Appl Phys A (2013) 112:143–148 DOI 10.1007/s00339-012-7216-8
Study of optical properties and biocompatibility of DLC films characterized by sp3 bonds Petr Písaˇrík · Miroslav Jelínek · Karel Smetana Jr. · Barbora Dvoˇránková · Tomáš Kocourek · Josef Zemek · Dagmar Chvostová
Received: 18 October 2011 / Published online: 5 September 2012 © Springer-Verlag 2012
Abstract Optical and biomedical properties of diamondlike carbon (DLC) films of various sp2 , sp3 bonds were studied. The layers were prepared by pulsed laser deposition (PLD) for laser energy densities from 4 J cm−2 to 14 J cm−2 . The percentage of sp2 and sp3 bonds was calculated using X-ray photoelectron spectroscopy (XPS). In dependence on density the films contained up to 70 % of sp3 bonds. Optical properties were measured using spectroscopic ellipsometry in region from 250 nm to 1000 nm (n = 2.6–2.7; k = 0.07–0.25) and by transmission measurement (from 200 nm to 1100 nm). The adhesion and growth of human fibroblasts and keratinocytes of DLC films were tested in vitro. 1 Introduction DLC films attaches great interest because of their unique physical properties as high hardness, negative electron affinP. Písaˇrík () · M. Jelínek · T. Kocourek · J. Zemek · D. Chvostová Institute of Physics ASCR v.v.i., Na Slovance 2, 182 21 Prague 8, Czech Republic e-mail:
[email protected] Fax: +420-286-890527 P. Písaˇrík · M. Jelínek · T. Kocourek Faculty of Biomedical Engineering, Czech Technical University in Prague, nám. Sítná 3105, 272 01 Kladno, Czech Republic K. Smetana Jr. · B. Dvoˇránková 1st Faculty of Medicine, Institute of Anatomy, Charles University, U nemocnice 3, 128 00 Prague 2, Czech Republic K. Smetana Jr. · B. Dvoˇránková 2nd Faculty of Medicine, Center of Cell Therapy and Tissue Repair, Charles University, V Úvalu 84, 150 06 Prague 5, Czech Republic
ity, good thermal conductivity, transparency, chemical stability, low friction coefficient, high wear resistance [1–3] and also biocompatibility [4–7]. DLC film consists of hybridization in different percentage of sp1 (acetylene-like), sp2 (graphite-like), and sp3 (diamond-like) bonding [3]. The content of sp2 , sp3 bonds exhibits influence on optical, electrical and mechanical properties [8]. It is supposed that sp2 and sp3 bonds can change also some biomedical properties. Biomedical tests of DLC layers have been performed by several authors [9–12]. These studies showed that the layers can be classified as biomaterials. There is no evidence that the DLC is toxic, exhibit inflammatory reactions or cause the loss of integrity of the cells. In some studies the attention was also paid to link of biological properties of DLC layers with sp2 , sp3 bonds [5–7, 12–19]. Ma [12, 13] studied absorption ratio of albumin: fibrinogen of hydrogenated DLC layers. Layers were deposited by PACVD with content of sp3 bonds 30–55 % (a higher albumin: fibrinogen ratio reduced thrombogenesis characteristics). The highest ratio of albumin:fibrinogen adsorption was found for a layer with 55 % sp3 bonds. Chen [14] studied blood compatibility of hydrogenated DLC films synthesized by plasma immersion ion implantation-deposition. He claimed that hemocompatibility is influenced by ratio of sp3 /sp2 , not by the absolute sp3 or sp2 content. The hemocompatibility became worse when sp3 /sp2 ratio increases. Logothetidis [15, 16] reached different results. He prepared a-C and a-C:H films were prepared by rf magnetron sputtering and the ta-C film using filtered cathodic vacuum arc deposition. He studied a-C:H films with the sp3 fraction of 20–45 %, and ta-C of 80 %. He found the optimum hemocompatibility for 45 % of sp3 fraction. Stan [17] has grown a-C:H films by RF CVD. The sp3 fraction was in the range of 49–52 %. Excellent results of activated partial thromboplastin time were obtained for the DLC with 52 % of sp3 bonds. Zhao [18] prepared DLC
144
layers by a magnetron sputtering with sp2 content from 25 % to 67 %. Bacterial attachment decreased with increasing ratio of sp3 /sp2 bonds. Jelínek [6, 19] studied cytotoxicity, adhesion and proliferation using human fibroblasts and keratinocytes. DLC layers with sp3 content in the region from 52 % to 62 % were prepared by PLD. The best adhesion of fibroblast was found for layers with 60 % of sp3 bonds. He also studied [5, 19] the biocompatibility of textile blood vessels covered with DLC layers by in vivo tests using the sheep. The sp3 content varied from 40 % to 60 %. Better results were achieved for higher sp3 content. We see that a-C and a-C:H layers were fabricated by several techniques, and for specific sp3 content a various biomedical properties were partly tested. In our contribution we study the optical and biological properties of films prepared by PLD for a wide range of energy densities, i.e. sp2 , sp3 bonds. The goal is to find the influence of sp3 content on physical (optical) properties and also on biomedical properties. The biological properties were evaluated by in vitro long-term cultivation of human fibroblasts and keratinocytes.
2 Experimental Deposition DLC films were prepared by KrF excimer PLD. Laser beam was focused onto a high purity graphite target with energy densities from 4 J cm−2 to 14 J cm−2 . Substrate (fused silica or Si) was placed in a distance of 45 mm from the target. The DLC films were created at room substrate temperature. The base vacuum of the coating system was 1 × 10−4 Pa. The films were deposited in argon ambient (0.25 Pa) [5, 7]. Before deposition the substrates were RF cleaned in 5 Pa of argon for two minutes. Characterization of films physical and optical properties The percentage of sp2 and sp3 hybridized carbon atoms was estimated by XPS. The C 1s spectra were measured by an ADES-400 photoelectron spectrometer (VG Scientific, UK) using Mg Kα excitation. Optical transmission was investigated by Shimadzu spectrometer UV 1601 (200–1100 nm). The refractive index n and extinction coefficient k were derived by a variable angle spectroscopic ellipsometry (VASE, J.A. Woollam Co.) working in rotating analyzer mode. In vitro test—the long-term cultivation of fibroblasts and keratinocytes For long-term cultivation tests the human fibroblasts and keratinocytes were isolated from residual skin from plastic surgery with the informed consent of donor [20]. The samples were sterilized at 100 ◦ C for one hour and placed to the 24-well plate. All procedures were performed under sterile conditions. Fibroblasts were seeded in the density of 1500 cells cm−2 and cultured in DMEM
P. Písaˇrík et al.
Fig. 1 Deconvoluted C 1s XPS spectra of DLC films (for two energy densities)
(Dulbecco’s modified Eagle’s medium) with 10 % fetal bovine serum for 14 days. Keratinocytes were seeded in the density of 50,000 cells cm−2 and cultured in the keratinocyte medium [20] for 8 days. Both cell types were cultured at 37 ◦ C and 5 % CO2 . Culture medium was changed every 2 days. Samples for immunohistochemical staining were fixed by paraformaldehyde and then stained with specific antibody for detection of fibronectin and keratin 10, panel of keratins and for keratin 14. The fluorescein isothiocyanate-labeled anti-sera were used to visualize the result of immunocytochemic reaction. The nuclei were counterstained by DAPI (4 ,6-diamidino-2-phenylindole), specific for DNA. Samples were analyzed by Nikon Eclipse 90i fluorescence microscope equipped with a CCD camera and the pictures were harvested using the computerassisted image analysis system LUCIA [21]. The toxicity of the tested samples was evaluated according to the survival of cells on the surface of tested material, including their morphology and fibronectin or keratin expression without statistical evaluation.
3 Results and discussion DLC films were fabricated for laser energy densities of 4, 6, 8, 10, 12 and 14 J cm−2 . Films prepared on Si substrates were used for thickness, ellipsometric and XPS measurements. Films prepared on fused silica were used for in vitro testing and for transmission measurements. Thickness of films was in the range from 30 nm (for in vitro tests) to 140–170 nm (other tests). XPS Only carbon and oxygen peaks were found in the recorded photoelectron spectra of the films. Therefore, nearsurface composition was quantified from peak areas of C 1s
Study of optical properties and biocompatibility of DLC films characterized by sp3 bonds Table 1 Composition and carbon atoms hybridization measurement using XPS (C 1s)
145
Laser energy density (J cm−2 )
Composition (XPS) Carbon (at%)
Oxygen (at%)
sp2
sp3
DLC-14
14
93.4
6.6
30
70
DLC-12
12
94.6
5.4
33
67
DLC-10
10
94.3
5.7
36
64
DLC-08
8
94.1
5.9
35
65
DLC-06
6
93.8
6.2
41
59
DLC-04
4
94.3
5.7
42
58
Sample
and O 1s transitions corrected for the photoelectron crosssections [22], the inelastic mean free paths [23], and the transmission function [24] of the spectrometer used. Details of the procedure are described in [6] and the results are summarized in Table 1. The percentage of the sp3 hybridized carbon atoms in the analyzed volume of the films was revealed from a peak-fit of the high-resolution C 1s spectra. As illustrated in Fig. 1 three Gaussian lines correspond to the C sp2 bonding peaked at 284.3 eV, C sp3 at 285.2 eV, and the C–O at about 286 eV. Although the C sp2 peak position and C–O contribution agree well with the data of the literature for graphite [25], the C sp3 peak position can hardly be compared to the C 1s peak position for diamond because the latter strongly depends on the Fermi level position of diamond. Therefore, we rely on the C 1s binding energy shifts between C sp2 and sp3 contributions calculated for amorphous carbon [26]. As shown in Table 1, the content of sp3 bonds increases with the laser energy density. The resolution depth of XPS was 4.8 nm [6], the accuracy of measurement was about 10 % [27]. Optical properties There is claimed that optical properties of DLC thin films depend strongly on sp2 , sp3 hybridization and on the amount of hydrogen, but no numerical data are given [28]. Some contributions [1, 3, 29–37] give information on transmission and optical constants but without connection to sp2 , sp3 bonds. We measured the transmission and optical constants of a-C films in relation to various values of sp3 , sp2 bonds: (a) Transmission In general, optical transmittance is higher for samples with higher sp3 fraction [3]. Hydrogenated films prepared by CVD [28, 29] exhibit higher transparency compared to hydrogen-free films prepared by PLD [3, 31, 32]. CVD films are characterized by high slope of optical transmission at the beginning of the visible region, while PLD films exhibit gradual rise over the region [3, 28–31]. The transmission spectrum prepared by MW CVD of methane and ethane [29] exhibited transmission (T > 80 %) in the range 500-900 nm. The best transmission (about 90 %) in the range 400–2000 nm was measured for films prepared by remote plasma enhancement CVD [38].
XPS (C 1s)
Fig. 2 Wavelength dependence of transmission
The hydrogen-free films prepared by PLD exhibited increasing transparency from 300 nm to 800 nm (up to 80 %) with a slow decrease at longer wavelength [3]. The transmittance spectra of our DLC films are shown in Fig. 2. Transmission gradually increases from 200 nm to maximum at 700–1000 nm (depending on energy density). For films created at lower densities (4 J cm−2 ) the maximum transmission was lower. With increasing density the transmission was higher and the peak moved to ultraviolet region. The film deposited at 14 J cm−2 showed the highest transmission (87 %) while the lowest transmission (72 %) was obtained from the film deposited at 4 J cm−2 . In the transmission curves we observe the undulation of rise slope which is changing with density—see Fig. 2. A similar behavior was observed also in [3]. The reason is unclear. (b) Spectroscopic ellipsometry of a-C:H was studied in [33–35]. The maximum refractive index n(λ) was 1.8 [33] or 2.2 [34, 35]. Spectroscopic ellipsometry of hydrogen-free DLC films was investigated in [36, 37]. They studied [36] the influence of sp3 bonds. The variables n and k exhibited monotonic behavior. Films with sp3 > 50 % were characterized by a refraction index n(λ) of ∼2.5 and an extinction coefficient k(λ) of ∼0.03 for λ = 600 nm. Films with sp3 < 50 % were characterized by strongly decreasing n with decreasing λ while k remained almost constant. In [37], n and k exhibited maximum at ∼2.75 or ∼0.7.
146
Our ellipsometric measurement was carried out in the spectral range 250–1000 nm at three incidence angles 65◦ , 70◦ , 75◦ at room temperature. Ellipsometric angles ψ and Δ are related to the complex ratio of the Fresnel reflection coefficients Rp and Rs for the polarization states p and s by Rp /Rs = tan(ψ) exp(iΔ). Experimental data ψ and Δ were fitted using several dispersion models for di-
Fig. 3 Wavelength dependence of refractive index n and coefficient extinction k
Fig. 4 Immunohistochemical detection of fibronectin (green signal), produced by fibroblasts cultured on tested surfaces after 14 days of cultivation. Cell nuclei stained with DAPI (blue signal). The extensive production of extracellular matrix-rich for fibronectin is visible. Magnification 200 × 336
P. Písaˇrík et al.
electric function of DLC material and the optical constants and thickness of layer were determined. Figure 3 shows behavior of refractive index n(λ) and extinction coefficient k(λ) of our DLC films. The refractive index n is in the range of 2.53–2.72. The maximum is shifted with the increasing content of sp3 bonds from 375 nm to 580 nm. The extinction coefficient k(λ) is monotonically decreasing with wavelength and decreasing with laser energy density. Our results are comparable to or better than the data of the literature [33–37]. Our measurement was also performed on a wider scale of sp3 . In vitro test—the long-term cultivation of fibroblasts and keratinocytes Used substrate (fused silica) or carbon films, deposited on it, were not cytotoxic and allowed cell adhesion and growth. The continuous growth of cells was observed on all DLC films during the long-term fibroblast cultivation. Fibroblasts produced the extracellular matrix containing fibronectin (Fig. 4) that indicated their physiological status. This molecule is known to be a substantial molecule of extracellular matrix in vivo. There were no significant differences in the phenotype of keratinocytes on experimental DLC films and on tissue grade polystyrene used as a control. Immunohistochemical detection of keratins demonstrated the favorable formation of colonies of keratinocytes on surface of film created at 8 J cm−2 (Fig. 5). The cells exhibiting keratin 14 (typical for proliferating cells) were also observed only over the surface of DLC-08 film (Fig. 6). On the other hand only a few keratinocytes were terminally differentiated exhibiting keratin 10 in their cytoplasma (Fig. 6). The best results of cell cultivation were achieved using film prepared at 8 J cm−2 . The explanation of this positive effect
Study of optical properties and biocompatibility of DLC films characterized by sp3 bonds
147
Fig. 5 Immunohistochemical detection of keratins (green signal; antibody against pankeratin) in keratinocytes after 8 days of cultivation, cell nuclei stained with DAPI (blue signal). The large colonies were detected on films DLC-08 in comparison with other surfaces (DLC-04, DLC-10 and fused silica). Magnification 200× (DLC-08) and 100× (DLC-04, DLC-10 and fused silica)
Fig. 6 Immunohistochemical detection of cytokeratin 10 (terminally differentiated cells) and cytokeratin 14 (cells able to proliferate) in cytoplasm of keratinocytes adhering to the layer DLC-08 after 8 days of cultivation (green signal), cell nuclei stained with DAPI (blue signal). Magnification 200×
of DLC-08 films on growth and differentiation of human keratinocytes needs further investigation but, theoretically, the surface properties of this sample can influence the spectrum and conformation of adsorbed serum proteins [38] that can positively influence the growth of epithelial cells [39].
4 Conclusions Hydrogen-free diamond-like carbon films with different content of sp3 bonds were prepared by PLD using energy densities from 4 J cm−2 to 14 J cm−2 . Maximum of sp3 bonds (70 %) was measured by XPS for energy density of 14 J cm−2 . Optical properties depend on the content of sp3 bonds. Transmission increased with of sp3 bonds. The best transmission (87 %) was achieved for the energy density of
14 J cm−2 . Our films exhibited higher refractive index n (2.63–2.72) and lower extinction coefficient k (0.07–0.25) at 600 nm compared to published data [36, 37]. The measurement was also performed on a wider scale of sp3 bonds. DLC films were not cytotoxic. During the long-term fibroblast cultivation fibronectin was produced in all cases. Nevertheless the growth of keratinocytes expressing keratin 14 was achieved predominantly on layer prepared at 8 J cm−2 , i.e. with sp3 bonds of 67 %. We proved that physical properties (transmission and n, k coefficients) of DLC films are varied with sp3 bonds. DLC films are biocompatible and we can indicate the influence of sp3 content on their biological behavior. Acknowledgements We thank the Institutional research plan AV010100522 and grant of the Czech Technical University in Prague No. SGS10/222/OHK4/2T/17 and No. SGS11/078/OHK4/1T/17.
148
References 1. J. Robertson, Mater. Sci. Eng. R. 37(4–6), 129 (2002) 2. A. Grill, Diam. Relat. Mater. 8(2–5), 428 (1999) 3. X.-L. Ding, Q.-S. Li, X.-H. Kong, Chin. Phys. Lett. 26(2), 027802 (2009) 4. J. Podlaha, M. Dvoˇrák, V. Žižková, R. Dvoˇrák, R. Kabeš, M. Jelínek, K. Veselý, Acta Vet. Brno 78(1), 115 (2009) 5. T. Kocourek, M. Jelínek, V. Vorlíˇcek, J. Zemek, T. Janˇca, V. Žížková, J. Podlaha, C. Popov, Appl. Phys. A 93(3), 627 (2008) 6. M. Jelínek, K. Smetana, T. Kocourek, B. Dvoˇránková, J. Zemek, J. Remsa, T. Luxbacher, Mater. Sci. Eng. B 169(1–3), 89 (2010) 7. M. Jelínek, T. Kocourek, J. Remsa, J. Mikšovský, J. Zemek, K. Smetana, B. Dvoˇránková, T. Luxbacher, Appl. Phys. A 101(4), 579 (2010) 8. N. Dwivedi, S. Kumar, H.K. Malik Govind, C.M.S. Rauthan, O.S. Panwar, Appl. Surf. Sci. 257(15), 6804 (2011) 9. V. Karagkiozaki, S. Logothetidis, A. Laskarakis, G. Giannoglou, S. Lousinian, Mater. Sci. Eng. B, Solid-State Mater. Adv. Technol. 152(1–3), 16 (2008) 10. S.C.H. Kwok, P.C.T. Ha, D.R. McKenzie, M.M.M. Bilek, P.K. Chu, Diam. Relat. Mater. 15(4–8), 893 (2006) 11. T. Hasebe, T. Ishimaru, A. Kamijo, Y. Yoshimoto, T. Yoshimura, S. Yohena, H. Kodama, A. Hotta, K. Takahashi, T. Suzuki, Diam. Relat. Mater. 16(4–7), 1343 (2007) (Spec. iss.) 12. W.J. Ma, A.J. Ruys, R.S. Mason, P.J. Martin, A. Bendavid, Z. Liu, M. Ionescu, H. Zreiqat, Biomaterials 28(9), 1620 (2007) 13. W.J. Ma, A.J. Ruys, H. Zreiqat, R. Mason, S.P. Ringer, Z.W. Liu, V. Keast, P.J. Martin, A. Bendavid, ICONN 2006 art. no. 4143385, p. 247 (2006) 14. J.Y. Chen, L.P. Wang, K.Y. Fu, N. Huang, Y. Leng, Y.X. Leng, P. Yang, J. Wang, G.J. Wan, H. Sun, X.B. Tian, P.K. Chu, Surf. Coat. Technol. 156(1–3), 289 (2002) 15. S. Logothetidis, Diam. Relat. Mater. 16(10), 1847 (2007) 16. S. Logothetidis, M. Gioti, S. Lousinian, S. Fotiadou, Thin Solid Films 482(1–2), 126 (2005) 17. G.E. Stan, D.A. Marcov, A.C. Popa, M.A. Husanu, J. Nanomater. Biostruct. 5(3), 705 (2010) 18. Q. Zhao, Y. Liu, C. Wang, S. Wang, Diam. Relat. Mater. 16(8), 1682 (2007) 19. M. Jelinek, J. Podlaha, T. Kocourek, V. Žížková, IFMBE Proc. 22, 2173 (2008) ˇ ˇ Vlˇcek, B. Dvoˇránková, 20. H. Strnad, L. Lacina, M. Koláˇr, Z. Cada, C. J. Betka, J. Plzák, M. Chovanec, J. Šáchová, J. Valach, M. Urbanová, K. Smetana, Cell Biol. 133(2), 201 (2010)
P. Písaˇrík et al. 21. J. Vacík, B. Dvoˇránková, J. Michálek, M. Pˇrádný, E. Krumbholcová, T. Fenclová, K. Smetana, J. Mater. Sci. Mater. Med. 19(2), 883 (2008) 22. I.M. Band, L.Y. Kharitonov, M.B. Trzhaskovskaya, At. Data Nucl. Data Tables 23(5), 443 (1979) 23. S. Tanuma, C.J. Powell, D.R. Penn, Surf. Interface Anal. 43(3), 689 (2011) 24. P. Jiˇríˇcek, Czechoslov. J. Phys. 44(3), 261 (1994) 25. J.F. Moulder, W.F. Sticle, P.E. Sobol, K.D. Bomben, Handbook of X-ray Photoelectron Spectroscopy (Perkin-Elmer, Eden Prairie, 1992) 26. J.T. Titantah, D. Lamoen, Carbon 43(6), 1311 (2005) 27. J. Zemek, J. Zalman, A. Luches, Appl. Surf. Sci. 133(1–2), 27 (1998) 28. D.K. Rai, D. Datta, R. Gupta, S. Kumar, Phys. Status Solidi C 5(5), 1198 (2008) 29. S. Adhikary, X.M. Tian, S. Adhikari, A.M.M. Omer, H. Uchida, M. Umeno, Diam. Relat. Mater. 14(11–12), 1832 (2005) 30. M.-S. Hwang, E.-T. Kim, J.-B. Kim, C. Lee, Diam. Relat. Mater. 10(11), 2063 (2001) 31. P.S. Banks, L. Dinh, B.C. Stuart, M.D. Feit, A.M. Komashko, A.M. Rubenchik, M.D. Perry, W. McLean, Appl. Phys. A 69(7), 347 (1999) 32. T. Nakamiya, S. Aoqui, K. Ebihara, Diam. Relat. Mater. 10(3–7), 905 (2001) 33. D.K. Rai, D. Datta, S.K. Ram, S. Sarkar, R. Gupta, S. Kumar, Solid State Sci. 12(8), 1449 (2010) 34. N. Kato, H. Mori, N. Takahashi, Phys. Status Solidi C 5(5), 1117 (2008) 35. J. Hong, S. Lee, C. Cardinaud, G. Turban, J. Non-Cryst. Solids 265(1–2), 125 (2000) 36. Y. Lifshitz, G.D. Lempert, E. Grossman, H.J. Scheibe, S. Voellmar, B. Schultrich, A. Breskin, R. Chechik, E. Shefer, D. Bacon, R. Kalish, A. Hoffman, Diam. Relat. Mater. 6(5–7), 687 (1997) 37. M. Bereznai, J. Budai, I. Hanyecz, J. Kopniczky, M. Veres, M. Koós, Z. Toth, Thin Solid Films 519(9), 2989 (2011) 38. C.H. Bamford, S.L. Cooper, T. Tsuruta (eds.), The Vroman Effect (VSP, Utrecht, 1992), pp. 1–191 39. J. Vacík, B. Dvoˇránková, J. Michálek, M. Pˇrádný, E. Krumbholcová, T. Fenclová, K. Smetana Jr., J. Mater. Sci. Mater. Med. 19(2), 883 (2008)
