- Email: [email protected]
Optical properties of diamond like carbon films containing copper, grown by high power pulsed magnetron sputtering and direct current magnetron sputtering: Structure and composition effects
Optical properties of diamond like carbon film containing copper, grown by high power pulsed magnetron sputtering and direct current magnetron sputtering: Structure and composition effects Š. Meškinis, A. Čiegis, A. Vasiliauskas, K. Šlapikas, T. Tamulevičius, A. Tamulevič ienė , S. Tamulevič ius PII: DOI: Reference:
S0040-6090(14)01165-1 doi: 10.1016/j.tsf.2014.11.045 TSF 33915
To appear in:
Thin Solid Films
Please cite this article as: Š. Meškinis, A. Č iegis, A. Vasiliauskas, K. Šlapikas, T. Tamulevičius, A. Tamulevi čienė, S. Tamulevičius, Optical properties of diamond like carbon films containing copper, grown by high power pulsed magnetron sputtering and direct current magnetron sputtering: Structure and composition effects, Thin Solid Films (2014), doi: 10.1016/j.tsf.2014.11.045
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Optical properties of diamond like carbon films containing copper, grown by high power pulsed magnetron sputtering and direct current magnetron sputtering:
IP T
structure and composition effects
Š. Meškinis, A. Čiegis, A. Vasiliauskas, K. Šlapikas, T. Tamulevičius,
CR
A. Tamulevičienė, S. Tamulevičius
US
Kaunas University of Technology, Institute of Materials Science, Baršausko 59, Kaunas,
MA N
Lithuania
In the present study chemical composition, structure and optical properties of hydrogenated diamond like carbon films containing copper (DLC:Cu films) deposited by reactive magnetron sputtering were studied. Different modes of deposition – direct current (DC)
ED
sputtering and high power pulsed magnetron sputtering (HIPIMS) as well as two configurations of the magnetron magnetic field (balanced and unbalanced) were applied. X-
PT
ray diffractometry, Raman scattering spectroscopy, energy-dispersive X-ray spectroscopy and atomic force microscopy were used to study structure and composition of the films. It was
CE
shown that by using HIPIMS mode contamination of the cathode during deposition of DLC:Cu films can be suppressed. In all cases oxygen atomic concentration in the films was in
AC
5-10 at.% range and it increased with the copper atomic concentration. The highest oxygen content was observed in the films deposited employing low ion/neutral ratio balanced DC magnetron sputtering process. According to the analysis of the parameters of Raman scattering spectra, sp3/sp2 bond ratio decreased with the increase of Cu atomic concentration in the DLC films. Clear dependence of the extinction, absorbance and reflectance spectra on copper atomic concentration in the films was observed independently of the method of deposition. Surface plasmon resonance effect was observed only when Cu atomic concentration in DLC:Cu film was at least 15 at.%. The maximum of the surface plasmon resonance peak of the absorbance spectra of DLC:Cu films was in 600-700 nm range and redshifted with the increase of Cu amount. Ratio between the intensities of the plasmonic peak and hydrogenated amorphous carbon related peak at ~220 nm both in the extinction and absorbance spectra as well as peak to background ratio of DLC:Cu films increased linearly with Cu amount in the investigated 0-40 atomic percents range. Reflectance of the plasmonic
1
ACCEPTED MANUSCRIPT DLC:Cu films were in 30-50% range that could be important in practical optical applications
AC
CE
PT
ED
MA N
US
CR
IP T
of DLC:Cu films.
2
ACCEPTED MANUSCRIPT 1. Introduction
Nanoparticles of the group IB metals such as copper, silver and gold received a
IP T
significant interest as plasmonic nanomaterials. In the case of Cu and Au, surface plasmon resonance effect can be observed in very similar range of electromagnetic waves [1]. In this case Cu has some advantages over gold. Copper is substantially cheaper material. It is
CR
compatible with semiconductor device technology [2]. However problem of the copper surface oxidation arises [3]. Nanocomposites where nanoparticles are embedded in a matrix
US
can be used to solve this problem. Particularly, copper containing diamond like carbon (DLC:Cu) films are deposited in the form of the metal (metal oxide) nanoparticles inserted
MA N
into the diamond like carbon matrix [4-12]. Similar tendencies were reported for the silver containing and gold containing DLC films, too [12-15]. Despite growing interest to copper as a plasmonic nanomaterial, along with most often used silver and gold, there are very few studies on optical properties of the copper containing DLC films [6,9,11]. In the absorbance
ED
spectra of DLC:Cu films deposited by reactive diode sputtering only traces of the plasmonic peaks were observed [11]. In [6,9] optical properties of DLC:Cu films synthesized by electrochemical deposition were studied and plasmonic peaks were observed in the
PT
absorbance spectra of some samples. However, in these studies no data on chemical composition of the films as well as on the structure of the carbon matrix were provided. In
CE
addition, it should be mentioned that electrochemical deposition is rarely used for synthesis of DLC films and it is only suitable for conductive substrates.
AC
It should be mentioned that diamond like carbon (DLC) is amorphous allotrope of carbon consisting of the sp2 bonded (graphite-like) carbon nanoclusters embedded into the sp3 bonded (diamond-like) carbon matrix [16-18]. It may contain from zero up to several tens of atomic percents of hydrogen [16-18]. Usually optical transmittance of DLC in a visible light range as well as hardness and Young’s modulus increase with sp3/sp2 bond ratio. In such a way, hardness of DLC can reach up to 80% of the diamond hardness, while optical transmittance may be comparable with transmittance of the nanocrystalline diamond films [19]. In the present study the influence of composition and structure on optical properties of DLC:Cu films deposited by reactive magnetron sputtering were investigated. Along with direct current (DC) magnetron sputtering, innovative deposition method - high power pulsed magnetron sputtering (HIPIMS) - was applied to deposit the diamond like carbon films. HIPIMS deposition method provides high ion/neutral ratio [20] preferable for formation of 3
ACCEPTED MANUSCRIPT DLC films containing high amount of sp3 bonded carbon [16]. Plasma density in HIPIMS is similar to the case of the cathodic arc deposition and pulsed laser ablation – synthesis methods that are most often used to deposit high sp3/sp2 bond ratio films [20]. Yet main drawbacks of the cathodic arc deposition and pulsed laser ablation (contamination of the film
IP T
by droplets of target and problems with large area deposition) are avoided [20].
CR
2. Experimental techniques
US
In the present study reactive magnetron sputtering was used for deposition of DLC:Cu films. Two modes of operation (direct current (DC) and HIPIMS) were used. Two
MA N
configurations of the magnetron magnetic field (balanced and unbalanced) were applied. In all cases mixture of the hydrocarbons (acetylene) and argon gas was used. The diameter of magnetron was 3″, the copper target was used.
In the case of HIPIMS deposition various pulse currents in 2.4–30 A range were used.
ED
Pulse on time in all cases was set 100 µs, duty cycle 1%, frequency 100 Hz. Base pressure was 5×10-4 Pa and work pressure (4±1)×10-1 Pa was maintained throughout the deposition
PT
process. In the case of DC sputtering, magnetron target current was 0.1 A, base pressure was
5×10-4 Pa and work pressure was (4±1)×10-1 Pa. Thin films were deposited on
CE
monocrystalline silicon and quartz substrates. In all experiments substrate-target gap was set at 0.1 m, the substrates were grounded. Thickness of the deposited films in all cases was in the
AC
range 50-100 nm.
Optical properties of the films were investigated using an optical spectrometer Avantes that is composed of a deuterium halogen light source (AvaLight DHc) and spectrometer (Avaspec-2048). Extinction (complete optical losses due to reflection and absorption of light), absorbance and reflectance of the films were analyzed in the wavelength region from 180nm to 1100nm. Microstructure of DLC:Cu films was studied and linear dimensions of the copper nanoparticles/clusters in DLC matrix were estimated by employing field emission scanning electron microscope (FE-SEM) FEI Quanta 200 FEG in high vacuum mode. Theoretical resolution of the FESEM at 30 kV accelerating voltage is 1.2 nm in high vacuum mode. Chemical composition of the
films was studied by energy-dispersive X-ray spectrometer Bruker Quantax system with XFlash 4030 detector attached on the field emission scanning electron microscope (FEISEM).
4
ACCEPTED MANUSCRIPT Raman scattering measurements were performed using Raman microscope inVia (Renishaw) with 532 nm excitation. Integration time was 100s, power 0.3 mW, grating groove density 2400 grooves/mm. Structure of the crystalline copper nanoclusters was studied by X-ray diffractometer
IP T
D8 Advance (Bruker AXS, Germany). Grazing incidence angle arrangement combined with parallel beam X-ray diffraction geometry was used. Multilayer Ni/graphite parabolic
CR
monochromator was placed in front of the sample. X-ray diffraction patterns were recorded using Cu cathode at 40 kV, anode current 40 mA, scanning step D2Q=0.04° and average time of integration was 15s.
US
The morphology of the surface was analyzed with atomic force microscope NanoWizard®3 (JPK, Germany) working in AC Mode. Silicon probes with a reflective
MA N
backside Al coating (ACTA-10, APPNano, USA) with a resonance frequency of 200-400 kHz and force constant of 13-77 N/m were used. Nominal tip radius was less than 10 nm. The scanning rate of 0.8 Hz was selected.
PT
ED
3. Experimental results
CE
3.1 Deposition conditions effects on target contamination
Target contamination by the products of reactions with reactive gas components is one
AC
of the main problems related to the reactive magnetron sputtering [21]. In the present study during the deposition of DLC:Cu films containing copper (up to 30 at.%) by DC magnetron sputtering resulted in contamination of the sputtering target by carbon film and resultant subsequent drop of the deposition rate during growth of the film. According to our findings this problem can be avoided by using HIPIMS. For DLC:Cu films deposited by reactive HIPIMS such a problem was observed only for the films containing small amounts of copper (up to several atomic percents).
3.2 Structure and composition
Raman scattering spectra of the DLC:Cu films deposited by unbalanced HIPIMS are presented in Fig. 1. In all cases the spectra were typical for diamond like carbon [16-18]. There were no clear differences between the Raman scattering spectra of DLC:Cu films 5
ACCEPTED MANUSCRIPT deposited by both balanced and unbalanced HIPIMS as well as balanced DC magnetron sputtering. In all spectra G peak (stretching vibration mode of sp2 bonded carbon) at ~15001600 cm-1 and shoulder at ~200-1400 cm-1 related to disorder-induced D peak (breathing modes of sp2 bonded carbon rings) [16-18] were observed.
IP T
SEM study revealed nanocomposite microstructure of DLC:Cu films: copper nanoclusters embedded into the diamond like carbon matrix. From the FE-SEM micrographs
CR
of DLC:Cu nanocomposite film (Fig. 2) one can see 10-20nm sized bright features that could be attributed to the copper nanoparticles as the metal gives higher electron contrast then dielectric matrix i.e. DLC.
US
X-ray diffraction (XRD) difractograms of DLC:Cu films are presented in Fig. 3. In the case of DLC:Cu film deposited by balanced HIPIMS and containing 26 at.% of copper,
MA N
Cu(111) XRD peak as well as weak Cu(200) peak can be seen. Particularly the peaks became well expressed in the case of the thicker films (400 nm), but they are seen in the thinner films (100 nm), too. Nanocrystallite size evaluated from Cu(111) peak according to Sherrer equation was 5.1 and 4.1 nm in the case of the thinner and thicker DLC:Cu film respectively.
ED
It should be mentioned that other researchers reported similar results for DLC:Cu films. In [22] for DLC:Cu films deposited by reactive cathodic arc evaporation broad Cu(111) XRD
PT
peak was reported when Cu atomic concentration in the film was ~6.5 at.%. Cu(111) peak along with weak Cu(200) peak was observed when Cu amount reached 10 at.% [22]. In the
CE
case of DLC:Cu films deposited by reactive magnetron sputtering and containing >27 at.% Cu stronger Cu(111) peak and weaker Cu(200) XRD peaks were observed [23]. In [6,9] Cu(111)
AC
XRD peak was reported for DLC:Cu films deposited by electrochemical synthesis. In [11] weak Cu(111) peak was observed in XRD spectra of DLC:Cu films deposited by reactive RF diode sputtering. In [22] average crystallite size of DLC:Cu film deposited by cathodic arc deposition was reported to be 15 nm for the film containing 10 at.% Cu and 3.5 nm for film containing 3 at.% Cu. Such a difference from the results observed in the present study possibly can be explained by peculiarities of the cathodic arc deposition process such as presence of the droplets of Cu target in growing DLC:Cu film resulting in larger average crystallite size. In [23] average crystallite size of DLC:Cu film deposited by radio frequency plasma enhanced reactive magnetron sputtering was 10 nm for the film containing 27.85 at.% Cu (nearly the same amount as in the present study). In this case differences in crystallite size can be explained by differences in deposition process used such as application of methane instead of acetylene in [23] resulting in larger hydrogen contents during films growth as well as use of the additional RF discharge and substrate bias. It should be mentioned that in [24] it
6
ACCEPTED MANUSCRIPT was shown that in the case of DLC:Ag films application of the substrate bias can result in substantially increased average crystallite size.
IP T
3.3 The influence of Cu atomic concentration
In the following section we will analyse properties of the DLC:Cu films versus Cu
CR
atomic concentration. We have found that such a parameter enables to analyse properties of the films deposited by different modes of the operation of magnetron.
US
It can be seen in Fig. 4 that oxygen atomic concentration was in 5-10 at.% range and correlated with the content of copper. It increased with copper atomic concentration for the films deposited by all methods used in the present study (unbalanced HIPIMS, balanced
MA N
HIPIMS, balanced DC magnetron sputtering). Tendency of the decrease of oxygen content in film with the increase of energy flux during the deposition process was observed. The highest oxygen content was found in the films deposited employing low ion/neutral ratio balanced DC magnetron sputtering process. In the case of the high ion/neutral ratio balanced HIPIMS
ED
deposition process, oxygen atomic concentration was lower. The lowest oxygen concentration was observed in the DLC:Cu films deposited by unbalanced HIPIMS process providing both
PT
high ion/neutral ratio and higher ion energy. D/G peak area ratio of DLC:Cu films increased with the Cu atomic concentration (Fig.
CE
5). It should be mentioned that such a result is in good accordance with numerous studies where increased D/G ratio as well as decreased sp 3/sp2 ratio with increase metal concentration
AC
in DLC film was reported for metal containing diamond like carbon films [13,15,25-29]. No clear dependence of D/G ratio on DLC:Cu film deposition method used was observed. Surface root mean square roughness (RRMS) of the DLC:Cu films was low and didn‘t exceed 2 nm (Fig. 6). No clear dependence of RRMS of DLC:Cu films on copper amount or deposition method used was observed. It should be mentioned that surface roughness of „conventional“ hydrogenated DLC films containing no other chemical elements except carbon and hydrogen (see e.g. [30]) is usually at the same level as the lowest surface roughness reported in the present study.
3.4 Optical properties
Optical properties of DLC:Cu films were studied relating them to the Cu concentration and method of deposition. Extinction spectra of DLC:Cu films are presented in Fig. 7. Optical 7
ACCEPTED MANUSCRIPT properties of the films were independent of the deposition method used but clear dependence of the extinction (reflection+absorbance), absorbance and reflectance spectra on copper atomic concentration in the films was observed. It should be noted that some surface plasmon resonance peak was registered in the
IP T
absorption spectrum only when Cu atomic concentration in the film was at least 15 at.% (Fig. 7). Such a behavior is very different from the case of DLC:Ag [24] and DLC:Au [27] films
CR
where surface plasmon resonance peak was detected when silver atomic or gold atomic concentration was several percents (3 at.% or 2 at.% respectively). Surface plasmon resonance peak for DLC:Cu films was in 600-700 nm range. Similar
US
position of the optical absorption plasmonic peak was reported for DLC:Cu films deposited by electrochemical deposition [6,9]. Plasmonic peak redshifted with the increase of Cu atomic
MA N
concentration in the films. It can be mentioned that similar behavior (redshift of plasmonic peak with Ag atomic concentration) was reported for DLC:Ag films [24]. In all cases plasmonic peaks were relatively broad in comparison with ones reported for DLC:Ag and DLC:Au films [24,27,28,29,31]. It is in good accordance with data on
ED
DLC:Cu films deposited by electrochemical deposition [6,9]. Similarly in [3] broad plasmonic peak was reported for the oxidized Cu nanoparticles, while removal of copper
PT
oxide resulted in the increased intensity and increase of the width of Cu plasmonic peak that was comparable with one measured in the case of silver and gold nanoparticles. Taking into
CE
account data on chemical composition of the deposited nanocomposite films presented in this study, it seems that for DLC:Cu films even 5-10 atomic percents of oxygen is enough for
AC
formation of the copper oxide overlayer on the copper nanoclusters and resultant decrease of the surface plasmon resonance peak intensity. It should be mentioned that plasmonic peak can be seen as well in the reflectance spectra of the DLC:Cu films containing >15 at.% Cu. Position of the maximum of this peaks is redshifted in comparison with the absorbance spectra. Tendency of the increased reflectance up to 30-50% with Cu amount should be mentioned. Ratio between the intensities of the plasmonic peak and hydrogenated amorphous carbon (a-C:H) related peak at ~220 nm both for the extinction and absorbance spectra of DLC:Cu films increased with Cu amount linearly in 0-40 atomic percents range (Fig. 8). Particularly, it means that measurement of DLC:Cu film optical extinction or absorbance spectra can be used for rough estimation of copper atomic concentration in the film. On the other hand, plasmonic film could be used as selective optical filter [32]. For these reasons ratio of the plasmonic peak to the background of extinction and absorbance spectra of DLC:Cu films was calculated as a measure of contrast of such potential selective 8
ACCEPTED MANUSCRIPT optical filter. It can be seen that in the investigated range of 0-40 atomic percents of Cu, peak to background ratio increased linearly (Fig. 9). In the case of the optical absorption spectra this ratio was higher in comparison with the case of the extinction spectra. It can be clearly seen in Fig. 9, as well. Thus in real optical applications of the DLC:Cu films, large reflectance
IP T
of these films should be taken into account, not only the problem of the copper oxidation pointed out in [3].
CR
It should be mentioned that oxygen atomic concentration, structure of the amorphous carbon matrix and roughness of the film may have significant influence on the peak to background ratio. According to [3], removal of oxide overlayer from Cu nanoparticles
US
resulted in significant improvement of plasmonic peak to background ratio. In [33] modeling of absorbance spectra of DLC:Ag films revealed that increased sp3/sp2 bond ratio should
MA N
result in the increased peak to background ratio of the plasmonic peak. According to [34] increased surface roughness may result in longer optical way of the light and increased absorption. In addition larger content of the bonded hydrogen in DLC film results in increased optical transparency of the film [35]. According to our results (Fig. 4-6) in some cases
ED
significant differences of the structure of amorphous carbon matrix as well as surface roughness and oxygen atomic concentration were observed for the DLC:Cu films containing
PT
the same amount of copper. However peak to background ratio of DLC:Cu films containing the same content of copper differed no more than 10%. Analysis of the parameters mentioned
CE
above, revealed that observed disparities can not be explained by any separate parameters mentioned above. I.e. this dependence is complex and is influenced by all the discussed
AC
parameters. It should be mentioned as well that to improve plasmonic properties of DLC:Cu films one should use the mode of deposition enabling large amount of sp3 phase and low concentration of oxygen (preferable below 5 at.% ).
Conclusions
Structure and optical properties of diamond like carbon films containing copper (DLC:Cu) deposited by reactive magnetron sputtering using different modes (high power pulsed magnetron sputtering (HIPIMS) and direct current (DC) magnetron sputtering) were studied. It was shown that contamination of cathode taking place during DC magnetron sputtering of DLC:Cu films (containing up to 30 at.% of copper) can be avoided using the HIPIMS deposition mode. For both used modes of the deposition, Raman scattering spectra of the investigated DLC:Cu films were typical for diamond like carbon. XRD patterns included 9
ACCEPTED MANUSCRIPT Cu(111) peak and low intensity Cu(200) peak. In all cases oxygen atomic concentration in the films was in 5-10% range and tendency of the increase of oxygen atomic concentration with the increase of copper atomic concentration was observed. The highest oxygen content was found in the films deposited by low ion/neutral ratio balanced DC magnetron sputtering
IP T
process. Analysis of D/G ratio in the Raman scattering spectra revealed decreased sp 3/sp2 ratio with the increase of Cu atomic concentration in the film. No clear dependence of surface
CR
roughness (RRMS) of DLC:Cu films on copper amount was observed, but optical properties of the films, i.e. the extinction, absorbance and reflectance spectra were dependent on copper atomic concentration in the films as well as on deposition mode. Surface plasmon resonance
US
effect was observed only when Cu atomic concentration in the film was at least 15 at.%. Peak in the absorption spectra due to the surface plasmon resonance was in 600-700 nm range and
MA N
redshifted with the increase of Cu amount. Reflectance of the plasmonic DLC:Cu films was in the 30-50% range. Ratio between the intensities of the plasmonic peak and a-C:H related peak (registered at ~220 nm) in the extinction (and absorbance spectra) as well as ratio of the surface plasmon resonance peak to the background ratio for DLC:Cu films increased with Cu
ED
amount linearly in the investigated 0-40 atomic percent’s range. It was shown that this kind of measurement of optical extinction (or absorbance) spectra of DLC:Cu film can be used for
PT
estimation of copper atomic concentration in the film. On the other hand, large reflectance of
AC
CE
DLC:Cu films should be taken into account in different optical applications.
10
ACCEPTED MANUSCRIPT Acknowledgements
This research was funded by the European Social Fund under the Global Grant
CR
References
IP T
measure (project No VP1-3.1-ŠMM-07-K-03-057).
[1] B. Sharma, R.R. Frontiera, A.-I. Henry, E. Ringe, R.P. Van Duyne, SERS: Materials,
US
applications, and the future, Mater. Today 15 (2012) 16-25.
[2] S.M. Merchant, S. H. Kang, M. Sanganeria, B. van Schravendijk, T. Mountsier, Copper
MA N
Interconnects for Semiconductor Devices, JOM 53 (2001) 43-48. [3] G.H. Chan, J. Zhao, E.M. Hicks, G.C. Schatz, R.P. Van Duyne, Plasmonic Properties of Copper Nanoparticles Fabricated by Nanosphere Lithography, Nano Lett. 7 (2007) 19471952.
ED
[4] C.-C. Chen, F. Chau-Nan Hong, Structure and properties of diamond-like carbon nanocomposite films containing copper nanoparticles, Appl. Surf. Sci. 242 (2005) 261–269
PT
[5] Hua Pang, Xingquan Wang, Guling Zhang, Huan Chen, Guohua Lv, Size Yang, Characterization of diamond-like carbon films by SEM, XRD and Raman spectroscopy, Appl.
CE
Surf. Sci. 256 (2010) 6403–6407. [6] S. Hussain, A.K. Pal, Surface plasmon effect in nanocrystalline copper/DLC composite
AC
filmsby electrodeposition technique, Bull. Mater. Sci., 29 (2006) 553–557. [7] V.I. Ivanov-Omskii, L.K. Panina, S.G. Yastrebov, Amorphous hydrogenated carbon doped with copper as antifungal protective coating, Carbon 38 (2000) 495–499. [8] Meng-Yuan Tsai, Mao-SuanHuang, Li-KaiChen, Yun-DunShen, Ming-Hong Lin, YiChen Chiang, Keng-Liang Ou, Shih-Fu Ou Surface properties of copper-incorporated diamond-like carbon films deposited by hybrid magnetron sputtering, Ceram. Int. 39 (2013) 8335–8340. [9] S Hussain, R K Roy and A K Pal, Surface plasmon effect in nanocrystalline copper/DLC composite films by an electrodeposition technique, J. Phys. D: Appl. Phys. 38 (2005) 900– 908. [10] A.S. Chaus , T.N. Fedosenko, A.V. Rogachev, L. Čaplovič, Surface, microstructure and optical properties of copper-doped diamond-like carbon coating deposited in pulsed cathodic arc plasma, Diam. Relat. Mater. 42 (2014) 64–70.
11
ACCEPTED MANUSCRIPT [11] N. Dwivedi, S. Kumar, H.K. Malik, C.Sreekumar, S. Dayal, C.M.S. Rauthan, O.S.Panwar, Investigation of properties of Cu containing DLC films produced by PECVD process, J. Phys. Chem. of Solids 73 (2012) 308–316. [12] I. Gerhards, C. Ronning, U. Vetter, H. Hofsass, H. Gibhardt, G. Eckold, Q. Li, S.T. Lee,
IP T
Y.L. Huang, M. Seibt, Ion beam synthesis of amorphous carbon thin films containing metallic nanoclusters, Surf. Coat. Technol. 158-159 (2002) 114–119.
CR
[13] Heon Woong Choi, Jung-Hae Choi, Kwang-Ryeol Lee, Jae-Pyoung Ahn, Kyu Hwan Oh, Structure and mechanical properties of Ag-incorporated DLC films prepared by a hybrid ion beam deposition system, Thin Solid Films 516 (2007) 248-251.
US
[14] K. Baba, R. Hatada, S. Flege, W. Ensinger, Preparation and Properties of Ag-Containing
Eng. 2012 (2012) Art. ID 536853.
MA N
Diamond-Like Carbon Films by Magnetron Plasma Source Ion Implantation, Adv. Mater. Sci.
[15] Heon Woong Choi, R.H. Dauskardt, Seung-Cheol Lee, Kwang-Ryeol Lee, Kyu Hwan Oh, Characteristic of silver doped DLC films on surface properties and protein adsorption, Diam. Relat. Mater. 17 (2008) 252-257.
ED
[16] J. Robertson, Diamond Like Amorphous Carbon, Mater. Sci. Eng. R, 37 (2002), pp. 129281.
J. Robertson, Comparison of diamond-like carbon to diamond for applications,
PT
[17]
Phys. Stat. Sol. A, 205 (2008) 2233-2244. A.C. Ferrari, J. Robertson, Raman spectroscopy of amorphous, nanostructured,
CE
[18]
diamond-like carbon, and nanodiamond, Phil. Trans. R. Soc. Lond. A, 362 (2004) 2477–2512.
AC
[19] B. Mednikarov, G. Spasov, Tz. Babeva, J. Pirov, M. Sahatchieva, C. Popova, W.Kulischa, Optical properties of diamond-like carbon and nanocrystalline diamond films, J. Optoelectron. Adv. Mat., 7 (2005) 1407 – 1413. [20] K. Sarakinos, J. Alami, S. Konstantinidis, High power pulsed magnetron sputtering: A review on scientific and engineering state of the art, Surf. Coat. Technol. 204 (2010) 1661– 1684. [21] S. Berg, T. Nyberg, Fundamental understanding and modelling of reactive sputtering of oxides. Thin Solid Films 476 (2005) 215-230. [22] Haiqiang Zhang, Yiming Chen, Bin Liao, Xianying Wu, Huixing Zhang, Xu Zhang, Effect of C2H2 flow rate on microstructure and properties of nc–Cu/a–C:H nanocomposite films prepared by filtered cathodic vaccum arc technique, Nucl. Instrum. Meth. B 307 (2013) 137–142.
12
ACCEPTED MANUSCRIPT [23] Yu-Hao Chan, Chung-Fang Huang, Keng-Liang Ou, Pei-Wen Peng, Mechanical properties and antibacterial activity of copper doped diamond-like carbon films, Surf. Coat. Technol. 206 (2011) 1037–1040. [24] Š. Meškinis, A. Čiegis, A. Vasiliauskas, A. Tamulevičienė, K. Šlapikas, R. Juškėnas, G.
IP T
Niaura, S. Tamulevičius, Plasmonic Properties of Silver Nanoparticles Embedded in Diamond Like Carbon Films: Influence of Structure and Composition, Appl. Surf. Sci. 317 (2014)
CR
1041-1046.
[25] Z. Montiel-González, S.E. Rodil, S. Muhl, A. Mendoza-Galván, L. RodríguezFernández, Amorphous Carbon Gold Nanocomposite Thin Films: Structural and Spectro-
US
ellipsometric Analysis, Thin Solid Films, 519 (2011) 5924–5932. [26] P. Vijai Bharathy, D. Nataraj, Paul K. Chu, Huaiyu Wang, Q. Yang, M.S.R.N. Kiran,
MA N
J. Silvestre-Albero, D. Mangalaraj, Effect of titanium incorporation on the structural, mechanical and biocompatible properties of DLC thin films prepared by reactive-biased target ion beam deposition method, Appl. Surf. Sci. 257 (2010) 143–150. [27] R. Paul, S.R. Bhattacharyya, R. Bhar, A.K. Pal, Modulation of residual stress in diamond
ED
like carbon films with incorporation of nanocrystalline gold, Appl. Surf. Sci. 257 (2011) 10451– 10458.
PT
[28] R. Paul, S. Hussain, A.K. Pal, Characterization of nanocrystalline gold/DLC composite films synthesized by plasma CVD technique, Appl. Surf. Sci. 255 (2009) 8076–8083.
CE
[29] R. Paul, S. Hussain, S. Majumder, S. Varma, A.K. Pal, Surface plasmon characteristics of nanocrystalline gold/DLC composite films prepared by plasma CVD technique, Mater. Sci.
AC
Eng. B 164 (2009) 156–164. [30] J. Robertson. Ultrathin carbon coatings for magnetic storage technology, Thin Solid Films 383 (2001) 81-88. [31] S. Hussain, R.K. Roy, A.K. Pal, Incorporation of silver nanoparticles in DLC matrix and surface plasmon resonance effect, Mater. Chem. Phys. 99 (2006) 375–381. [32] G. Kedawat, B.K. Gupta, P. Kumar, J. Dwivedi, A. Kumar, N.K. Agrawal, S.S. Kumar, Y.K. Vijay, Fabrication of a Flexible UV Band-Pass Filter Using Surface Plasmon Metal−Polymer Nanocomposite Films for Promising Laser Applications, Appl. Mater. Interfaces 6 (2014) 8407−8414. [33] H. Zoubos, L.E.Koutsokeras, D.F.Anagnostopoulos, E.Lidorikis, S.A.Kalogirou, A.R.Wildes, P.C.Kelires, P.Patsalas, Broadband optical absorption of amorphous carbon/Ag nanocomposite films and its potential for solar harvesting applications, Sol. Energ. Mat. Sol. C. 117 (2013) 350–356.
13
ACCEPTED MANUSCRIPT [34] S. Mokkapati, K. R. Catchpole, Nanophotonic light trapping in solar cells, J. Appl. Phys. 112 (2012) 101101. [35] C. Casiraghi, F. Piazza, A.C. Ferrari, D. Grambole, J. Robertson, Bonding in hydrogenated diamond-like carbon by Raman spectroscopy, Diam. Relat. Mater. 14 (2005)
AC
CE
PT
ED
MA N
US
CR
IP T
1098– 1102.
14
ACCEPTED MANUSCRIPT List of the captures
Fig. 1. Typical Raman scattering spectra of DLC:Cu films (all films presented were deposited
IP T
by unbalanced HIPIMS).
CR
Fig. 2. Typical SEM microgram of DLC:Cu film. The film was deposited by unbalanced
US
HIPIMS and contains 34 at.% Cu.
Fig. 3. X-ray diffraction patterns of DLC:Cu films deposited by balanced HIPIMS and
MA N
containing 26 at.% Cu.
ED
Fig. 4. Oxygen atomic concentration Vs Cu atomic concentration in DLC:Cu films.
PT
Fig. 5. D/G ratio peak area ratio Vs Cu atomic concentration.
CE
Fig. 6. Surface roughness of DLC:Cu films.
AC
Fig. 7. Typical optical extinction (absorbance+reflectance) (a), absorbance (b) and reflectance (c) spectra of DLC:Cu films containing different amounts of copper.
Fig. 8. Dependence of intensity of the plasmonic peak normalized to a-C:H absorption peak intensity Vs Cu atomic concentration in DLC:Cu film. Ratio was calculated for the extinction (a) and absorbance (b) spectra.
Fig. 9. Ratio of the plasmonic peak to background of extinction (a) and absorbance (b) spectra of DLC:Cu films Vs Cu atomic concentration in DLC:Cu films.
15
AC
CE
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
Figure 1
16
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
Figure 2
17
CE
AC
Figure 3
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
18
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 4
19
AC
CE
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
Figure 5
20
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 6
21
AC
CE
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
Figure 7a
22
AC
CE
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
Figure 7b
23
CE
AC
Figure 7c
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
24
CE
AC
Figure 8a
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
25
CE
AC
Figure 8b
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
26
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 9a
27
AC
CE
Figure 9b
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
28
ACCEPTED MANUSCRIPT Highlights Optical properties of diamond like carbon (DLC:Cu) films containing Cu were
IP T
studied.
CR
Reactive high power pulsed magnetron sputtering was used for deposition.
US
Extinction, absorbance and reflectance spectra of DLC:Cu films depend on Cu content.
AC
CE
PT
ED
MA N
Plasmonic peak intensity and position of the peak depend on Cu amount.
29
S0040-6090(14)01165-1 doi: 10.1016/j.tsf.2014.11.045 TSF 33915
To appear in:
Thin Solid Films
Please cite this article as: Š. Meškinis, A. Č iegis, A. Vasiliauskas, K. Šlapikas, T. Tamulevičius, A. Tamulevi čienė, S. Tamulevičius, Optical properties of diamond like carbon films containing copper, grown by high power pulsed magnetron sputtering and direct current magnetron sputtering: Structure and composition effects, Thin Solid Films (2014), doi: 10.1016/j.tsf.2014.11.045
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Optical properties of diamond like carbon films containing copper, grown by high power pulsed magnetron sputtering and direct current magnetron sputtering:
IP T
structure and composition effects
Š. Meškinis, A. Čiegis, A. Vasiliauskas, K. Šlapikas, T. Tamulevičius,
CR
A. Tamulevičienė, S. Tamulevičius
US
Kaunas University of Technology, Institute of Materials Science, Baršausko 59, Kaunas,
MA N
Lithuania
In the present study chemical composition, structure and optical properties of hydrogenated diamond like carbon films containing copper (DLC:Cu films) deposited by reactive magnetron sputtering were studied. Different modes of deposition – direct current (DC)
ED
sputtering and high power pulsed magnetron sputtering (HIPIMS) as well as two configurations of the magnetron magnetic field (balanced and unbalanced) were applied. X-
PT
ray diffractometry, Raman scattering spectroscopy, energy-dispersive X-ray spectroscopy and atomic force microscopy were used to study structure and composition of the films. It was
CE
shown that by using HIPIMS mode contamination of the cathode during deposition of DLC:Cu films can be suppressed. In all cases oxygen atomic concentration in the films was in
AC
5-10 at.% range and it increased with the copper atomic concentration. The highest oxygen content was observed in the films deposited employing low ion/neutral ratio balanced DC magnetron sputtering process. According to the analysis of the parameters of Raman scattering spectra, sp3/sp2 bond ratio decreased with the increase of Cu atomic concentration in the DLC films. Clear dependence of the extinction, absorbance and reflectance spectra on copper atomic concentration in the films was observed independently of the method of deposition. Surface plasmon resonance effect was observed only when Cu atomic concentration in DLC:Cu film was at least 15 at.%. The maximum of the surface plasmon resonance peak of the absorbance spectra of DLC:Cu films was in 600-700 nm range and redshifted with the increase of Cu amount. Ratio between the intensities of the plasmonic peak and hydrogenated amorphous carbon related peak at ~220 nm both in the extinction and absorbance spectra as well as peak to background ratio of DLC:Cu films increased linearly with Cu amount in the investigated 0-40 atomic percents range. Reflectance of the plasmonic
1
ACCEPTED MANUSCRIPT DLC:Cu films were in 30-50% range that could be important in practical optical applications
AC
CE
PT
ED
MA N
US
CR
IP T
of DLC:Cu films.
2
ACCEPTED MANUSCRIPT 1. Introduction
Nanoparticles of the group IB metals such as copper, silver and gold received a
IP T
significant interest as plasmonic nanomaterials. In the case of Cu and Au, surface plasmon resonance effect can be observed in very similar range of electromagnetic waves [1]. In this case Cu has some advantages over gold. Copper is substantially cheaper material. It is
CR
compatible with semiconductor device technology [2]. However problem of the copper surface oxidation arises [3]. Nanocomposites where nanoparticles are embedded in a matrix
US
can be used to solve this problem. Particularly, copper containing diamond like carbon (DLC:Cu) films are deposited in the form of the metal (metal oxide) nanoparticles inserted
MA N
into the diamond like carbon matrix [4-12]. Similar tendencies were reported for the silver containing and gold containing DLC films, too [12-15]. Despite growing interest to copper as a plasmonic nanomaterial, along with most often used silver and gold, there are very few studies on optical properties of the copper containing DLC films [6,9,11]. In the absorbance
ED
spectra of DLC:Cu films deposited by reactive diode sputtering only traces of the plasmonic peaks were observed [11]. In [6,9] optical properties of DLC:Cu films synthesized by electrochemical deposition were studied and plasmonic peaks were observed in the
PT
absorbance spectra of some samples. However, in these studies no data on chemical composition of the films as well as on the structure of the carbon matrix were provided. In
CE
addition, it should be mentioned that electrochemical deposition is rarely used for synthesis of DLC films and it is only suitable for conductive substrates.
AC
It should be mentioned that diamond like carbon (DLC) is amorphous allotrope of carbon consisting of the sp2 bonded (graphite-like) carbon nanoclusters embedded into the sp3 bonded (diamond-like) carbon matrix [16-18]. It may contain from zero up to several tens of atomic percents of hydrogen [16-18]. Usually optical transmittance of DLC in a visible light range as well as hardness and Young’s modulus increase with sp3/sp2 bond ratio. In such a way, hardness of DLC can reach up to 80% of the diamond hardness, while optical transmittance may be comparable with transmittance of the nanocrystalline diamond films [19]. In the present study the influence of composition and structure on optical properties of DLC:Cu films deposited by reactive magnetron sputtering were investigated. Along with direct current (DC) magnetron sputtering, innovative deposition method - high power pulsed magnetron sputtering (HIPIMS) - was applied to deposit the diamond like carbon films. HIPIMS deposition method provides high ion/neutral ratio [20] preferable for formation of 3
ACCEPTED MANUSCRIPT DLC films containing high amount of sp3 bonded carbon [16]. Plasma density in HIPIMS is similar to the case of the cathodic arc deposition and pulsed laser ablation – synthesis methods that are most often used to deposit high sp3/sp2 bond ratio films [20]. Yet main drawbacks of the cathodic arc deposition and pulsed laser ablation (contamination of the film
IP T
by droplets of target and problems with large area deposition) are avoided [20].
CR
2. Experimental techniques
US
In the present study reactive magnetron sputtering was used for deposition of DLC:Cu films. Two modes of operation (direct current (DC) and HIPIMS) were used. Two
MA N
configurations of the magnetron magnetic field (balanced and unbalanced) were applied. In all cases mixture of the hydrocarbons (acetylene) and argon gas was used. The diameter of magnetron was 3″, the copper target was used.
In the case of HIPIMS deposition various pulse currents in 2.4–30 A range were used.
ED
Pulse on time in all cases was set 100 µs, duty cycle 1%, frequency 100 Hz. Base pressure was 5×10-4 Pa and work pressure (4±1)×10-1 Pa was maintained throughout the deposition
PT
process. In the case of DC sputtering, magnetron target current was 0.1 A, base pressure was
5×10-4 Pa and work pressure was (4±1)×10-1 Pa. Thin films were deposited on
CE
monocrystalline silicon and quartz substrates. In all experiments substrate-target gap was set at 0.1 m, the substrates were grounded. Thickness of the deposited films in all cases was in the
AC
range 50-100 nm.
Optical properties of the films were investigated using an optical spectrometer Avantes that is composed of a deuterium halogen light source (AvaLight DHc) and spectrometer (Avaspec-2048). Extinction (complete optical losses due to reflection and absorption of light), absorbance and reflectance of the films were analyzed in the wavelength region from 180nm to 1100nm. Microstructure of DLC:Cu films was studied and linear dimensions of the copper nanoparticles/clusters in DLC matrix were estimated by employing field emission scanning electron microscope (FE-SEM) FEI Quanta 200 FEG in high vacuum mode. Theoretical resolution of the FESEM at 30 kV accelerating voltage is 1.2 nm in high vacuum mode. Chemical composition of the
films was studied by energy-dispersive X-ray spectrometer Bruker Quantax system with XFlash 4030 detector attached on the field emission scanning electron microscope (FEISEM).
4
ACCEPTED MANUSCRIPT Raman scattering measurements were performed using Raman microscope inVia (Renishaw) with 532 nm excitation. Integration time was 100s, power 0.3 mW, grating groove density 2400 grooves/mm. Structure of the crystalline copper nanoclusters was studied by X-ray diffractometer
IP T
D8 Advance (Bruker AXS, Germany). Grazing incidence angle arrangement combined with parallel beam X-ray diffraction geometry was used. Multilayer Ni/graphite parabolic
CR
monochromator was placed in front of the sample. X-ray diffraction patterns were recorded using Cu cathode at 40 kV, anode current 40 mA, scanning step D2Q=0.04° and average time of integration was 15s.
US
The morphology of the surface was analyzed with atomic force microscope NanoWizard®3 (JPK, Germany) working in AC Mode. Silicon probes with a reflective
MA N
backside Al coating (ACTA-10, APPNano, USA) with a resonance frequency of 200-400 kHz and force constant of 13-77 N/m were used. Nominal tip radius was less than 10 nm. The scanning rate of 0.8 Hz was selected.
PT
ED
3. Experimental results
CE
3.1 Deposition conditions effects on target contamination
Target contamination by the products of reactions with reactive gas components is one
AC
of the main problems related to the reactive magnetron sputtering [21]. In the present study during the deposition of DLC:Cu films containing copper (up to 30 at.%) by DC magnetron sputtering resulted in contamination of the sputtering target by carbon film and resultant subsequent drop of the deposition rate during growth of the film. According to our findings this problem can be avoided by using HIPIMS. For DLC:Cu films deposited by reactive HIPIMS such a problem was observed only for the films containing small amounts of copper (up to several atomic percents).
3.2 Structure and composition
Raman scattering spectra of the DLC:Cu films deposited by unbalanced HIPIMS are presented in Fig. 1. In all cases the spectra were typical for diamond like carbon [16-18]. There were no clear differences between the Raman scattering spectra of DLC:Cu films 5
ACCEPTED MANUSCRIPT deposited by both balanced and unbalanced HIPIMS as well as balanced DC magnetron sputtering. In all spectra G peak (stretching vibration mode of sp2 bonded carbon) at ~15001600 cm-1 and shoulder at ~200-1400 cm-1 related to disorder-induced D peak (breathing modes of sp2 bonded carbon rings) [16-18] were observed.
IP T
SEM study revealed nanocomposite microstructure of DLC:Cu films: copper nanoclusters embedded into the diamond like carbon matrix. From the FE-SEM micrographs
CR
of DLC:Cu nanocomposite film (Fig. 2) one can see 10-20nm sized bright features that could be attributed to the copper nanoparticles as the metal gives higher electron contrast then dielectric matrix i.e. DLC.
US
X-ray diffraction (XRD) difractograms of DLC:Cu films are presented in Fig. 3. In the case of DLC:Cu film deposited by balanced HIPIMS and containing 26 at.% of copper,
MA N
Cu(111) XRD peak as well as weak Cu(200) peak can be seen. Particularly the peaks became well expressed in the case of the thicker films (400 nm), but they are seen in the thinner films (100 nm), too. Nanocrystallite size evaluated from Cu(111) peak according to Sherrer equation was 5.1 and 4.1 nm in the case of the thinner and thicker DLC:Cu film respectively.
ED
It should be mentioned that other researchers reported similar results for DLC:Cu films. In [22] for DLC:Cu films deposited by reactive cathodic arc evaporation broad Cu(111) XRD
PT
peak was reported when Cu atomic concentration in the film was ~6.5 at.%. Cu(111) peak along with weak Cu(200) peak was observed when Cu amount reached 10 at.% [22]. In the
CE
case of DLC:Cu films deposited by reactive magnetron sputtering and containing >27 at.% Cu stronger Cu(111) peak and weaker Cu(200) XRD peaks were observed [23]. In [6,9] Cu(111)
AC
XRD peak was reported for DLC:Cu films deposited by electrochemical synthesis. In [11] weak Cu(111) peak was observed in XRD spectra of DLC:Cu films deposited by reactive RF diode sputtering. In [22] average crystallite size of DLC:Cu film deposited by cathodic arc deposition was reported to be 15 nm for the film containing 10 at.% Cu and 3.5 nm for film containing 3 at.% Cu. Such a difference from the results observed in the present study possibly can be explained by peculiarities of the cathodic arc deposition process such as presence of the droplets of Cu target in growing DLC:Cu film resulting in larger average crystallite size. In [23] average crystallite size of DLC:Cu film deposited by radio frequency plasma enhanced reactive magnetron sputtering was 10 nm for the film containing 27.85 at.% Cu (nearly the same amount as in the present study). In this case differences in crystallite size can be explained by differences in deposition process used such as application of methane instead of acetylene in [23] resulting in larger hydrogen contents during films growth as well as use of the additional RF discharge and substrate bias. It should be mentioned that in [24] it
6
ACCEPTED MANUSCRIPT was shown that in the case of DLC:Ag films application of the substrate bias can result in substantially increased average crystallite size.
IP T
3.3 The influence of Cu atomic concentration
In the following section we will analyse properties of the DLC:Cu films versus Cu
CR
atomic concentration. We have found that such a parameter enables to analyse properties of the films deposited by different modes of the operation of magnetron.
US
It can be seen in Fig. 4 that oxygen atomic concentration was in 5-10 at.% range and correlated with the content of copper. It increased with copper atomic concentration for the films deposited by all methods used in the present study (unbalanced HIPIMS, balanced
MA N
HIPIMS, balanced DC magnetron sputtering). Tendency of the decrease of oxygen content in film with the increase of energy flux during the deposition process was observed. The highest oxygen content was found in the films deposited employing low ion/neutral ratio balanced DC magnetron sputtering process. In the case of the high ion/neutral ratio balanced HIPIMS
ED
deposition process, oxygen atomic concentration was lower. The lowest oxygen concentration was observed in the DLC:Cu films deposited by unbalanced HIPIMS process providing both
PT
high ion/neutral ratio and higher ion energy. D/G peak area ratio of DLC:Cu films increased with the Cu atomic concentration (Fig.
CE
5). It should be mentioned that such a result is in good accordance with numerous studies where increased D/G ratio as well as decreased sp 3/sp2 ratio with increase metal concentration
AC
in DLC film was reported for metal containing diamond like carbon films [13,15,25-29]. No clear dependence of D/G ratio on DLC:Cu film deposition method used was observed. Surface root mean square roughness (RRMS) of the DLC:Cu films was low and didn‘t exceed 2 nm (Fig. 6). No clear dependence of RRMS of DLC:Cu films on copper amount or deposition method used was observed. It should be mentioned that surface roughness of „conventional“ hydrogenated DLC films containing no other chemical elements except carbon and hydrogen (see e.g. [30]) is usually at the same level as the lowest surface roughness reported in the present study.
3.4 Optical properties
Optical properties of DLC:Cu films were studied relating them to the Cu concentration and method of deposition. Extinction spectra of DLC:Cu films are presented in Fig. 7. Optical 7
ACCEPTED MANUSCRIPT properties of the films were independent of the deposition method used but clear dependence of the extinction (reflection+absorbance), absorbance and reflectance spectra on copper atomic concentration in the films was observed. It should be noted that some surface plasmon resonance peak was registered in the
IP T
absorption spectrum only when Cu atomic concentration in the film was at least 15 at.% (Fig. 7). Such a behavior is very different from the case of DLC:Ag [24] and DLC:Au [27] films
CR
where surface plasmon resonance peak was detected when silver atomic or gold atomic concentration was several percents (3 at.% or 2 at.% respectively). Surface plasmon resonance peak for DLC:Cu films was in 600-700 nm range. Similar
US
position of the optical absorption plasmonic peak was reported for DLC:Cu films deposited by electrochemical deposition [6,9]. Plasmonic peak redshifted with the increase of Cu atomic
MA N
concentration in the films. It can be mentioned that similar behavior (redshift of plasmonic peak with Ag atomic concentration) was reported for DLC:Ag films [24]. In all cases plasmonic peaks were relatively broad in comparison with ones reported for DLC:Ag and DLC:Au films [24,27,28,29,31]. It is in good accordance with data on
ED
DLC:Cu films deposited by electrochemical deposition [6,9]. Similarly in [3] broad plasmonic peak was reported for the oxidized Cu nanoparticles, while removal of copper
PT
oxide resulted in the increased intensity and increase of the width of Cu plasmonic peak that was comparable with one measured in the case of silver and gold nanoparticles. Taking into
CE
account data on chemical composition of the deposited nanocomposite films presented in this study, it seems that for DLC:Cu films even 5-10 atomic percents of oxygen is enough for
AC
formation of the copper oxide overlayer on the copper nanoclusters and resultant decrease of the surface plasmon resonance peak intensity. It should be mentioned that plasmonic peak can be seen as well in the reflectance spectra of the DLC:Cu films containing >15 at.% Cu. Position of the maximum of this peaks is redshifted in comparison with the absorbance spectra. Tendency of the increased reflectance up to 30-50% with Cu amount should be mentioned. Ratio between the intensities of the plasmonic peak and hydrogenated amorphous carbon (a-C:H) related peak at ~220 nm both for the extinction and absorbance spectra of DLC:Cu films increased with Cu amount linearly in 0-40 atomic percents range (Fig. 8). Particularly, it means that measurement of DLC:Cu film optical extinction or absorbance spectra can be used for rough estimation of copper atomic concentration in the film. On the other hand, plasmonic film could be used as selective optical filter [32]. For these reasons ratio of the plasmonic peak to the background of extinction and absorbance spectra of DLC:Cu films was calculated as a measure of contrast of such potential selective 8
ACCEPTED MANUSCRIPT optical filter. It can be seen that in the investigated range of 0-40 atomic percents of Cu, peak to background ratio increased linearly (Fig. 9). In the case of the optical absorption spectra this ratio was higher in comparison with the case of the extinction spectra. It can be clearly seen in Fig. 9, as well. Thus in real optical applications of the DLC:Cu films, large reflectance
IP T
of these films should be taken into account, not only the problem of the copper oxidation pointed out in [3].
CR
It should be mentioned that oxygen atomic concentration, structure of the amorphous carbon matrix and roughness of the film may have significant influence on the peak to background ratio. According to [3], removal of oxide overlayer from Cu nanoparticles
US
resulted in significant improvement of plasmonic peak to background ratio. In [33] modeling of absorbance spectra of DLC:Ag films revealed that increased sp3/sp2 bond ratio should
MA N
result in the increased peak to background ratio of the plasmonic peak. According to [34] increased surface roughness may result in longer optical way of the light and increased absorption. In addition larger content of the bonded hydrogen in DLC film results in increased optical transparency of the film [35]. According to our results (Fig. 4-6) in some cases
ED
significant differences of the structure of amorphous carbon matrix as well as surface roughness and oxygen atomic concentration were observed for the DLC:Cu films containing
PT
the same amount of copper. However peak to background ratio of DLC:Cu films containing the same content of copper differed no more than 10%. Analysis of the parameters mentioned
CE
above, revealed that observed disparities can not be explained by any separate parameters mentioned above. I.e. this dependence is complex and is influenced by all the discussed
AC
parameters. It should be mentioned as well that to improve plasmonic properties of DLC:Cu films one should use the mode of deposition enabling large amount of sp3 phase and low concentration of oxygen (preferable below 5 at.% ).
Conclusions
Structure and optical properties of diamond like carbon films containing copper (DLC:Cu) deposited by reactive magnetron sputtering using different modes (high power pulsed magnetron sputtering (HIPIMS) and direct current (DC) magnetron sputtering) were studied. It was shown that contamination of cathode taking place during DC magnetron sputtering of DLC:Cu films (containing up to 30 at.% of copper) can be avoided using the HIPIMS deposition mode. For both used modes of the deposition, Raman scattering spectra of the investigated DLC:Cu films were typical for diamond like carbon. XRD patterns included 9
ACCEPTED MANUSCRIPT Cu(111) peak and low intensity Cu(200) peak. In all cases oxygen atomic concentration in the films was in 5-10% range and tendency of the increase of oxygen atomic concentration with the increase of copper atomic concentration was observed. The highest oxygen content was found in the films deposited by low ion/neutral ratio balanced DC magnetron sputtering
IP T
process. Analysis of D/G ratio in the Raman scattering spectra revealed decreased sp 3/sp2 ratio with the increase of Cu atomic concentration in the film. No clear dependence of surface
CR
roughness (RRMS) of DLC:Cu films on copper amount was observed, but optical properties of the films, i.e. the extinction, absorbance and reflectance spectra were dependent on copper atomic concentration in the films as well as on deposition mode. Surface plasmon resonance
US
effect was observed only when Cu atomic concentration in the film was at least 15 at.%. Peak in the absorption spectra due to the surface plasmon resonance was in 600-700 nm range and
MA N
redshifted with the increase of Cu amount. Reflectance of the plasmonic DLC:Cu films was in the 30-50% range. Ratio between the intensities of the plasmonic peak and a-C:H related peak (registered at ~220 nm) in the extinction (and absorbance spectra) as well as ratio of the surface plasmon resonance peak to the background ratio for DLC:Cu films increased with Cu
ED
amount linearly in the investigated 0-40 atomic percent’s range. It was shown that this kind of measurement of optical extinction (or absorbance) spectra of DLC:Cu film can be used for
PT
estimation of copper atomic concentration in the film. On the other hand, large reflectance of
AC
CE
DLC:Cu films should be taken into account in different optical applications.
10
ACCEPTED MANUSCRIPT Acknowledgements
This research was funded by the European Social Fund under the Global Grant
CR
References
IP T
measure (project No VP1-3.1-ŠMM-07-K-03-057).
[1] B. Sharma, R.R. Frontiera, A.-I. Henry, E. Ringe, R.P. Van Duyne, SERS: Materials,
US
applications, and the future, Mater. Today 15 (2012) 16-25.
[2] S.M. Merchant, S. H. Kang, M. Sanganeria, B. van Schravendijk, T. Mountsier, Copper
MA N
Interconnects for Semiconductor Devices, JOM 53 (2001) 43-48. [3] G.H. Chan, J. Zhao, E.M. Hicks, G.C. Schatz, R.P. Van Duyne, Plasmonic Properties of Copper Nanoparticles Fabricated by Nanosphere Lithography, Nano Lett. 7 (2007) 19471952.
ED
[4] C.-C. Chen, F. Chau-Nan Hong, Structure and properties of diamond-like carbon nanocomposite films containing copper nanoparticles, Appl. Surf. Sci. 242 (2005) 261–269
PT
[5] Hua Pang, Xingquan Wang, Guling Zhang, Huan Chen, Guohua Lv, Size Yang, Characterization of diamond-like carbon films by SEM, XRD and Raman spectroscopy, Appl.
CE
Surf. Sci. 256 (2010) 6403–6407. [6] S. Hussain, A.K. Pal, Surface plasmon effect in nanocrystalline copper/DLC composite
AC
filmsby electrodeposition technique, Bull. Mater. Sci., 29 (2006) 553–557. [7] V.I. Ivanov-Omskii, L.K. Panina, S.G. Yastrebov, Amorphous hydrogenated carbon doped with copper as antifungal protective coating, Carbon 38 (2000) 495–499. [8] Meng-Yuan Tsai, Mao-SuanHuang, Li-KaiChen, Yun-DunShen, Ming-Hong Lin, YiChen Chiang, Keng-Liang Ou, Shih-Fu Ou Surface properties of copper-incorporated diamond-like carbon films deposited by hybrid magnetron sputtering, Ceram. Int. 39 (2013) 8335–8340. [9] S Hussain, R K Roy and A K Pal, Surface plasmon effect in nanocrystalline copper/DLC composite films by an electrodeposition technique, J. Phys. D: Appl. Phys. 38 (2005) 900– 908. [10] A.S. Chaus , T.N. Fedosenko, A.V. Rogachev, L. Čaplovič, Surface, microstructure and optical properties of copper-doped diamond-like carbon coating deposited in pulsed cathodic arc plasma, Diam. Relat. Mater. 42 (2014) 64–70.
11
ACCEPTED MANUSCRIPT [11] N. Dwivedi, S. Kumar, H.K. Malik, C.Sreekumar, S. Dayal, C.M.S. Rauthan, O.S.Panwar, Investigation of properties of Cu containing DLC films produced by PECVD process, J. Phys. Chem. of Solids 73 (2012) 308–316. [12] I. Gerhards, C. Ronning, U. Vetter, H. Hofsass, H. Gibhardt, G. Eckold, Q. Li, S.T. Lee,
IP T
Y.L. Huang, M. Seibt, Ion beam synthesis of amorphous carbon thin films containing metallic nanoclusters, Surf. Coat. Technol. 158-159 (2002) 114–119.
CR
[13] Heon Woong Choi, Jung-Hae Choi, Kwang-Ryeol Lee, Jae-Pyoung Ahn, Kyu Hwan Oh, Structure and mechanical properties of Ag-incorporated DLC films prepared by a hybrid ion beam deposition system, Thin Solid Films 516 (2007) 248-251.
US
[14] K. Baba, R. Hatada, S. Flege, W. Ensinger, Preparation and Properties of Ag-Containing
Eng. 2012 (2012) Art. ID 536853.
MA N
Diamond-Like Carbon Films by Magnetron Plasma Source Ion Implantation, Adv. Mater. Sci.
[15] Heon Woong Choi, R.H. Dauskardt, Seung-Cheol Lee, Kwang-Ryeol Lee, Kyu Hwan Oh, Characteristic of silver doped DLC films on surface properties and protein adsorption, Diam. Relat. Mater. 17 (2008) 252-257.
ED
[16] J. Robertson, Diamond Like Amorphous Carbon, Mater. Sci. Eng. R, 37 (2002), pp. 129281.
J. Robertson, Comparison of diamond-like carbon to diamond for applications,
PT
[17]
Phys. Stat. Sol. A, 205 (2008) 2233-2244. A.C. Ferrari, J. Robertson, Raman spectroscopy of amorphous, nanostructured,
CE
[18]
diamond-like carbon, and nanodiamond, Phil. Trans. R. Soc. Lond. A, 362 (2004) 2477–2512.
AC
[19] B. Mednikarov, G. Spasov, Tz. Babeva, J. Pirov, M. Sahatchieva, C. Popova, W.Kulischa, Optical properties of diamond-like carbon and nanocrystalline diamond films, J. Optoelectron. Adv. Mat., 7 (2005) 1407 – 1413. [20] K. Sarakinos, J. Alami, S. Konstantinidis, High power pulsed magnetron sputtering: A review on scientific and engineering state of the art, Surf. Coat. Technol. 204 (2010) 1661– 1684. [21] S. Berg, T. Nyberg, Fundamental understanding and modelling of reactive sputtering of oxides. Thin Solid Films 476 (2005) 215-230. [22] Haiqiang Zhang, Yiming Chen, Bin Liao, Xianying Wu, Huixing Zhang, Xu Zhang, Effect of C2H2 flow rate on microstructure and properties of nc–Cu/a–C:H nanocomposite films prepared by filtered cathodic vaccum arc technique, Nucl. Instrum. Meth. B 307 (2013) 137–142.
12
ACCEPTED MANUSCRIPT [23] Yu-Hao Chan, Chung-Fang Huang, Keng-Liang Ou, Pei-Wen Peng, Mechanical properties and antibacterial activity of copper doped diamond-like carbon films, Surf. Coat. Technol. 206 (2011) 1037–1040. [24] Š. Meškinis, A. Čiegis, A. Vasiliauskas, A. Tamulevičienė, K. Šlapikas, R. Juškėnas, G.
IP T
Niaura, S. Tamulevičius, Plasmonic Properties of Silver Nanoparticles Embedded in Diamond Like Carbon Films: Influence of Structure and Composition, Appl. Surf. Sci. 317 (2014)
CR
1041-1046.
[25] Z. Montiel-González, S.E. Rodil, S. Muhl, A. Mendoza-Galván, L. RodríguezFernández, Amorphous Carbon Gold Nanocomposite Thin Films: Structural and Spectro-
US
ellipsometric Analysis, Thin Solid Films, 519 (2011) 5924–5932. [26] P. Vijai Bharathy, D. Nataraj, Paul K. Chu, Huaiyu Wang, Q. Yang, M.S.R.N. Kiran,
MA N
J. Silvestre-Albero, D. Mangalaraj, Effect of titanium incorporation on the structural, mechanical and biocompatible properties of DLC thin films prepared by reactive-biased target ion beam deposition method, Appl. Surf. Sci. 257 (2010) 143–150. [27] R. Paul, S.R. Bhattacharyya, R. Bhar, A.K. Pal, Modulation of residual stress in diamond
ED
like carbon films with incorporation of nanocrystalline gold, Appl. Surf. Sci. 257 (2011) 10451– 10458.
PT
[28] R. Paul, S. Hussain, A.K. Pal, Characterization of nanocrystalline gold/DLC composite films synthesized by plasma CVD technique, Appl. Surf. Sci. 255 (2009) 8076–8083.
CE
[29] R. Paul, S. Hussain, S. Majumder, S. Varma, A.K. Pal, Surface plasmon characteristics of nanocrystalline gold/DLC composite films prepared by plasma CVD technique, Mater. Sci.
AC
Eng. B 164 (2009) 156–164. [30] J. Robertson. Ultrathin carbon coatings for magnetic storage technology, Thin Solid Films 383 (2001) 81-88. [31] S. Hussain, R.K. Roy, A.K. Pal, Incorporation of silver nanoparticles in DLC matrix and surface plasmon resonance effect, Mater. Chem. Phys. 99 (2006) 375–381. [32] G. Kedawat, B.K. Gupta, P. Kumar, J. Dwivedi, A. Kumar, N.K. Agrawal, S.S. Kumar, Y.K. Vijay, Fabrication of a Flexible UV Band-Pass Filter Using Surface Plasmon Metal−Polymer Nanocomposite Films for Promising Laser Applications, Appl. Mater. Interfaces 6 (2014) 8407−8414. [33] H. Zoubos, L.E.Koutsokeras, D.F.Anagnostopoulos, E.Lidorikis, S.A.Kalogirou, A.R.Wildes, P.C.Kelires, P.Patsalas, Broadband optical absorption of amorphous carbon/Ag nanocomposite films and its potential for solar harvesting applications, Sol. Energ. Mat. Sol. C. 117 (2013) 350–356.
13
ACCEPTED MANUSCRIPT [34] S. Mokkapati, K. R. Catchpole, Nanophotonic light trapping in solar cells, J. Appl. Phys. 112 (2012) 101101. [35] C. Casiraghi, F. Piazza, A.C. Ferrari, D. Grambole, J. Robertson, Bonding in hydrogenated diamond-like carbon by Raman spectroscopy, Diam. Relat. Mater. 14 (2005)
AC
CE
PT
ED
MA N
US
CR
IP T
1098– 1102.
14
ACCEPTED MANUSCRIPT List of the captures
Fig. 1. Typical Raman scattering spectra of DLC:Cu films (all films presented were deposited
IP T
by unbalanced HIPIMS).
CR
Fig. 2. Typical SEM microgram of DLC:Cu film. The film was deposited by unbalanced
US
HIPIMS and contains 34 at.% Cu.
Fig. 3. X-ray diffraction patterns of DLC:Cu films deposited by balanced HIPIMS and
MA N
containing 26 at.% Cu.
ED
Fig. 4. Oxygen atomic concentration Vs Cu atomic concentration in DLC:Cu films.
PT
Fig. 5. D/G ratio peak area ratio Vs Cu atomic concentration.
CE
Fig. 6. Surface roughness of DLC:Cu films.
AC
Fig. 7. Typical optical extinction (absorbance+reflectance) (a), absorbance (b) and reflectance (c) spectra of DLC:Cu films containing different amounts of copper.
Fig. 8. Dependence of intensity of the plasmonic peak normalized to a-C:H absorption peak intensity Vs Cu atomic concentration in DLC:Cu film. Ratio was calculated for the extinction (a) and absorbance (b) spectra.
Fig. 9. Ratio of the plasmonic peak to background of extinction (a) and absorbance (b) spectra of DLC:Cu films Vs Cu atomic concentration in DLC:Cu films.
15
AC
CE
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
Figure 1
16
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
AC
CE
PT
ED
Figure 2
17
CE
AC
Figure 3
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
18
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 4
19
AC
CE
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
Figure 5
20
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 6
21
AC
CE
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
Figure 7a
22
AC
CE
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
Figure 7b
23
CE
AC
Figure 7c
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
24
CE
AC
Figure 8a
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
25
CE
AC
Figure 8b
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
26
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
AC
CE
PT
Figure 9a
27
AC
CE
Figure 9b
PT
ED
MA N
US
CR
IP T
ACCEPTED MANUSCRIPT
28
ACCEPTED MANUSCRIPT Highlights Optical properties of diamond like carbon (DLC:Cu) films containing Cu were
IP T
studied.
CR
Reactive high power pulsed magnetron sputtering was used for deposition.
US
Extinction, absorbance and reflectance spectra of DLC:Cu films depend on Cu content.
AC
CE
PT
ED
MA N
Plasmonic peak intensity and position of the peak depend on Cu amount.
29