Investigation of the Growth Mechanisms of a-CHₓ Coatings Deposited by Pulsed Reactive Magnetron Sputtering

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ This document is the Accepted Manuscript version of a Published Work that appeared in final form in The Journal of Physical Chemistry C (2012) vol.116 n.22, copyright © American Chemical Society after peer review and technical editing by the publisher. To access the final edited and published work see https://doi.org/10.1021/jp300697s This is an Accepted Manuscript of an article published by IOP Publishing in: Nanotechnology (2016), vol. 27, no. 39 available at: https://doi.org10.1088/0957-4484/27/39/395702 ©IOP. En idUS Licencia Creative Commons CC BY NC ND 1 INVESTIGATION OF THE GROWTH MECHANISMS OF a-CHX COATINGS DEPOSITED BY PULSED REACTIVE MAGNETRON SPUTTERING C. Lopez-Santos1*, J.L. Colaux1, J.C. Gonzalez2, S. Lucas1 1) Namur Research Institute for Life Sciences (NARILIS), Research Centre for the Physics of Matter and Radiation (PMR-LARN) University of Namur- FUNDP Rue de Bruxelles, 61, B-5000 Namur, Belgium 2) Institute of Materials Science of Seville (CSIC – University of Seville). Américo Vespucio 49, E-41092 Seville, Spain * corresponding author: Tel: 32 (0)81 72 54 79 Email: [email protected] (C Lopez- Santos) Abstract The study of the growth mechanisms of amorphous hydrogenated carbon coatings (a- CHx) deposited by reactive pulsed magnetron discharge in Ar+C2H2, Ar+H2 and Ar+C2H2+H2 low pressure atmospheres is presented in this work. Hydrogen-containing species of the reactant gas affect the microstructure and surface properties of the a-CHx thin films. The dynamic scaling theory has been used to relate the main reactive species involved in the deposition process to the growth mechanisms of the thin film by means of the analysis of the roughness evolution. Anomalous scaling effects have been observed in smooth a-CHx coatings. Dynamic scaling exponents ,  and z indicate a general growth controlled by surface diffusion mechanisms. Hydrogen species have an influence on the lateral growth of the a-CHx coatings and are involved in the development of a polymericlike structure. Meanwhile, hydrocarbon species promote the generation of higher 2 aggregates, which increases the roughness of a more sp2 clustering structure of the a-CHx coating. Keywords: a-CHx coating, smoothening, dynamic scaling, growth mechanism, magnetron sputtering 1. Introduction Amorphous hydrogenated carbon coatings (a-CHx) have a notable impact in a wide range of industrial applications due to the combination of special properties such as the hardness, high wear resistance, chemical inertness and low friction coefficient, among others 1. The special structure of the a-CHx coatings is composed of carbon, bonded in sp2 and sp3 electronic configurations, and hydrogen. This microstructure determines their versatility 2 to be used like protective coating for optical and magnetic storage devices 3, mechanical components 4 or biomedical instruments 5. The development of a suitable deposition method to produce a-CHx coatings directed at a specific application has been one of the main current goals of investigation. The well-established systems of a-CHx deposition include plasma-enhanced chemical vapour deposition 6, ion beam deposition 7, pulsed laser ablation 8 or magnetron sputtering 9. Deposition of a-C:Hx coatings by magnetron discharge has been studied because such technology is suitable for many industrial applications 9-11. An improvement in the process can be achieved through a pulsed configuration and the use of graphite targets with a hydrocarbon gas precursor 9,12. However, the effect of the predominant growth processes on the obtained microstructure of the a-CHx material in a basic pulsed magnetron sputtering deposition has not been completely understood. Experimental conditions have a strong influence on the reaction 3 mechanisms during the growth of the a-CHx material and subsequently on its microstructure. For example, smooth a-CHx films have been obtained under low pressure work conditions, whereas the development of rougher microstructures was possible by increasing the pressure 13. Other experimental parameters can provide different ratios of sp2 and sp3 carbon bonds and hydrogen contents 14-18. Control of the a-CHx growth mechanisms at low plasma pressure becomes a key goal of study depending on the final application. In order to understand the physical-chemical reactions taking place during the growth of the a-CHx material, several models have been proposed 1,19-21. However, the growth of a-CHx films is a complicated non-equilibrium set of processes. Most of them happen on the surface and in the subsurface layers during simultaneous deposition of carbon species and ion bombardment. This general hypothesis is based on the preferential displacement of sp2 bonded atoms, the densification by recoil implantation and relaxation and the subplantation of atoms and stress relaxation. Therefore, the influence of each mechanism can be affected by the experimental parameters of deposition, such as the ion flux or the substrate temperature. Thus, growth studies based on optical spectroscopy 1,22, spectroscopy ellipsometry 20, mass spectrometry 23,24 and energetic considerations about the impinging species 25 can be found. Other proposed models come from the analysis of the a-CHx coating properties. The development of a surface morphology can be affected by the competition between surface roughening and smoothening mechanisms during film growth, which can be explained through dynamic surface growth models 26. The dynamic scaling hypothesis assumes that the growth of a self-affine surface does not scale the same way in the vertical direction as in the lateral direction. A self-affine surface seems to have both 4 short-range space scaling and long-range time scaling. By measuring the roughness evolution through the scaling coefficients ( roughness and  growth exponents), it is possible to deduce the main growth mechanisms, such as a “ballistic deposition” (direct chemisorption of a growth precursor upon impact) or a “surface diffusion mediated deposition” (chemisorption, which is mediated by a preceding surface diffusion step) 21. The dynamic scaling hypothesis has been recently applied to the formation of smooth a- C:Hx films 20,21,27-30. For example, Zhu et al. 27 have proposed a set of processes responsible for the formation of amorphous diamond-like carbon under ion beam growth conditions: a dense diamond-like phase generation takes place during the slowing-down process of ion collision simultaneously with some relaxation mechanisms, such as the asymmetric displacement of sp2/sp3 carbon atoms, thermally activated density, and stress relaxation processes. In the case of a-CHx coatings grown by sputtering techniques, there are only a few suggested growth models. Logothetidis et al. 20 have described the kinetics of carbon sputter deposition as a function of the effect of the ion energy bombardment. Whereas Pei et al. 29 have studied the growth of a similar material like TiC/a-C nanocomposites coatings deposited by pulsed DC magnetron sputtering with a biased substrate configuration. The common conclusion relates the growth of smooth a-CHx coatings to the existence of an ion-induced downhill current of material, providing a local growth rate dependence on the local surface curvature 28,30-32. However, when the growth of a-CHx coatings occurs without the presence of ion bombarding, other processes must be involved. This is the case with smooth a-CHx thin films deposited by magnetron sputtering without ion beam assistance or a biased substrate. For this purpose, an exhaustive study of the growth of a-CHx coatings through the application of the dynamic 5 scaling theory is presented in this work. From the values of the experimental scaling exponents, theoretical and numerical simulations of nonequilibrium process deposition have been proposed. The Edwards-Wilkinson model is based on the random deposition with surface relaxation (obtaining α=0 and β=0) 33, while the Kardar–Parisi–Zhang model 34 includes the relaxation of the interface by a surface tension, and the random fluctuation in the incoming flux (with α=0.5 and β=0.33). Drotar et al. 35 established a shadowing model through Monte Carlo simulations (resulting in a roughening growth exponent β=1) and the Das Sarma and Tamborenea 36 model relates the relaxation of the deposited atoms to the ratio between the deposition rate and the hopping rate. Taking them into account, the aim of the present work pursues to contribute to the interpretation of the physicalchemical mechanisms affecting the final microstructure of the a-CHx coatings, deposited by magnetron sputtering, according to the more influential hydrocarbon species. The identification of these reactive radicals, as well as the surface characterization of the a- CHx thin films by means of the analysis of the chemical composition, microstructure and topography is used to support the present study. 2. Experimental 2.1 Deposition system Deposition reactor was a vacuum cylinder chamber of stainless steel, assisted by a unipolar pulsed magnetron sputtering source with two confocal 2” graphite targets (99.9% purity). Graphite targets are oriented at 45° with respect to the surface sample normal direction. The residual vacuum was below 10-4 Pa. Several mixtures of gasses were used: 20 sccm of argon (99.9% purity) and 4 sccm of acetylene (99.9% purity), 6 (Ar+C2H2 mixture), 20 sccm of argon and 4 sccm of hydrogen (99.9% purity), (Ar+H2 mixture) and 20 sccm of argon with acetylene and hydrogen, both 4 sccm (Ar+C2H2+H2 mixture). A differentially pumped RC RGA Analyzer HAL7 quadrupole mass spectrometer (Hiden PSM) was employed to characterize the plasma radicals and sputtered species impinging on the substrate during the magnetron sputtering deposition. The study was carried out in residual gas analysis mode, where the ions are formed through electron impact ionisation with a hot filament and focused to the quadrupole mass filter. They are classified by their mass / charge ratio (m/z). Pressure during the deposition was maintained between 2.9 - 4.4 x 10-1 Pa, depending on the flux of reactive gasses. Both targets were connected in parallel on the same pulsed power supply, which delivered a mean power of 250 W with a pulse time-on and a duty cycle set to 50 s and 50 %, respectively. The used substrates were glass, well-cleaned by two ultrasound baths in water and ethanol. 2.2 a-CHx coatings characterization Composition depth profiles were obtained by ion beam analysis (IBA) through the combination of Elastic Recoil Detection Analysis (ERDA) and Rutherford Backscattering Spectroscopy (RBS). An impinging beam of 4He+ at 2.4 MeV was used and the sample was tilted at an angle of 70° between the normal surface direction and the incident beam direction. The detectors were placed at 135° (RBS) and 30° (ERDA). The composition depth profiles of the a-CHx films were obtained by fitting the RBS and ERDA spectra using SIMTarget 37 and SIMNRA 38 codes. 7 Raman spectroscopy was performed by a RH800 Horiba Jovin-Yvon -Raman spectrometer with the IK Series He-Cd laser (=325nm). A grating with 1800g.mm-1, a CCD detector cooling by air and an Olympus BX-41 metallographic microscope attached to the spectrometer were employed. Raman spectra were collected in backscattering configuration, using a silicon wafer for the calibration. Thickness of the a-CHx layer was estimated with a stylus profilometer from Veeco Instruments (Surface Profile Measuring System Dektak). By knowing the thickness and the atomic density provided by nuclear techniques, the density was estimated. Surface topography and roughness were analyzed by a Multimode 8 Atomic Force Microscopy AFM (Veeco diNanoScope V). AFM images were acquired with Si tips of 350-380 kHz of resonant frequency and 8-20 N/m of spring constant working in tapping mode. AFM analysis was performed with the WSxM free available software from Nanotec 39. 2.3 Dynamic scaling theory Growth processes for the evolution of amorphous hydrogenated carbon films can be studied through the dynamic scaling theory. According to this model, for a self-affine surface, the roughness R(L,t) can be scaled as a function of both the scale of measurement (L) and deposition time (t), following the Family-Vicsec relation 40,41:   / t t L R(L, t) L f t / L ~ L t L                 (1) Here  and  parameters are the roughness and growth exponents. Scaling exponents can be estimated from the dimensional analysis by AFM measurements, using the evolution 8 of the roughness over time or through the height-height correlation function HCF(L) 41. HCF(L) is an averaged distribution of heights depending on the lateral distance of the scanned surface. From this function it is possible to deduce the lateral correlation length , which represents the typical lateral length below which surface heights are statistically correlated and is described as ~t1/z. This new exponent is called the coarsening exponent and usually obeys the equation z = / 42. Also, it is useful to obtain the power spectral density PSD(k) by means of the Fourier transform of the topography. These functions present the following asymptotic tendencies 41,42: 2 (2 d) L L HCF(L) ~ const L const k PSD(k) ~ k k                 (2) where d is the dimension of estimation for the PSD(k). The presence of a maximum () on the PSD(k) indicates a wavelength selection on the surface, corresponding to the typical distance between mounds 40,41. 3. Results and discussion 3.1 Influence of the different gas species and sputtered particles on the a-CHx coating’s chemical composition and structure Mass spectrometry analysis provides the identification of the main reactive species at the different experimental deposition conditions. Atomic species such as Ar and H, molecules such as H2, C2 or C2H2 and hydrocarbon radicals type CH, C2H and CH2 were the main species detected during the a-CHx thin films formation. The presence of reactive argon species facilitates the ignition of the graphite sputtering 23,43, improving the 15 are abundant in the gas phase mixture. Samples deposited with the Ar+H2 precursor showed a higher growth exponent of 0.21. This difference between depositions with hydrogen and/or acetylene mixtures may indicate the ability of the main species inducing smoothening/roughening processes to modify the surface morphology, as will be discussed in the last section. The  roughness exponent can be obtained from the height-height correlation function HCF(L) or the power spectral density PSD(k), as shown in Fig. 6. Slopes at low k values were around -4, being slightly higher for the sample deposited from the Ar+H2 mixture. This result agrees with a prevalent surface diffusion process likely responsible for the observed smooth topography 28. A transition wave number kc=1/Lc of the PSD(k) at the power law zone can be related to the typical distance between mounds originated on the a-CHx surface. At a low deposition time, this distance corresponds to the averaged film grain size in a compact surface (~60 nm), while it is related to large aggregate structures with a “cauliflower-like” topography (~300 nm) for thicker samples. From the same growth stage in Fig. 6a, it could be deduced that the introduction of hydrogen to the mixture with acetylene tended to increase this typical distance from 125 nm to 200 nm. Meanwhile, the sample deposited from the Ar+H2 mixture presented an intermediate distance between mounds around 170 nm. Its corresponding PSD(k) is below those of the a-CHx coatings grown with acetylene precursors, which agrees with a lower surface roughness. So, it could be presumed that the main hydrocarbon species provide an increase of the roughness. The joint effect of hydrogen and hydrocarbon species generated a higher roughness and a higher distance between mounds on the a-CHx surface. 16 From the HCF(L) function shape, the increasing tendency at small L values is related to the kinetic roughening, and the saturation at large L values suggests the absence of lateral correlation in the surface roughness. The height-height correlation function HCF(L) corresponding to the a-CHx coating deposited with the Ar+H2 mixture also appears below the data of the samples grown from the acetylene precursor at large length scale. HCF(L) generally shifts upwards as the deposition thickness is increased, as it is possible to observe from the position of the maximum in Fig. 6b. Lateral correlation distance reached the value of 70 nm for the a-CHx deposited with the acetylene precursor, and was higher in the cases where hydrogen was present: 90 nm for the sample grown from the Ar+H2 mixture and 140 nm in the coating grown from the Ar+C2H2+H2 mixture. These results confirm a different local surface growth. The evolution of  during the different deposition processes is presented in Fig. 7. It should be considered a correction factor due to the size of the AFM tip and its convolution with the surface topography 56. The first information comes from the different exponent values deduced from both HCF(L) and PSD(k), as evidence of the existence of local and global exponents, respectively. This is due to different growth behaviors at long and short length scales, known as anomalous scaling 42. A lower global roughness exponent was found for the a-CHx layer deposited with the Ar+H2+C2H2 mixture, whereas the deposition from the Ar+H2 precursor involved a higher global  value. This tendency was inverted in the case of the local roughness exponent. Thus, main hydrogen species might increase the global roughness of the surface, while the hydrocarbon radicals seemed to affect the local microstructure development of the grown a-CHx coating. The variation of  with the deposition time can also be considered as a 17 characteristic of the anomalous dynamic scaling, as shown Fig. 7. Larger variations of  with time deposition were observed in the a-CHx coating deposited with the Ar+H2 mixture. According to Yanguas et al. 42, the change of  value with time indicates that the surface growth mechanisms do not preserve the fractal properties of the surface. Anomalous behavior is usually related to the shadowing processes constant in deposition geometries where species come from off-normal directions, as in the present magnetron sputtering configuration. From these assumptions, the deposition of a-CHx coatings from the mixture of Ar+H2 should be more affected by shadowing effects than those thin films deposited with an acetylene precursor. Anomalous dynamic scaling introduces a modification of the surface roughness evolution, dependent on the scale measurement (l) and expressed as R(l,t)~t if t<<lz and R(l,t)~t l’ if t>>lz 41,57. Now, ’ coefficient is the local roughness exponent related to the scaling behaviour of a self-affine surface. It is noteworthy that, in these cases, the use of the α exponent deduced from the PSD of a mounded surface, not self-affine, is inappropriate 41. Time dependence of the roughness is described by =-/z. It is known from the literature that low roughness of the a-CHx surface is associated with low  and  values 28,30, due to a smoothening phenomena according to the Edwards and Wilkinson (EW) model 33. The EW model is based on a random deposition with surface relaxation, making reference to evaporation/condensation processes, since those regions with local positive surface curvatures (depressions) grow faster than those with local negative surface curvatures (protrusions). In this case, roughness must be scaled as ~(logt)0.5 for short times, and ~(logl)0.5 for long times. For example, the linear representation shown in Fig. 5b confirms the tendency towards the EW model for the 18 deposition time before saturation, as it was previously proposed by Buijnsters et al 28. Rougher surfaces have been obtained with acetylene in the mixture of reactive gasses, due to the random arrival of relatively low energetic radicals, preferentially at the surface protrusions rather than at the valleys. This occurrence is typical of a shadowing effect, which gives preference to growth in the vertical direction. However, since  and  exponents have not been close to null like the EW model predicted, other deposition mechanisms take place simultaneously. Low  values obtained for the a-CHx coatings deposited from the Ar+C2H2 and Ar+H2 mixtures are close to those corresponding to the nonlinear Kardar–Parisi–Zhang (KPZ) model 41. According to this approximation, the main growth processes are stabilization mechanisms such as surface diffusion, lateral growth, or surface tension, providing a conformal growth. A small  value may also indicate the prevalence of surface diffusion, as was previously observed in Si3N4 thin films grown by sputtering techniques 31, due to the combination of a very low working pressure with a low deposition rate. Surface diffusion favors a smooth growing surface instead of a shadowing effect. Chu et al. 58 have indicated that shadowing, stress and step-edge barrier effects can be disregarded during TiN film growth by sputtering deposition at both low temperature and pressure with a small  value (~0.25). Thus, they proposed as a roughening mechanism the shot noise effect, whereas the smoothening mechanism was the surface diffusion. However, a roughness exponent close to 1 is expected for a predominant surface diffusion process, which does not agree with the results of these a-CHx coatings. Global roughness exponent values, around 1, agree with the surface diffusion model that could be applied to large length scale growth of the a-CHx coatings 41. But only the sample deposited from the 19 mixture with acetylene and hydrogen presents a high  value, where it would be clear an important contribution of surface diffusion, but not enough. As an alternative, the evaporation-recondensation process involves a very low  value, although this process is usually weak at room temperature depositions. The combination of both smoothening mechanisms corresponds to a roughness exponent lower than 1 (~2/3), according to the Mullins diffusion model 41. This approximation includes a non-linear surface diffusion process, due to its lateral growth, together with a weak evaporation-recondensation at earlier growth times, in order to describe a growth in low-lying, large curvature areas of the surface. At larger growth times, Mullins approximations tends to follow the EW model 58, which seems to agree with the case of the a-CHx coating grown from the mixtures with acetylene. The Mullins diffusion model predicts a low coarsening exponent of ¼. Nevertheless, the dynamic exponent does not present a universal behaviour since it depends on the experimental deposition conditions, through an influence on the diffusion strength or the sticking coefficient of the species 41. The coarsening exponent can be estimated from the evolution of the lateral correlation length with time in a potential relation type t1/z 41,59. Fig. 7 presents different values of the coarsening exponent for the smooth a-CHx coatings. The use of the Ar+C2H2 mixture provides a higher z-1 value of 0.39, indicating the mound coarsening on the surface 26. Meanwhile, samples deposited from the Ar+H2 precursor showed similar values to those previously observed in smooth DLC coatings (~0.21) 27. In the case of the combination of acetylene and hydrogen for the deposition of a-CHx thin films, the coarsening exponent was found to be significantly lower. It is possible to deduce that the hydrogen flux 20 presented in the reactive mixture reduced the coarsening of the a-CHx microstructure, i.e. the evolution of the lateral correlation length was delayed. 3.4 Relationship between the main reactive species and the a-CHx coating growth mechanisms As it is proposed in Fig. 8, to summarize the main effects of reactive species during the growth of the a-CHx coatings, the prevalence of a Mullins diffusion mechanism suggests that the fragments of a precursor reach the surface with certain mobility. Similar scaling exponents at larger deposition times between the a-CHx coatings grown from the two mixtures with acetylene suggests that preferential hydrocarbon radicals are arriving at the surface, despite the high additional hydrogen presence, as is represented in Fig. 8b and 8c. It is possible to assume that the concentration of active species have an influence on the growth rate, while the type of radicals determines the surface microstructure 42. Fig 8a indicates that hydrogen species may affect the initial growth of independent grains, while the development of higher surface aggregates, observed in the other cases, seems to depend on the hydrocarbon radicals, as the scaling exponents have suggested. Therefore, Fig. 8b and 8c present how radicals with a low sticking coefficient, such as C2H (10-2~10- 4), are involved in a surface diffusion mechanism 21. Meanwhile, species with higher sticking coefficients close to 1, such as CH molecules (in Fig. 8b) or C atoms (in Fig. 8a), cause a fast chemisorption mechanism and, subsequently, less surface smoothening. Posterior material coalescence is supported by the common off-normal arrival of particles at the experimental deposition conditions. 21 Because a shadowing effect has been associated with a high sticking coefficient, while a lower sticking coefficient tends to favour reemission processes, the hydrogen mixture might present a shadowing effect on the a-CHx grown surface. However, a shadowing prevalence would increase the roughness, which has not been observed in that sample. Instead of that, hydrogen species can affect the surface morphology through an etching process, suggested in Fig. 8a. In that case, the influence of the sticking coefficient on the film growth is the opposite: a higher sticking coefficient is responsible for a strong etching and, subsequently a surface smoothening, mainly on the tops of the surface mounds 41. Species with lower sticking coefficients may produce an etching mechanism assisted by reemission processes. This allows easily reach the valleys of the surface improving the roughening, as it is included in Fig. 8b for the a-CHx deposition with the acetylene precursor. This effect is in agreement with the different roughness obtained for the a-CHx coatings deposited from the mixture with hydrogen or acetylene. In any case, it should be noted that the etching influence is generally low at room temperature. For the sample deposited from the Ar+C2H2+H2 mixture, growth mechanisms are a combination of the main processes of both depositions with hydrogen and acetylene, represented in Fig. 8c. There must be a competition between shadowing and reemission processes to increase the roughness. As is well known, an incident atom coming from a sputtering deposition has a high kinetic energy when reaching the surface and a sticking coefficient different than 1. After the surface contact, the corresponding energy loss facilitates the reemission of the atom increasing the sticking probability. This results in a global increase in the surface roughness. Hydrogen atoms might also have an influence on the effective sticking coefficient of the hydrocarbon radicals due to the formation of dangling 22 bonds as preferred chemisorption sites 21, accompanied by the improvement of the roughness. 4. Conclusion Deposition of a-CHx coatings by reactive pulsed magnetron sputtering at low pressure has resulted in a high hydrogen incorporation, uniform in-depth chemical composition and smooth surface. From the dynamic scaling approximation, a-CHx samples presented the ,  and z-1 values of 0.29, 0.13 and 0.39 from the deposition with the Ar+C2H2 mixture; 0.60, 0.13 and 0.05 for the sample corresponding to the Ar+C2H2+H2 mixture; and 0.18, 0.21 and 0.15 with the Ar+H2 precursor. Growth of the a-CHx coatings can be described by the Mullins diffusion model. Hydrogen species seemed to improve the smoothening through a lateral growth in a conformal way supported by etching processes. This favored the generation of a polymeric-like microstructure, accompanied by a small increase in the bulk density. Meanwhile, the main hydrocarbon radicals promoted the roughening by means of surface diffusion in large curvature regions and the contribution of reemission mechanisms. These hydrocarbon species improved the deposition rate and provided the formation of transpolyacetylene aggregations included in a more sp2-clustering structure of the a-CHx coating. Acknowledgments This work was partially supported by the Belgian Walloon Region, DG06 project TRIBOFUTUR, and by the European Commission Programme FP7 REGPOT-CT-2011- 23 285895, AL-NANOFUNC. We thank Pf. D. Bonifazi from the Department of Chemistry, University of Namur (FUNDP) for the access to the AFM equipment. 24 REFERENCES (1) Robertson, J. Materials Science & Engineering R-Reports 2002, 37, 129. (2) Rybachuk, M.; Bell, J. M. Carbon 2009, 47, 2481. (3) Piazza, F.; Grambole, D.; Zhou, L.; Talke, F.; Casiraghi, C.; Ferrari, A. C.; Robertson, J. Diamond and Related Materials 2004, 13, 1505. (4) Hauert, R. Diamond and Related Materials 2003, 12, 583. (5) Roy, R. K.; Lee, K. R. Journal of Biomedical Materials Research Part B- Applied Biomaterials 2007, 83B, 72. (6) Casiraghi, C.; Piazza, F.; Ferrari, A. C.; Grambole, D.; Robertson, J. Diamond and Related Materials 2005, 14, 1098. (7) Ibenskas, A.; Galdikas, A.; Meskinis, S.; Andrulevicius, M.; Tamulevicius, S. Diamond and Related Materials 2011, 20, 693. (8) Eskusson, J.; Jaaniso, R.; Lust, E. Applied Surface Science 2009, 255, 7104. (9) Czyzniewski, A. Surface & Coatings Technology 2009, 203, 1027. (10) Hang, R. Q.; Ma, S. L.; Chu, P. K. Diamond and Related Materials 2010, 19, 1230. (11) Lousinian, S.; Logothetidis, S.; Laskarakis, A.; Gioti, M. Biomolecular Engineering 2007, 24, 107. (12) Keunecke, M.; Weigel, K.; Bewilogua, K.; Cremer, R.; Fuss, H. G. Thin Solid Films 2009, 518, 1465. (13) Miladinovic, S. M.; De Vriendt, V.; Robotham, S. A.; Maseri, F.; Lucas, S.; Wilkins, C. L. Journal of the American Society for Mass Spectrometry 2010, 21, 411. (14) Chowdhury, S.; Laugier, M. T.; Rahman, I. Z. Journal of Materials Processing Technology 2004, 153, 804. (15) Caschera, D.; Cossari, P.; Federici, F.; Kaciulis, S.; Mezzi, A.; Padeletti, G.; Trucchi, D. M. Thin Solid Films 2011, 519, 4087. (16) Hosseini, S. I.; Shokri, B.; Firouzjah, M. A.; Kooshki, S.; Sharifian, M. Thin Solid Films 2011, 519, 3090. (17) Sonderby, S.; Berthelsen, A. N.; Almtoft, K. P.; Christensen, B. H.; Nielsen, L. P.; Bottiger, J. Diamond and Related Materials 2011, 20, 682. (18) Wang, Y. X.; Ye, Y. P.; Li, H. X.; Ji, L.; Chen, J. M.; Zhou, H. D. Applied Surface Science 2011, 257, 1990. (19) Maheswaran, R.; Ramaswamy, S.; Thiruvadigal, D. J.; Gopalakrishnan, C. Journal of Non-Crystalline Solids 2011, 357, 1710. (20) Logothetidis, S.; Patsalas, P.; Gioti, M.; Galdikas, A.; Pranevicius, L. Thin Solid Films 2000, 376, 56. (21) Kim, I. Y.; Hong, S. H.; Consoli, A.; Benedikt, J.; von Keudell, A. Journal of Applied Physics 2006, 100. (22) Acsente, T.; Ionita, E. R.; Stancu, C.; Ionita, M. D.; Dinescu, G.; Grisolia, C. Surface & Coatings Technology 2011, 205, S402. (23) Baby, A.; Mahony, C. M. O.; Maguire, P. D. Plasma Sources Science & Technology 2011, 20.