka, M. Yoshida, M. Shinohara, T. Takagi, J. Vac. Sci. Technol., A, Vac. Surf. Films 20 (2002) 625.
M. Ueda et al. / Surface & Coatings Technology 186 (2004) 295–298
 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.  K. Inagawa, T. Zeniya, N. Hibino, Surf. Coat. Techol 169 – 170 (2003) 344.  S. Watanabe, M. Shinohara, H. Kodama, T. Tanaka, M. Yoshida, T. Tagaki, Thin Solid Films 420 – 421 (2002) 253.  M. Yoshida, T. Tanaka, S. Watanabe, M. Shinohara, J.W. Lee, T. Tagaki, Surf. Coat. Technol. 174 – 175 (2003) 1033.  P. Yang, S.C.H. Kwok, P.K. Chu, Y.X. Leng, J.Y. Chen, J. Wang, N. Huang, Nucl. Instrum. Methods Phys. Res., B Beam Interact. Mater. Atoms 206 (2003) 721.
 N. Sakudo, D. Mizutani, Y. Ohmura, H. Endo, R. Yoneda, N. Ikenaga, H. Takikawa, Nucl. Instrum. Methods Phys. Res., B Beam Interact. Mater. Atoms 206 (2003) 687.  M. Ueda, L.A. Berni, R.M. Castro, Surf. Coat. Technol. 156 (2002) 71.  A.C. Ferrari, Diam. Relat. Mater. 11 (2002) 1053.  J. Robertson, Mat. Sci. Eng., R Rep. 37 (2002) 129.  K. Baba, R. Hatada, Surf. Coat. Technol. 103 – 104 (1998) 235.  S. Prawer, K.W. Nugent, Y. Lifshitz, G.D. Lempert, E. Grossman, J. Kulik, I. Avigal, R. Kalish, Diam. Relat. Mater. 5 (1996) 433.