- Email: [email protected]
Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy
G Model
ARTICLE IN PRESS
APSUSC-35650; No. of Pages 5
Applied Surface Science xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Full Length Article
Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy Tomáˇs Kocourek a,b,∗ , Miroslav Jelínek a,b , Petr Písaˇrík a,b , Jan Remsa a,b , Michaela Janovská c , Michal Landa c , Josef Zemek a , Vladimír Havránek d a
Institute of Physics AS CR, Na Slovance 2, 182 21, Praha 8, Czech Republic Czech Technical University, Faculty of Biomedical Engineering, Nám. Sítná 3105, 27201 Kladno, Czech Republic c Institute of Thermomechanics AS CR, Dolejˇskova 1402/5, 182 00, Praha 8, Czech Republic d ˇ z 130, 250 68 Reˇ ˇ z, Czech Republic Nuclear Physics Institute AS CR, Reˇ b
a r t i c l e
i n f o
Article history: Received 27 October 2016 Received in revised form 20 March 2017 Accepted 30 March 2017 Available online xxx Keywords: In-situ ion bombardment Pulsed laser deposition Diamond-like carbon Hybrid technology Film modification
a b s t r a c t Biocompatible Diamond-Like Carbon (DLC) films were prepared by Pulsed Laser Deposition technique using the laser energy density of 10 J cm−2 on the graphite target. The surface of the grown film was modified during the deposition by bombardment with argon, xenon, nitrogen or oxygen ions. The ion energy (up to 150 eV) was changed by gun voltage and by ionic current. The films with high and low diamond/graphite content were prepared. Physical and mechanical properties of biocompatible DLC thin layers prepared by hybrid laser technology were studied. The composition of layers and the content trace elements were determined by the methods of Rutherford Backscattering Spectrometry and Particle Induced X-ray Emission. The content of sp2 and sp3 bonds was measured using X-ray Photoelectron Spectroscopy. For different energy of argon and oxygen ions the maximum of sp3 bonds content was found (83.63% of sp3 bonds for argon ions). All films were smooth, which was confirmed by profilometry and Atomic Force Microscopy measurements. Maximum roughness Ra and RMS was did not exceed 1 nm. The Young´ıs and shear moduli were studied by Resonant Ultrasound Spectroscopy. The Young’s Modulus attained the value of 601 GPa and the shear Modulus attained the value of 253 GPa at the energy of 30 eV of Ar ions. The influence of ion bombardment on DLC film properties is discussed. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Diamond-like carbon (DLC) is one of the most interesting films with high hardness, optical transparency in the visible and infrared regions, electrical and thermal conductivities, wear resistance, low friction coefficient, and excellent biocompatibility [1,2]. The main aim for utilization of ion bombardment during deposition process was the sp3 bonds content increase and the improvement of elastic properties of the layers. As a result, we have expected general improvement in practical use of DLC coating, namely for medical implants. The goal of this work was achieving optimal deposition conditions of DLC layers modified by ion bombardment, and the study of elastic properties by newly developed method Resonant Ultrasound Spectroscopy (RUS), which has potential to be widely used because of its non-destructiveness [3].
∗ Corresponding author at: Institute of Physics AS CR, Na Slovance 2, Praha 8, 182 21, Czech Republic. E-mail address: [email protected] (T. Kocourek).
Ion beam assisted processing can be used for direct deposition films, such as Ion Beam Assisted Deposition technology (IBAD) or for modification of the fabricated or grown films [4,5]. In the case of IBAD, usually ions energy higher than 1 keV (typically up to 2 keV) and gases or precursors such as methane, ethylene/ethane, acetylene, CH4 /H2 , HMDSO, fluorinated carbon, H2 S, O2 and N2 are used. For modification, especially for a-C and a-C:H DLC films, the Kaufmann source or Van der Graaf accelerator using gases as Ar, Xe, N2 and Kr were applied [6–9]. For surface modification (or graphene synthesis) also bias voltage was used [10,11]. Ion bombardment of DLC films strongly affects the film properties, such as crystallinity, density, sp3 content and surface morphology. By bombardment the mobility of adatoms is increased and weak carbon–carbon bonds are removed. The processes that are under way in ion bombarded growing film are not yet fully understood. Publications indicates the importance of other parameters than ion energy, such as the ion mass and possible chemical activity of the ion. Even though ions with higher energy deposit more energy into growing layer, reported results show that there are optimal energy values (tenths
http://dx.doi.org/10.1016/j.apsusc.2017.03.274 0169-4332/© 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: T. Kocourek, et al., Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.03.274
G Model APSUSC-35650; No. of Pages 5
ARTICLE IN PRESS T. Kocourek et al. / Applied Surface Science xxx (2017) xxx–xxx
2
Fig. 1. Scheme of the PLD deposition system with ion bombardment and a photo of the chamber.
of eV) that are most efficient in influencing the growth. The same is valid for ion mass [12–15]. In our study we combine PLD technique with in situ ion bombardment of growing DLC films to produce films of higher sp3 content. We use KrF excimer laser, Kaufman ion source and gases as argon, xenon, nitrogen and oxygen. For analysis (composition, bonds, and morphology) of thin films following methods were used: Rutherford Backscattering Spectrometry (RBS), Particle Induced X-ray Emission (PIXE), X-ray Photoelectron Spectroscopy (XPS), profilometric and Atomic Force Microscopy (AFM). And for determination of elastic properties (Young’s and shear modulus) we used the method RUS.
2. Experimental The DLC layers were prepared using PLD with simultaneous bombardment of growing film with Ar, Xe, O2 or N2 ions of various energies (see Fig. 1.). The laser (KrF excimer; = 248 nm; pulse duration = 20 ns) energy density was 10 J cm−2 (spot size 2 × 1 mm) and the repetition rate was 10 Hz. The number of pulses was adjusted to reach approximately the same layer thickness (1500 nm for layers on Ti6Al4 V cylindrical substrate (for RUS measurement) and 100 nm on Si (111) substrate). The graphite target was rotated (0.5 Hz) and the target-substrate distance was 45 mm. DLC films were created with substrate at room temperature. Base vacuum of the coating system was 1 × 10−4 Pa. Films were deposited in Ar ambient of 0.25 Pa. Before the deposition process the substrates were cleaned by RF discharge (13.56 MHz) in 5 Pa of argon for two minutes. The layers were created with or without ion bombardment. The ion gun eH200 (Kaufman and Robinson, Inc.) was used for the bombardment. The operating parameters of the ion gun were held at a working pressure of 0.25 Pa (for gases: Ar, Xe, O2 , N2 ) and at a cathode current of 0.15 A. The ion energy was varied from 30 eV to 150 eV for Ar and Xe, and from 50 eV to 150 eV for O2 and N2 . A standard measurement set for determination of DLC layers physical properties was conducted, i.e.: thickness, roughness, elemental composition, bonds. Then on selected layers the elastic properties were evaluated to determine the optimal fabrication conditions. Thickness and surface roughness were measured by mechanical profilometer Tencor AlphaStep 500. The layers topography was characterized by the atomic force microscopy type Solver NEXT (NT-MDT) in a dynamic regime with HA NC tips (tip radius ∼ 6 nm). The roughness average (Ra ), adapted from ISO 4287/1, was calculated from 100 m2 area with software NOVA P9.
X-ray photoelectron spectra were measured by an ADES-400 photoelectron spectrometer (VG Scientific, UK) using Mg K␣ radiation (1253.6 eV). The spectra were recorded for wide-survey and narrow scans in C 1 s and O 1 s regions with a pass energy of 100 eV or 20 eV (C 1 s line). Inelastic electron background was subtracted using Shirley’s procedure [2,16]. The composition of layers and the content of trace elements were determined by Particle Induced X-ray Emission, and Rutherford Backscattering Spectrometry methods using 3 MV Tandetron ˇ z with the energy of protons equal to 4130MC accelerator at INP Reˇ 2.94 MeV. Elastic properties of DLC layers were evaluated by Resonant Ultrasound Spectroscopy. This method is based on comparison of the frequency spectra of free vibration of the substrate measured before and after layer deposition, and determines in-plane elastic properties of the layer from frequency shifts of the individual resonant peaks [17]. Plane stress state is assumed in the vibrated layer and takes into account the elastic and mass contributions of the layer as a small perturbation to the elasticity and mass of the substrate. Thus only six independent elastic coefficients Qij describing in-plane elasticity in a general case can be obtained by this method. Assuming isotropic elasticity of the deposited layers, only two parameters (Q11 , Q33 ) were iteratively refined by matching of calculated and measured frequency shifts. Subsequently, they were recalculated [17] to G (shear) and E (Young’s) modulus. Vibration of the samples was excited by a thermoacoustic source with infrared pulsed laser. The resonant frequencies and shapes of individual modes were evaluated from signals recorded by the scanning laser vibrometer implemented in the optical microscope (Polytec Micro System Analyzer MSA 500). For each sample, 20–30 resonance modes were detected in the frequency range 0.3–2 MHz.
3. Results and discussion 3.1. The thickness and roughness The thickness of all DLC layers was in the region from 60 nm to 140 nm for Si (111) substrates and from 1200 nm to 1540 nm for Ti6Al4 V substrates. The lowest growth rate of the layers was for energy of ions from 30 eV to 50 eV for all the gases. Surface roughness Ra and RMS was not higher that 1 nm (measured on the Si substrates by mechanical profilometer). The layer topography was characterized by AFM, too. The layers were smooth with a small number of small droplets. (see Fig. 2.) The average surface roughness Ra was not higher than 0.4 nm (calculated from 10 m × 10 m area).
Please cite this article in press as: T. Kocourek, et al., Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.03.274
G Model
ARTICLE IN PRESS
APSUSC-35650; No. of Pages 5
T. Kocourek et al. / Applied Surface Science xxx (2017) xxx–xxx
3
Fig. 2. The layer topography was characterized by the AFM (the sample prepared with energy of 30 eV of Ar ions, on the Si substrate).
Fig. 3. The composition of elements of bombarded DLC films (determined by RBS on the Si substrates).
3.2. The composition The elemental composition was determined by RBS and by PIXE. Both methods have reach several micrometers in depth. Since the layers were roughly 0.1 m thick, the information is from the entire layer. The representation of other elements in the C layers is not significant, except for the layers bombarded by ions O2 or N2 , which exhibited increase of O or N elements on a significant level – the highest content for the lowest ion energy. The higher concentration of O and N was caused by implantation of these elements into the layer when utilizing low energies. In case of high energy ions (O2 , N2 ) the sputtering effect of the growing layer is stronger so the element easily escape. (as determined by RBS, see Fig. 3.) The content of trace elements was at most up to 0.3 at.% (Ar, Xe, W, Fe, Ni, Cr, etc.) determined by PIXE. Bombarded layers with the lowest content of trace elements (mainly O and N) were obtained with Ar ions.
3.3. The percentage of bonds The percentage of bonds (Csp3 , Csp2 , C O, C O, and C N) was determined by XPS. Number of bonds C O was approximately constant for all prepared layers. The bonds of C O were present in all layers except layers prepared with N2 ions. In layers prepared with N2 ions was found a significant number of C N bonds, which were not detected in the other layers. The Xe ions bombardment
with energy of 30 eV increased the content of Csp3 and Csp2 at the expense of C O and C O bonds, but the content of Csp2 bonds increased. The difference in mechanism between Xe and Ar ions can be contributed to different ion mass so different collision mechanism (see Fig. 4.). 3.4. The ratio of sp3 /sp2 bonds Maximum sp3 (83.63%) was observed for energy of 30 eV for Ar ions, which is more accurate than our past results [2], where the minimum energy was 40 eV (lower energy could not be reached at the time). Information about the effects of ions with energies lower than 30 eV could not be obtained because the device does not allow these modes of operation. Our experiments showed that the Ar ions with energy from 30 to 40 eV impacting the growing layers increase the content of sp3 carbon bonds and decrease the sp2 content. The sp3 /sp2 ratio increased from 2.8 (not bombarded) to 5.1 for the layers bombarded with Ar ions with the energy of 30 eV. The C C bonds have energy 285.2–285.4 eV for sp3 and 284.2–284.5 eV for sp2 . A small increase of the sp3 /sp2 ratio was observed for O2 ions with energy close to 80 eV. For the ion types used here, higher energy did not influence the formation of sp3 bonds. On the other hand, Xe and N2 ion bombardment (with energy 30–50 eV and 50–80 eV, respectively) increased the sp2 content, which led to deterioration of the DLC layer properties. It can be concluded that
Please cite this article in press as: T. Kocourek, et al., Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.03.274
G Model APSUSC-35650; No. of Pages 5
ARTICLE IN PRESS T. Kocourek et al. / Applied Surface Science xxx (2017) xxx–xxx
4
Fig. 4. The survey of bonds of bombarded DLC films (determined by XPS on the Si substrates).
Fig. 5. Graph of the values of sp3 bonds (percentage – left axis) and sp3 /sp2 bonds ratio (right axis) of DLC films depending on the energy of Ar, O2 , Xe, and N2 ions (measured on Si substrates).
(which was our aim), these sample were not selected for the elastic properties measurement (RUS method) and we focused on Ar ions. The influence of ion bombardment on the development of sp3 bonds, E (Young’s modulus) and G (shear modulus) is evident from Fig. 7a, b, c. where the dependences are roughly similar. For the non-bombarded DLC layers the E modulus was determined to be 565 GPa and G modulus 238 GPa. The E and G moduli increased to 601 GPa and 253 GPa, respectively with the increase of sp3 content with ion energy of 30 eV. For ion energy of 40–50 eV, the sp3 content and E and G moduli was lowered. However, the values were lower than the nonbombarded layers even though the sp3 /sp2 ratio was higher or the same. This is implies an influence of ion bombardment on elastic properties which are not connected to sp3 /sp2 ratio (density, porosity, texture, etc.). Modulus E decreased to 529 GPa and modulus G to 224 GPa. For the layers prepared using higher energy ions (80–150 eV) the values of E, G and sp3 /sp2 ratio did not differ significantly from the non-bombarded DLC layers. 4. Conclusions
the ions with lower energy have greater influence on the sp3
or sp2
bonds content (see Fig. 5.). We can conclude that there is an optimal ion mass and energy for the increase of sp3 bonds or the decrease of sp2 bonds, or both processes simultaneously. 3.5. Elastic properties The elastic properties of the substrate were evaluated by the RUS method modified for cylindrical samples (Ti6Al4 V substrates were polished with high geometrical accuracy) [18]. Subsequently, the layer was deposited and the spectrum was re-measured at identical conditions. The corresponding peaks (Fig. 6.) in both spectra were found by means of comparison of shapes of vibration modes, and the frequency shifts of the resonances were determined. The measured resonant frequency shifts were by two orders of magnitude higher than the uncertainty in evaluation of the resonant frequencies (spectral peaks position). Mass density of DLC was determined from close relationship to the measured sp3 contents. Since Xe, O2 , and N2 ion bombardment did not improve the sp3 /sp2 ratio
With our experimental setup, we were able to fabricate DLC layers with high content of sp3 bonds, maximum of 83.63% (determined by XPS), by PLD method with Ar ion bombardment during layer growth for ion energy of 30 eV. Other tested conditions (30–150 eV, Xe, O2 , N2 ) were not deemed perspective. The composition study showed negligible content of oxygen and nitrogen, and other element traces that are present in non-bombarded layers (determined by RBS and PIXE). The layers were smooth with small number of droplets (profilometer and AFM). For the cylindrical samples we showed the possibility of DLC layers measurement by non-destructive method RUS. For the layers fabricated with Ar ion bombardment with ion energy of 30 eV the Young modulus and shear modulus values reached 601 GPa and 253 GPa, respectively. We can conclude that it is possible to increase the sp3 bonds content and improve elastic properties of DLC layer via ion bombardment of low energy argon ions and that the RUS method bring a possibility to measure sample without damaging the surface. Next research in this field can bring a better understanding of processes taking place during ion bombardment of growing layer, as well as the develop-
Please cite this article in press as: T. Kocourek, et al., Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.03.274
G Model APSUSC-35650; No. of Pages 5
ARTICLE IN PRESS T. Kocourek et al. / Applied Surface Science xxx (2017) xxx–xxx
5
Fig. 6. The resonant spectra of free vibrating substrate before and after DLC deposition (measured on the Ti6Al4 V substrates, Ar ions, 30 eV).
References
Fig. 7. a, b, c. The graph of values of sp3 bonds (a), E (Young’s modulus) (b), and G (shear modulus) (c) of DLC films depending on the energy of Ar ions (measured on the Ti6Al4 V substrates).
ment in ultrasonic spectroscopic methods can bring the possibility of elastic properties measurement during manufacturing process, such as implant coatings.
Acknowledgments
[1] J. Robertson, Diamond-like amorphous carbon, Mater. Sci. Eng. R 37 (2002) 129–281. [2] M. Jelinek, P. Písaˇrík, T. Kocourek, J. Zemek, J. Lukeˇs, Influence of ion bombardment on growth and properties of PLD created DLC films, Appl. Phys. A 110 (2013) 943–947. ´ J. Kopeˇcek, ˚ zek, M. Landa, M. Jelínek, J. Mikˇsovsky, [3] T. Kocourek, M. Ruˇ Evaluation of elastic properties of DLC layers using resonant ultrasound spectroscopy and AFM nanoindentation, Surf. Coat. Technol. 205 (2011) 67–70. [4] X.L. Peng, Z.H. Barber, T.W. Clyne, Surface roughness of diamond-like carbon films prepared using various techniques, Surf. Coat. Technol. 138 (2001) 23–32. [5] Q. Jun, L. Jianbin, W. Shizhu, W. Jing, L. Wenzhi, Mechanical and tribological properties of non-hydrogenated DLC films synthesized by IBAD, Surf. Coat. Technol. 128–129 (2000) 324–328. [6] F. Rossi, Amorphisation and growth mechanisms of carbon films under ion beam irradiation, Chaos Solitons Fractals 10 (1999) 2019–2029. [7] A.R. Gonzalez-Elipe, F. Yubero, J.M. Sanz, Low Energy Ion Assisted Film Imperial Growth, College Press, 2003. [8] M. Rybachuk, J.M. Bell, Growth of diamond-like carbon films using low energy ion beam sputter – bombardment deposition with Ar ions, J. Phys.: Conf. Ser. 100 (2008) 082009. [9] I.S. Trakhtenberg, S.A. Plotnikov, V.B. Vykhodets, V.A. Pavlov, G.A. Raspopova, Modification of amorphous diamond-like films by bombardment with low-energy ions, Diamond Relat. Mater. 5 (1996) 943–946. [10] T. Yasui, T. Kotani, K. Fujiuchi, H. Tahara, T. Ypshikawa, Carbon nitride films deposited by reactive sputtering and pulsed laser ablation, Thin Solid Films 457 (2004) 133–138. [11] S.S. Tinchev, Surface modification of diamond-like carbon films to graphene under low energy ion beam irradiation, Appl. Surf. Sci. 258 (2012) 2931–2934. [12] J. Robertson, Deposition mechanisms for promoting sp3 bonding in diamond-like carbon, Diamond Relat. Mater. 2 (1993) 984–989. [13] P.A. Karaseov, O.A. Podsvirov, A.I. Titov, K.V. Karabeshkin, A. Ya Vinogradov, V.S. Belyakov, A.V. Arkhipov, L.M. Nikulina, A.L. Shakhmin, E.N. Shubina, N.N. Karasev, Effect of ion bombardment on the phase composition and mechanical properties of diamond-like carbon films, J. Surf. Invest. X-ray Synchrotron Neutron Tech. 8 (2014) 45–49. [14] C.A. Davis, A simple model for the formation of compressive stress in thin films by ion bombardment, Thin Solid Films 226 (1993) 30–34. [15] O.A. Podsvirov, P.A. Karaseov, A. Ya Vinogradov, A. Yu Azarov, N.N. Karasev, A.S. Smirnov, A.I. Titov, K.V. Karabeshkin, Residual stress in diamond-like carbon films- role of growth conditions and ion irradiation, J. Surf. Invest. X-Ray Synchrotron Neutron Tech. 4 (2010) 241–244. [16] D.A. Shirley, High-resolution X-ray photoemission spectrum of the valence bands of gold, Phys. Rev. B 5 (1972) 4709–4714. ˚ zek, P. Sedlák, H. Seiner, A. Kruisová, M. Landa, Linearized forward and [17] M. Ruˇ inverse problems of the resonant ultrasound spectroscopy for the evaluation of thin surface layers, Acoust. Soc. Am. 128 (2010) 3426–3437. [18] M. Janovská, P. Sedlák, H. Seiner, M. Landa, P. Marton, P. Ondrejkoviˇc, J. Hlinka, Anisotropic elasticity of DyScO3 substrates, J. Phys. Condens. Matter 24 (2012) 385404.
The project has been supported by Czech Grant Agency under GA15-05864S and the Grant Agency of the Czech Technical University in Prague under grant No. SGS16/190/OHK4/2T/17.
Please cite this article in press as: T. Kocourek, et al., Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.03.274
ARTICLE IN PRESS
APSUSC-35650; No. of Pages 5
Applied Surface Science xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Full Length Article
Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy Tomáˇs Kocourek a,b,∗ , Miroslav Jelínek a,b , Petr Písaˇrík a,b , Jan Remsa a,b , Michaela Janovská c , Michal Landa c , Josef Zemek a , Vladimír Havránek d a
Institute of Physics AS CR, Na Slovance 2, 182 21, Praha 8, Czech Republic Czech Technical University, Faculty of Biomedical Engineering, Nám. Sítná 3105, 27201 Kladno, Czech Republic c Institute of Thermomechanics AS CR, Dolejˇskova 1402/5, 182 00, Praha 8, Czech Republic d ˇ z 130, 250 68 Reˇ ˇ z, Czech Republic Nuclear Physics Institute AS CR, Reˇ b
a r t i c l e
i n f o
Article history: Received 27 October 2016 Received in revised form 20 March 2017 Accepted 30 March 2017 Available online xxx Keywords: In-situ ion bombardment Pulsed laser deposition Diamond-like carbon Hybrid technology Film modification
a b s t r a c t Biocompatible Diamond-Like Carbon (DLC) films were prepared by Pulsed Laser Deposition technique using the laser energy density of 10 J cm−2 on the graphite target. The surface of the grown film was modified during the deposition by bombardment with argon, xenon, nitrogen or oxygen ions. The ion energy (up to 150 eV) was changed by gun voltage and by ionic current. The films with high and low diamond/graphite content were prepared. Physical and mechanical properties of biocompatible DLC thin layers prepared by hybrid laser technology were studied. The composition of layers and the content trace elements were determined by the methods of Rutherford Backscattering Spectrometry and Particle Induced X-ray Emission. The content of sp2 and sp3 bonds was measured using X-ray Photoelectron Spectroscopy. For different energy of argon and oxygen ions the maximum of sp3 bonds content was found (83.63% of sp3 bonds for argon ions). All films were smooth, which was confirmed by profilometry and Atomic Force Microscopy measurements. Maximum roughness Ra and RMS was did not exceed 1 nm. The Young´ıs and shear moduli were studied by Resonant Ultrasound Spectroscopy. The Young’s Modulus attained the value of 601 GPa and the shear Modulus attained the value of 253 GPa at the energy of 30 eV of Ar ions. The influence of ion bombardment on DLC film properties is discussed. © 2017 Elsevier B.V. All rights reserved.
1. Introduction Diamond-like carbon (DLC) is one of the most interesting films with high hardness, optical transparency in the visible and infrared regions, electrical and thermal conductivities, wear resistance, low friction coefficient, and excellent biocompatibility [1,2]. The main aim for utilization of ion bombardment during deposition process was the sp3 bonds content increase and the improvement of elastic properties of the layers. As a result, we have expected general improvement in practical use of DLC coating, namely for medical implants. The goal of this work was achieving optimal deposition conditions of DLC layers modified by ion bombardment, and the study of elastic properties by newly developed method Resonant Ultrasound Spectroscopy (RUS), which has potential to be widely used because of its non-destructiveness [3].
∗ Corresponding author at: Institute of Physics AS CR, Na Slovance 2, Praha 8, 182 21, Czech Republic. E-mail address: [email protected] (T. Kocourek).
Ion beam assisted processing can be used for direct deposition films, such as Ion Beam Assisted Deposition technology (IBAD) or for modification of the fabricated or grown films [4,5]. In the case of IBAD, usually ions energy higher than 1 keV (typically up to 2 keV) and gases or precursors such as methane, ethylene/ethane, acetylene, CH4 /H2 , HMDSO, fluorinated carbon, H2 S, O2 and N2 are used. For modification, especially for a-C and a-C:H DLC films, the Kaufmann source or Van der Graaf accelerator using gases as Ar, Xe, N2 and Kr were applied [6–9]. For surface modification (or graphene synthesis) also bias voltage was used [10,11]. Ion bombardment of DLC films strongly affects the film properties, such as crystallinity, density, sp3 content and surface morphology. By bombardment the mobility of adatoms is increased and weak carbon–carbon bonds are removed. The processes that are under way in ion bombarded growing film are not yet fully understood. Publications indicates the importance of other parameters than ion energy, such as the ion mass and possible chemical activity of the ion. Even though ions with higher energy deposit more energy into growing layer, reported results show that there are optimal energy values (tenths
http://dx.doi.org/10.1016/j.apsusc.2017.03.274 0169-4332/© 2017 Elsevier B.V. All rights reserved.
Please cite this article in press as: T. Kocourek, et al., Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.03.274
G Model APSUSC-35650; No. of Pages 5
ARTICLE IN PRESS T. Kocourek et al. / Applied Surface Science xxx (2017) xxx–xxx
2
Fig. 1. Scheme of the PLD deposition system with ion bombardment and a photo of the chamber.
of eV) that are most efficient in influencing the growth. The same is valid for ion mass [12–15]. In our study we combine PLD technique with in situ ion bombardment of growing DLC films to produce films of higher sp3 content. We use KrF excimer laser, Kaufman ion source and gases as argon, xenon, nitrogen and oxygen. For analysis (composition, bonds, and morphology) of thin films following methods were used: Rutherford Backscattering Spectrometry (RBS), Particle Induced X-ray Emission (PIXE), X-ray Photoelectron Spectroscopy (XPS), profilometric and Atomic Force Microscopy (AFM). And for determination of elastic properties (Young’s and shear modulus) we used the method RUS.
2. Experimental The DLC layers were prepared using PLD with simultaneous bombardment of growing film with Ar, Xe, O2 or N2 ions of various energies (see Fig. 1.). The laser (KrF excimer; = 248 nm; pulse duration = 20 ns) energy density was 10 J cm−2 (spot size 2 × 1 mm) and the repetition rate was 10 Hz. The number of pulses was adjusted to reach approximately the same layer thickness (1500 nm for layers on Ti6Al4 V cylindrical substrate (for RUS measurement) and 100 nm on Si (111) substrate). The graphite target was rotated (0.5 Hz) and the target-substrate distance was 45 mm. DLC films were created with substrate at room temperature. Base vacuum of the coating system was 1 × 10−4 Pa. Films were deposited in Ar ambient of 0.25 Pa. Before the deposition process the substrates were cleaned by RF discharge (13.56 MHz) in 5 Pa of argon for two minutes. The layers were created with or without ion bombardment. The ion gun eH200 (Kaufman and Robinson, Inc.) was used for the bombardment. The operating parameters of the ion gun were held at a working pressure of 0.25 Pa (for gases: Ar, Xe, O2 , N2 ) and at a cathode current of 0.15 A. The ion energy was varied from 30 eV to 150 eV for Ar and Xe, and from 50 eV to 150 eV for O2 and N2 . A standard measurement set for determination of DLC layers physical properties was conducted, i.e.: thickness, roughness, elemental composition, bonds. Then on selected layers the elastic properties were evaluated to determine the optimal fabrication conditions. Thickness and surface roughness were measured by mechanical profilometer Tencor AlphaStep 500. The layers topography was characterized by the atomic force microscopy type Solver NEXT (NT-MDT) in a dynamic regime with HA NC tips (tip radius ∼ 6 nm). The roughness average (Ra ), adapted from ISO 4287/1, was calculated from 100 m2 area with software NOVA P9.
X-ray photoelectron spectra were measured by an ADES-400 photoelectron spectrometer (VG Scientific, UK) using Mg K␣ radiation (1253.6 eV). The spectra were recorded for wide-survey and narrow scans in C 1 s and O 1 s regions with a pass energy of 100 eV or 20 eV (C 1 s line). Inelastic electron background was subtracted using Shirley’s procedure [2,16]. The composition of layers and the content of trace elements were determined by Particle Induced X-ray Emission, and Rutherford Backscattering Spectrometry methods using 3 MV Tandetron ˇ z with the energy of protons equal to 4130MC accelerator at INP Reˇ 2.94 MeV. Elastic properties of DLC layers were evaluated by Resonant Ultrasound Spectroscopy. This method is based on comparison of the frequency spectra of free vibration of the substrate measured before and after layer deposition, and determines in-plane elastic properties of the layer from frequency shifts of the individual resonant peaks [17]. Plane stress state is assumed in the vibrated layer and takes into account the elastic and mass contributions of the layer as a small perturbation to the elasticity and mass of the substrate. Thus only six independent elastic coefficients Qij describing in-plane elasticity in a general case can be obtained by this method. Assuming isotropic elasticity of the deposited layers, only two parameters (Q11 , Q33 ) were iteratively refined by matching of calculated and measured frequency shifts. Subsequently, they were recalculated [17] to G (shear) and E (Young’s) modulus. Vibration of the samples was excited by a thermoacoustic source with infrared pulsed laser. The resonant frequencies and shapes of individual modes were evaluated from signals recorded by the scanning laser vibrometer implemented in the optical microscope (Polytec Micro System Analyzer MSA 500). For each sample, 20–30 resonance modes were detected in the frequency range 0.3–2 MHz.
3. Results and discussion 3.1. The thickness and roughness The thickness of all DLC layers was in the region from 60 nm to 140 nm for Si (111) substrates and from 1200 nm to 1540 nm for Ti6Al4 V substrates. The lowest growth rate of the layers was for energy of ions from 30 eV to 50 eV for all the gases. Surface roughness Ra and RMS was not higher that 1 nm (measured on the Si substrates by mechanical profilometer). The layer topography was characterized by AFM, too. The layers were smooth with a small number of small droplets. (see Fig. 2.) The average surface roughness Ra was not higher than 0.4 nm (calculated from 10 m × 10 m area).
Please cite this article in press as: T. Kocourek, et al., Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.03.274
G Model
ARTICLE IN PRESS
APSUSC-35650; No. of Pages 5
T. Kocourek et al. / Applied Surface Science xxx (2017) xxx–xxx
3
Fig. 2. The layer topography was characterized by the AFM (the sample prepared with energy of 30 eV of Ar ions, on the Si substrate).
Fig. 3. The composition of elements of bombarded DLC films (determined by RBS on the Si substrates).
3.2. The composition The elemental composition was determined by RBS and by PIXE. Both methods have reach several micrometers in depth. Since the layers were roughly 0.1 m thick, the information is from the entire layer. The representation of other elements in the C layers is not significant, except for the layers bombarded by ions O2 or N2 , which exhibited increase of O or N elements on a significant level – the highest content for the lowest ion energy. The higher concentration of O and N was caused by implantation of these elements into the layer when utilizing low energies. In case of high energy ions (O2 , N2 ) the sputtering effect of the growing layer is stronger so the element easily escape. (as determined by RBS, see Fig. 3.) The content of trace elements was at most up to 0.3 at.% (Ar, Xe, W, Fe, Ni, Cr, etc.) determined by PIXE. Bombarded layers with the lowest content of trace elements (mainly O and N) were obtained with Ar ions.
3.3. The percentage of bonds The percentage of bonds (Csp3 , Csp2 , C O, C O, and C N) was determined by XPS. Number of bonds C O was approximately constant for all prepared layers. The bonds of C O were present in all layers except layers prepared with N2 ions. In layers prepared with N2 ions was found a significant number of C N bonds, which were not detected in the other layers. The Xe ions bombardment
with energy of 30 eV increased the content of Csp3 and Csp2 at the expense of C O and C O bonds, but the content of Csp2 bonds increased. The difference in mechanism between Xe and Ar ions can be contributed to different ion mass so different collision mechanism (see Fig. 4.). 3.4. The ratio of sp3 /sp2 bonds Maximum sp3 (83.63%) was observed for energy of 30 eV for Ar ions, which is more accurate than our past results [2], where the minimum energy was 40 eV (lower energy could not be reached at the time). Information about the effects of ions with energies lower than 30 eV could not be obtained because the device does not allow these modes of operation. Our experiments showed that the Ar ions with energy from 30 to 40 eV impacting the growing layers increase the content of sp3 carbon bonds and decrease the sp2 content. The sp3 /sp2 ratio increased from 2.8 (not bombarded) to 5.1 for the layers bombarded with Ar ions with the energy of 30 eV. The C C bonds have energy 285.2–285.4 eV for sp3 and 284.2–284.5 eV for sp2 . A small increase of the sp3 /sp2 ratio was observed for O2 ions with energy close to 80 eV. For the ion types used here, higher energy did not influence the formation of sp3 bonds. On the other hand, Xe and N2 ion bombardment (with energy 30–50 eV and 50–80 eV, respectively) increased the sp2 content, which led to deterioration of the DLC layer properties. It can be concluded that
Please cite this article in press as: T. Kocourek, et al., Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.03.274
G Model APSUSC-35650; No. of Pages 5
ARTICLE IN PRESS T. Kocourek et al. / Applied Surface Science xxx (2017) xxx–xxx
4
Fig. 4. The survey of bonds of bombarded DLC films (determined by XPS on the Si substrates).
Fig. 5. Graph of the values of sp3 bonds (percentage – left axis) and sp3 /sp2 bonds ratio (right axis) of DLC films depending on the energy of Ar, O2 , Xe, and N2 ions (measured on Si substrates).
(which was our aim), these sample were not selected for the elastic properties measurement (RUS method) and we focused on Ar ions. The influence of ion bombardment on the development of sp3 bonds, E (Young’s modulus) and G (shear modulus) is evident from Fig. 7a, b, c. where the dependences are roughly similar. For the non-bombarded DLC layers the E modulus was determined to be 565 GPa and G modulus 238 GPa. The E and G moduli increased to 601 GPa and 253 GPa, respectively with the increase of sp3 content with ion energy of 30 eV. For ion energy of 40–50 eV, the sp3 content and E and G moduli was lowered. However, the values were lower than the nonbombarded layers even though the sp3 /sp2 ratio was higher or the same. This is implies an influence of ion bombardment on elastic properties which are not connected to sp3 /sp2 ratio (density, porosity, texture, etc.). Modulus E decreased to 529 GPa and modulus G to 224 GPa. For the layers prepared using higher energy ions (80–150 eV) the values of E, G and sp3 /sp2 ratio did not differ significantly from the non-bombarded DLC layers. 4. Conclusions
the ions with lower energy have greater influence on the sp3
or sp2
bonds content (see Fig. 5.). We can conclude that there is an optimal ion mass and energy for the increase of sp3 bonds or the decrease of sp2 bonds, or both processes simultaneously. 3.5. Elastic properties The elastic properties of the substrate were evaluated by the RUS method modified for cylindrical samples (Ti6Al4 V substrates were polished with high geometrical accuracy) [18]. Subsequently, the layer was deposited and the spectrum was re-measured at identical conditions. The corresponding peaks (Fig. 6.) in both spectra were found by means of comparison of shapes of vibration modes, and the frequency shifts of the resonances were determined. The measured resonant frequency shifts were by two orders of magnitude higher than the uncertainty in evaluation of the resonant frequencies (spectral peaks position). Mass density of DLC was determined from close relationship to the measured sp3 contents. Since Xe, O2 , and N2 ion bombardment did not improve the sp3 /sp2 ratio
With our experimental setup, we were able to fabricate DLC layers with high content of sp3 bonds, maximum of 83.63% (determined by XPS), by PLD method with Ar ion bombardment during layer growth for ion energy of 30 eV. Other tested conditions (30–150 eV, Xe, O2 , N2 ) were not deemed perspective. The composition study showed negligible content of oxygen and nitrogen, and other element traces that are present in non-bombarded layers (determined by RBS and PIXE). The layers were smooth with small number of droplets (profilometer and AFM). For the cylindrical samples we showed the possibility of DLC layers measurement by non-destructive method RUS. For the layers fabricated with Ar ion bombardment with ion energy of 30 eV the Young modulus and shear modulus values reached 601 GPa and 253 GPa, respectively. We can conclude that it is possible to increase the sp3 bonds content and improve elastic properties of DLC layer via ion bombardment of low energy argon ions and that the RUS method bring a possibility to measure sample without damaging the surface. Next research in this field can bring a better understanding of processes taking place during ion bombardment of growing layer, as well as the develop-
Please cite this article in press as: T. Kocourek, et al., Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.03.274
G Model APSUSC-35650; No. of Pages 5
ARTICLE IN PRESS T. Kocourek et al. / Applied Surface Science xxx (2017) xxx–xxx
5
Fig. 6. The resonant spectra of free vibrating substrate before and after DLC deposition (measured on the Ti6Al4 V substrates, Ar ions, 30 eV).
References
Fig. 7. a, b, c. The graph of values of sp3 bonds (a), E (Young’s modulus) (b), and G (shear modulus) (c) of DLC films depending on the energy of Ar ions (measured on the Ti6Al4 V substrates).
ment in ultrasonic spectroscopic methods can bring the possibility of elastic properties measurement during manufacturing process, such as implant coatings.
Acknowledgments
[1] J. Robertson, Diamond-like amorphous carbon, Mater. Sci. Eng. R 37 (2002) 129–281. [2] M. Jelinek, P. Písaˇrík, T. Kocourek, J. Zemek, J. Lukeˇs, Influence of ion bombardment on growth and properties of PLD created DLC films, Appl. Phys. A 110 (2013) 943–947. ´ J. Kopeˇcek, ˚ zek, M. Landa, M. Jelínek, J. Mikˇsovsky, [3] T. Kocourek, M. Ruˇ Evaluation of elastic properties of DLC layers using resonant ultrasound spectroscopy and AFM nanoindentation, Surf. Coat. Technol. 205 (2011) 67–70. [4] X.L. Peng, Z.H. Barber, T.W. Clyne, Surface roughness of diamond-like carbon films prepared using various techniques, Surf. Coat. Technol. 138 (2001) 23–32. [5] Q. Jun, L. Jianbin, W. Shizhu, W. Jing, L. Wenzhi, Mechanical and tribological properties of non-hydrogenated DLC films synthesized by IBAD, Surf. Coat. Technol. 128–129 (2000) 324–328. [6] F. Rossi, Amorphisation and growth mechanisms of carbon films under ion beam irradiation, Chaos Solitons Fractals 10 (1999) 2019–2029. [7] A.R. Gonzalez-Elipe, F. Yubero, J.M. Sanz, Low Energy Ion Assisted Film Imperial Growth, College Press, 2003. [8] M. Rybachuk, J.M. Bell, Growth of diamond-like carbon films using low energy ion beam sputter – bombardment deposition with Ar ions, J. Phys.: Conf. Ser. 100 (2008) 082009. [9] I.S. Trakhtenberg, S.A. Plotnikov, V.B. Vykhodets, V.A. Pavlov, G.A. Raspopova, Modification of amorphous diamond-like films by bombardment with low-energy ions, Diamond Relat. Mater. 5 (1996) 943–946. [10] T. Yasui, T. Kotani, K. Fujiuchi, H. Tahara, T. Ypshikawa, Carbon nitride films deposited by reactive sputtering and pulsed laser ablation, Thin Solid Films 457 (2004) 133–138. [11] S.S. Tinchev, Surface modification of diamond-like carbon films to graphene under low energy ion beam irradiation, Appl. Surf. Sci. 258 (2012) 2931–2934. [12] J. Robertson, Deposition mechanisms for promoting sp3 bonding in diamond-like carbon, Diamond Relat. Mater. 2 (1993) 984–989. [13] P.A. Karaseov, O.A. Podsvirov, A.I. Titov, K.V. Karabeshkin, A. Ya Vinogradov, V.S. Belyakov, A.V. Arkhipov, L.M. Nikulina, A.L. Shakhmin, E.N. Shubina, N.N. Karasev, Effect of ion bombardment on the phase composition and mechanical properties of diamond-like carbon films, J. Surf. Invest. X-ray Synchrotron Neutron Tech. 8 (2014) 45–49. [14] C.A. Davis, A simple model for the formation of compressive stress in thin films by ion bombardment, Thin Solid Films 226 (1993) 30–34. [15] O.A. Podsvirov, P.A. Karaseov, A. Ya Vinogradov, A. Yu Azarov, N.N. Karasev, A.S. Smirnov, A.I. Titov, K.V. Karabeshkin, Residual stress in diamond-like carbon films- role of growth conditions and ion irradiation, J. Surf. Invest. X-Ray Synchrotron Neutron Tech. 4 (2010) 241–244. [16] D.A. Shirley, High-resolution X-ray photoemission spectrum of the valence bands of gold, Phys. Rev. B 5 (1972) 4709–4714. ˚ zek, P. Sedlák, H. Seiner, A. Kruisová, M. Landa, Linearized forward and [17] M. Ruˇ inverse problems of the resonant ultrasound spectroscopy for the evaluation of thin surface layers, Acoust. Soc. Am. 128 (2010) 3426–3437. [18] M. Janovská, P. Sedlák, H. Seiner, M. Landa, P. Marton, P. Ondrejkoviˇc, J. Hlinka, Anisotropic elasticity of DyScO3 substrates, J. Phys. Condens. Matter 24 (2012) 385404.
The project has been supported by Czech Grant Agency under GA15-05864S and the Grant Agency of the Czech Technical University in Prague under grant No. SGS16/190/OHK4/2T/17.
Please cite this article in press as: T. Kocourek, et al., Diamond-like carbon layers modified by ion bombardment during growth and researched by Resonant Ultrasound Spectroscopy, Appl. Surf. Sci. (2017), http://dx.doi.org/10.1016/j.apsusc.2017.03.274