- Email: [email protected]
Structural modifications of diamond like carbon films induced by MeV nitrogen ion irradiation
Applied Surface Science 255 (2009) 4796–4800
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Structural modifications of diamond like carbon films induced by MeV nitrogen ion irradiation S. Mathew a,1, U.M. Bhatta a, A.K.M. Maidul Islam b, M. Mukherjee b, N.R. Ray c, B.N. Dev a,d,* a
Institute of Physics, Sachivalaya Marg, Bhubaneswar 751005, India Surface Physics Division, Saha Institute of Nuclear Physics 1/AF, Bidhan Nagar, Kolkata 700064, India c Plasma Physics Division, Saha Institute of Nuclear Physics 1/AF, Bidhan Nagar, Kolkata 700064, India d Indian Association for the Cultivation of Science, 2A & 2B Raja S. C. Mullick Road, Jadavpur, Kolkota 700032, India b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 17 July 2008 Received in revised form 23 November 2008 Accepted 23 November 2008 Available online 30 November 2008
Diamond-like carbon (DLC) films were deposited on Si(1 0 0) substrates using plasma deposition technique. The deposited films were irradiated using 2 MeV N+ ions at fluences of 1 1014 , 1 1015 and 5 1015 ions/cm2. Samples have been characterized by using Raman spectroscopy, X-ray photoelectron spectroscopy (XPS) and high-resolution transmission electron microscopy (HRTEM). Analysis of Raman spectra shows a gradual shift of both D and G band peaks towards higher frequencies along with an increase of the intensity ratio, I(D)/I(G), with increasing ion fluence in irradiation. These results are consistent with an increase of sp2 bonding. XPS results also show a monotonic increase of sp2/sp3 hybridization ratio with increasing ion fluence. Plan view TEM images show the formation of clusters in the irradiated DLC films. HRTEM micrographs from the samples irradiated at a fluence of 5 1015 ions/ cm2 show the lattice image with an average interplanar spacing of 0.34 nm, revealing that the clusters are graphite clusters. The crystallographic planes in these clusters are somewhat distorted compared to the perfect graphite structure. ß 2009 Published by Elsevier B.V.
PACS: 81.05.Uw 78.70 g 82.80.Pv; 78.30. j Keywords: Diamond Interactions of particles and radiation with matter X-ray photoelectron spectroscopy (XPS) Raman spectroscopy
1. Introduction Diamond-like carbon (DLC) films are well known for their excellent properties, such as, low frictional coefficient, high hardness, chemical inertness, biocompatibility, optical transparency, etc. [1]. The unique combination of the above properties provides substantial potential for applications of DLC films in electromechanical systems, biomedical field, flat-panel displays, particle detectors, etc. [2–4]. DLC films are extensively used as protective coatings in both magnetic and optical data storage devices [4,5]. Ion beams have been used to modify the optical, electrical and structural properties of DLC films [6–8]. Baptista et al. reported the formation of sp2-bonded hard amorphous carbon network in 400 keV N+-irradiated amorphous carbon and fullerene films [9]. Ion irradiation using various ions of tens of keV
* Corresponding author. Tel: +91 33 2473 4971x200; fax: +91 33 2473 2805. E-mail addresses: [email protected], [email protected] (B.N. Dev). 1 Present address: Department of Electrical and Computer Engineering, National University of Singapore, 117576, Singapore. 0169-4332/$ – see front matter ß 2009 Published by Elsevier B.V. doi:10.1016/j.apsusc.2008.11.061
energy has been found to reduce the intrinsic stress present in DLC films [10]. The growth of microcrystalline graphite in DLC films following 85 MeV Ni ion irradiation has been reported [11]. Since DLC films find applications in electronic and electro-mechanical systems the behavior of these films under extreme conditions of heat, radiation, etc. is important for the reactor based and space applications of these devices. The existence of carbon in sp, sp2 and sp3 hybridization with the possibility to obtain systems having different percentages of carbon bonding, makes ion irradiation of various carbon allotropes interesting for both basic as well as applied fields of research [12]. Ion irradiation introduces a wide range of defects in a controlled manner and is used to tailor the structural and even the magnetic properties of materials. Recent reports of magnetism in ionirradiated carbon systems have stimulated renewed interest in ion irradiation of various carbon allotropes [14–16]. Changes in the structure and hybridization cause changes in physical, electrical and other properties of DLC films. In order to investigate the ion beam induced changes in the structure and hybridization in DLC films, we have performed ion irradiation experiments as a function of the fluence of 2 MeV N+ ions. The
S. Mathew et al. / Applied Surface Science 255 (2009) 4796–4800
projected range of 2 MeV nitrogen ions is more than the thickness of the DLC film and the modifications of structure and hybridization are mainly due to the defects produced during the passage of ions through the film. The ions are eventually buried in the underlying substrate. Using high-resolution transmission electron microscopy (HRTEM) and Raman spectroscopy we have investigated the structural changes in our DLC films. X-ray photoelectron spectroscopy (XPS) has been employed for understanding the nature and strength of the carbon bonding.
4797
Table 1 The D, G peak positions and I(D)/I(G) ratios for the pristine and 2 MeV N+-irradiated samples Irradiation fluence (ions cm 0 (pristine) 1 1014 1 1015 5 1015
2
)
D (cm 1320 1364 1404 1415
1
)
G (cm 1539 1538 1553 1564
1
)
I(D)/I(G) 0.44 0.66 1.40 2.0
The Raman spectrum of DLC films consists of a broad skewed peak centered at about 1500–1550 cm 1[1]. The Raman spectra from the pristine DLC film and the films irradiated at fluences of 1 1014 , 1 1015 and 5 1015 ions/cm2(hereafter samples A, B, C
and D, respectively) are shown in Fig. 1(a)–(d). The spectra are fitted using two Lorentzians and a linear background as shown in Fig. 1. The component in the low wave number region is identified as the D mode and the one at higher wave number region is the G mode. The G peak is due to bond stretching of all pairs of sp2 atoms in both carbon rings and chains. The D peak is caused by the breathing modes of sp2 atoms in the aromatic sixfold carbon rings. In the pristine sample the D and the G peaks are found to be at 1320 and 1539 cm 1, respectively. In the case of purely sp2 hybridized systems, such as graphite, the G peak is expected to be at 1580 cm 1. But in the case of DLC films the position of the G peak is found to shift to a lower wave number depending on the amount of tetrahedral bonding [13]. It can be seen that the D and G peaks are shifted to higher wave numbers in the irradiated samples [Fig. 1(a)–(d)]. The position of D and G peaks and the ratio of the intensities of D peak to G peak (I(D)/I(G)) are given in Table 1. The large width of the peaks can be associated to distortions of sp2 bond angle [9]. The I(D)/I(G) ratio in the pristine film (sample A) is 0.44 and it increases to 2.0 in the case of sample D. The Raman frequencies of idealized five-, six- and seven-membered p-bonded carbon rings have been calculated theoretically by Doyle and Jamson and the D mode is estimated to be at 1444 cm 1 in the case of fivefold rings [17]. Since five- and seven-membered ring structures lack inversion symmetry, the vibrational modes with A symmetry dominate, i.e., an increase in the intensity of D mode is expected [17,18]. The shifting of the G peak to higher wave number with increasing ion fluence indicates an increase of sp2 hybridization in the irradiated samples. The D mode is at 1320 cm 1 in the pristine sample. It has shifted to 1415 cm 1 in the sample irradiated at a fluence of 5 1015 ions/cm2. This shifting of D mode is an indication of the presence of non-six-membered ring structures in the sample [18]. All the features in the Raman spectra, namely the shifting of the D and G peaks along with the increase of I(D)/I(G) ratio point towards an increase of sp2 bonding and
Fig. 1. Raman spectrum of (a) pristine DLC/Si(1 0 0) sample, and samples irradiated at a fluence of (b) 1 1014 ions/cm2, (c) 1 1015 ions/cm2 and (d) 5 1015 ions/cm2.
Fig. 2. XPS spectra from (a) a pristine DLC/Si(1 0 0) sample, and samples irradiated at a fluence of (b) 1 1014 ions/cm2, (c) 1 1015 ions/cm2 and (d) 5 1015 ions/cm2.
2. Experiment Diamond-like carbon thin films were deposited on mirror polished Si(1 0 0) substrates at room temperature using asymmetric capacitively coupled rf (13.56 MHz) plasma system utilizing pure hydrogen (H2), helium (He) and methane (CH4) gases. The thickness of the films were estimated to be 100 nm. Ion irradiations were carried out using the 3 MV 9SDH2 tandem Pelletron accelerator facility in our institute. The samples were irradiated uniformly by rastering the ion beam on the sample. The ion fluence used for irradiation were 1 1014 , 1 1015 and 5 1015 ions/cm2. The beam current during irradiation was kept at 20 nA. The pressure in the irradiation chamber was 1 10 6 mbar. Raman spectra were recorded at room temperature in the back-scattering geometry using photons of 514.5 nm wavelength as exciting radiation, U1000 monochromator and a CCD detector. XPS measurements were made using Omicron Multiprobe spectrometer fitted with an EA 125 hemispherical analyzer. A monochromatized Al Ka X-ray source operating at a base pressure of 2 10 10 mbar was used for the measurements. Plan view TEM measurements were carried out using 200 kV (JEOL 2010) HRTEM using GATAN TV camera for data acquisition with point to point resolution of 0.19 nm and lattice resolution of 0.14 nm. The sample preparation for TEM was carried out by using an ultrasonic disc cutter, a laping and polishing system, a dimple grinder and a precision ion polishing system for the final thinning of the sample. 3. Results and discussions
4798
S. Mathew et al. / Applied Surface Science 255 (2009) 4796–4800
Fig. 3. XPS spectra of N 1s and F 1s electrons extracted from the spectra in Fig. 2. (a), (b), (c) and (d) correspond to those in Fig. 2.
graphite clustering in the irradiated samples. To estimate the nature of hybridization we have carried out XPS measurements and the irradiation induced graphite clustering is investigated using HRTEM. These results are discussed below. XPS spectra of the pristine and the irradiated samples are shown in Fig. 2. Apart from the C 1s peak, the pristine sample shows the presence of N 1s (not visible on this scale), O 1s and F 1s peaks. These peaks are also seen in the irradiated samples with varying intensities. The core level spectra of N 1s and F 1s are shown in Fig. 3. The C 1s spectra are shown in Fig. 4. In the pristine sample the ratio of intensity of F 1s to C 1s is 0.3 and N 1s to C 1s is 0.06. With irradiation, the intensities of F 1s and N 1s are found to be decreasing and in sample D both these peaks are absent. In sample D the F 1s peak is absent, while the N 1s peak has nearly vanished (Fig. 3). The fitting of the spectra in Fig. 4 is done by a x2 iteration programme using a convolution of Lorentzian–Gaussian function with a Sherly background [19]. The core level spectra of C 1s in all the samples were fitted with components positioned
around 284.3 and 285.0, 286.2 and 288.8 eV. The peaks observed at 284.3 and 285.0 eV correspond to sp2 and sp3 hybridized carbon atoms, respectively [20]. The peaks at 286.2 and 288.8 eV can be due to C–O and C O, respectively [21,22]. The above two peaks observed in the higher binding energy side of C 1s spectrum can also be due to CN and CF bondings, i.e., CN can give a peak at 286.2 eV and CF at 288.8 eV [22,23]. The intensity of the peak at 288.8 eV in the pristine sample is 2% of the total intensity of C 1s peak and is barely visible in samples B and C. In sample D this peak has an intensity of 7% of the total C 1s peak intensity and is clearly visible. The intensities of F 1s/C 1s and N 1s/C 1s have reduced with ion fluence and in sample D the fluorine and the nitrogen peaks were absent [Fig. 3(a)–(d)]. In C 1s spectra, the intensities of the 286.2 and 288.8 eV peaks have not reduced with increasing ion fluence. This indicates that F and N are surface impurities and the peaks at 286.2 and 288.8 eV were predominantly due to C–O and C O bond, respectively. In the pristine sample the intensity ratio of O 1s/C 1s is 0.4 and it has became 0.6 in sample D (in earlier XPS
Fig. 4. XPS spectra of C 1s electrons extracted from the spectra in Fig. 2: (a) pristine DLC/Si(1 0 0) sample, and samples irradiated at a fluence of (b) 1 1014 ions/cm2, (c) 1 1015 ions/cm2 and (d) 5 1015 ions/cm2.
Fig. 5. The variation of sp2/sp3 ratio of pristine and irradiated DLC films estimated from the spectra in Fig. 4.
S. Mathew et al. / Applied Surface Science 255 (2009) 4796–4800
experiments, this trend has also been observed for halogenadsorbed Si surfaces, even without ion irradiation. For example, F desorption from Si–F bonds leads to silicon oxide formation [25]). We do not observe a noticeable shift of binding energies of C 1s peak position after irradiation. The binding energy positions and FWHMs of all the constituent peaks were kept constant for fitting the spectra of the pristine and the irradiated samples as shown in Fig. 4. The line shape of the XPS spectrum gives information about the chemical bonding environments in the sample [20,24]. The ratio of the intensities of sp2 to sp3 (sp2/sp3) in the pristine sample is 0.93 and it increases to 1.33 in sample D. The variation of sp2/sp3 ratio is plotted in Fig. 5. Both XPS and Raman spectroscopy measurements clearly indicate an increase of trigonal bonding with the ion fluence. Our Raman spectroscopy results show the features, based on which earlier workers speculated the formation of graphite clusters in the ion-irradiated DLC films [11]. We have used HRTEM measurements for direct verification of the presence of graphite clusters in our irradiated samples. Transmission electron micrographs of the pristine sample (sample A) and the sample irradiated at a fluence of 5 1015 ions/cm2(sample D) are shown in Fig. 6(a) and (b), respectively. The appearance of nanometer (nm) sized ( 2 nm) clusters are clearly visible in the irradiated sample. A HRTEM image of the irradiated sample (D) is shown in Fig. 7. From HRTEM, the observed lattice spacing of 0.34 nm in clusters is found to be that of graphite. A filtered image [using a fast Fourier transform (FFT) technique] from a single cluster is shown in the inset of Fig. 7. This clearly shows that MeV nitrogen irradiation induces the formation of graphite clusters in DLC films. Very recently Choi et al. observed the formation graphite clusters in DLC films after rapid thermal annealing at 900 C in nitrogen ambient [26]. Talapatra et al. observed the formation of nm sized graphite
4799
Fig. 7. A HRTEM image of an irradiated DLC film — the same sample as in Fig. 6(b). A FFT filtered image of one of the observed graphite clusters is shown in the inset.
clusters in nanodiamonds after keV N and C ion irradiation [27] and also in thermally annealed samples [28]. sp2 clustering in tetrahedral amorphous carbon samples induced by thermal annealing have also been reported [29]. In the present study of MeV N+ irradiation of DLC films the N+ ions pass through the DLC film and get buried into the substrate silicon; the increase of trigonal bonding in the DLC films can be mainly due to the defects produced during ion irradiation. The nuclear and the electronic energy loss of 2 MeV N+ ions in an amorphous carbon target with the density of DLC is estimated (using SRIM 2003 [30]) to be 1.354 keV/nm and 0.648 keV/nm, respectively. The effect of ion irradiation on DLC films can be related to the film microstructure. Ion irradiation of any solid induces the breakage of bonds, creation of interstitials and vacancies, and also loss of hydrogen in the case of DLC films [11]. An estimate about the amount of hydrogen in the DLC samples can be obtained from the photoluminescence (PL) background in the first order Raman spectra. The ratio of the slope of the fitted linear background of the Raman spectrum to the intensity of the G peak can be used as a measure of the bonded H content in the film [31]. The amount of hydrogen content in the pristine film is 40% and the Raman spectra analysis did not show a significant loss of hydrogen in our irradiated samples. Amorphous carbons with the same sp2/sp3 ratio and H content are reported to show different optical, mechanical and electrical properties according to the clustering of sp2 phase [13]. The recent experiments had demonstrated room temperature ferromagnetism in heat-treated and ion-irradiated (100 keV C or N) nanodiamond samples [15,28]. In both of these cases, graphite nanoclusters were formed either by heat treatment or ion irradiation [27,28], and the authors attribute the observed ferromagnetism to the presence of graphite nanoclusters. In our case, from the formation of graphite nanoclusters along with a mixed sp2–sp3 hybridization in the ion irradiated DLC samples, one may speculate the possibility of magnetic ordering in these irradiated DLC films. 4. Conclusions
Fig. 6. Transmission electron micrograph of (a) a pristine DLC film and (b) a DLC film irradiated at a fluence of 5 1015 ions/cm2.
MeV N+ irradiation on DLC films leads to an increase of sp2/sp3 hybridization ratio. With increasing ion fluence the sp2/sp3 ratio is found to be increasing and samples irradiated at a fluence of 5 1015 ions/cm2 show a 40% increase in comparison with the pristine sample. Raman spectroscopy results, namely, the shifting of the positions of the D and the G peaks, along with the increase of the I(D)/I(G) ratio, are also an indication of an increase of sp2 bonding in the ion irradiated DLC samples. TEM images from the
4800
S. Mathew et al. / Applied Surface Science 255 (2009) 4796–4800
irradiated samples show the formation of nanometer sized clusters. HRTEM results reveal the details within the clusters and show that the clusters are indeed graphite clusters, consistent with the increase of sp2 bonding. From the observed formation of graphite nanoclusters, along with a mixed sp2–sp3 hybridization in the ion-irradiated DLC films, one is tempted to speculate about the possibility of magnetic ordering in the irradiated DLC films. Further experiments such as the quantification of hydrogen content in the film, conductivity and hardness measurements, investigation of magnetic ordering, etc. can provide further insights into the basic understanding and the technological applications of these ionirradiated DLC films.
[8] [9] [10] [11]
Acknowledgment
[19] [20]
We acknowledge Prof. S.N. Sahu and Mr. S.N. Sarangi for Raman spectroscopy measurements.
[21] [22]
References
[23] [24] [25]
[1] J. Robertson, Mater. Sci. Eng. R 37 (2002) 129. [2] R. Hauert, Diamond. Relat. Mater. 12 (2003) 583. [3] N.R. Ray (Ed.). Applications of Nanocrystalline Diamond and Diamond Like Carbon materials, Proceedings of IWAncDLC-2006 (Saha Institute of Nuclear Physics, Kolkata, 2006). [4] A. Erdemir, C. Donnet, J. Phys. D 39 (2006) R311. [5] C. Casiraghi, J. Robertson, A.C. Ferrari, Mater. Today 10 (2007) 44. [6] M.S. Dresselhaus, R. Kalish, Ion Implantation in Diamond, Graphite and Related Materials, Chapter 9, Springer Series in Materials Science 22, Berlin, 1992. [7] V.D. Vankar, N. Dilawar, Vacuum 47 (1996) 1275.
[12] [13] [14] [15] [16] [17] [18]
[26] [27] [28] [29] [30] [31]
D.K. Avasti, Vacuum 47 (1996) 165. D.L. Baptista, F.C. Zawislak, Diamond. Relat. Mater. 13 (2004) 1791. D.H. Lee, S. Fayeulle, K.C. Walter, M. Nastasi, Nucl. Instr. Meth. B 148 (1999) 216. H.C. Barshilia, S. Sah, B.R. Mehta, V.D. Vankar, D.K. Avasthi, G.K. Jaipal, Mehta, Thin Solid Films 258 (1995) 123. G. Compagnini, S. Battiato, O. Puglisi, G.A. Baratta, G. Strazzulla, Carbon 43 (2005) 3025. A.C. Ferrari, J. Robertson, Phys. Rev. B. 61 (2000) 14095. S. Mathew, B. Satpati, B. Joseph, B.N. Dev, R. Nirmala, S.K. Mallik, R. Kesavamoorthy, Phys. Rev. B 75 (2007) 75426. S. Talapatra, P.G. Ganesan, T. Kim, T.R. Vajtai, H. Huang, M. Shima, G. Ramanath, D. Srivastava, S.C. Deevi, P.M. Ajayan, Phys. Rev. Lett. 95 (2005) 97201. P. Esquinazi, D. Spemann, R. Hhne, A. Setzer, K.H. Han, T. Butz, Phys. Rev. Lett. 91 (2003) 227201. T.E. Doyle, J.R. Dennison, Phys. Rev. B. 51 (1995) 196. M.P. Siegal, D.R. Tallant, L.J. Martinez-Miranda, J.C. Barbour, R.L. Simpson, D.L. Overmyer, Phys. Rev. B 61 (2000) 10451. D.A. Shirly, Phys. Rev. 55 (1972) 4709. J.F. Morar, F.J. Himpsel, G. Hollinger, J.L. Jordan, G. Hughes, F.R. McFeely, Phys. Rev. B 33 (1986) 1340. Y. Taki, O. Takai, Thin Solid Films 316 (1998) 45. K. Yamamoto, Y. Koga, S. Fujiwara, F. Kokai, J.I. Kleiman, K.K. Kim, Thin Solid Films 339 (1999) 38. A. Tressaud, T. Shirasaki, G. Nans, E. Papirer, Carbon 40 (2002) 217. R. Haerle, E. Riedo, A. Pasquarello, A. Baldereschi, Phys. Rev. B 65 (2001) 45101. K. Sekar, G. Kuri, D.P. Mahapatra, B.N. Dev, J.V. Ramana, S. Kumar, V.S. Raju, Surf. Sci. 302 (1994) 25. W.S. Choi, B. Hong, Renew. Energ. 33 (2008) 226. S. Talapatra, J.-Y. Cheng, N. Chakrapani, S. Trasobares, A. Cao, R. Vajtai, M.B. Huang, P.M. Ajayan, Nanotechnology 17 (2006) 305. S. Talapatra, T. Kim, B.Q. Wei, S. Kar, R. Vajtai, G.V.S. Sastry, M. Shima, D. Srivastava, P.M. Ajayan, Nanopage 1 (2006) 315. J.O. Orwa, I. Andrienko, J.L. Peng, S. Prawer, Y.B. Zhang, S.P. Lau, J. Appl. Phys. 96 (2004) 6286. J.F. Ziegler, J.P. Biersack, U. Littmark, SRIM 2003—a version of the TRIM program. B. Marchon, J. Gui, J.K. Grannen, G.C. Rauch, IEEE Trans. Magn. 33 (1997) 3148.
Contents lists available at ScienceDirect
Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc
Structural modifications of diamond like carbon films induced by MeV nitrogen ion irradiation S. Mathew a,1, U.M. Bhatta a, A.K.M. Maidul Islam b, M. Mukherjee b, N.R. Ray c, B.N. Dev a,d,* a
Institute of Physics, Sachivalaya Marg, Bhubaneswar 751005, India Surface Physics Division, Saha Institute of Nuclear Physics 1/AF, Bidhan Nagar, Kolkata 700064, India c Plasma Physics Division, Saha Institute of Nuclear Physics 1/AF, Bidhan Nagar, Kolkata 700064, India d Indian Association for the Cultivation of Science, 2A & 2B Raja S. C. Mullick Road, Jadavpur, Kolkota 700032, India b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 17 July 2008 Received in revised form 23 November 2008 Accepted 23 November 2008 Available online 30 November 2008
Diamond-like carbon (DLC) films were deposited on Si(1 0 0) substrates using plasma deposition technique. The deposited films were irradiated using 2 MeV N+ ions at fluences of 1 1014 , 1 1015 and 5 1015 ions/cm2. Samples have been characterized by using Raman spectroscopy, X-ray photoelectron spectroscopy (XPS) and high-resolution transmission electron microscopy (HRTEM). Analysis of Raman spectra shows a gradual shift of both D and G band peaks towards higher frequencies along with an increase of the intensity ratio, I(D)/I(G), with increasing ion fluence in irradiation. These results are consistent with an increase of sp2 bonding. XPS results also show a monotonic increase of sp2/sp3 hybridization ratio with increasing ion fluence. Plan view TEM images show the formation of clusters in the irradiated DLC films. HRTEM micrographs from the samples irradiated at a fluence of 5 1015 ions/ cm2 show the lattice image with an average interplanar spacing of 0.34 nm, revealing that the clusters are graphite clusters. The crystallographic planes in these clusters are somewhat distorted compared to the perfect graphite structure. ß 2009 Published by Elsevier B.V.
PACS: 81.05.Uw 78.70 g 82.80.Pv; 78.30. j Keywords: Diamond Interactions of particles and radiation with matter X-ray photoelectron spectroscopy (XPS) Raman spectroscopy
1. Introduction Diamond-like carbon (DLC) films are well known for their excellent properties, such as, low frictional coefficient, high hardness, chemical inertness, biocompatibility, optical transparency, etc. [1]. The unique combination of the above properties provides substantial potential for applications of DLC films in electromechanical systems, biomedical field, flat-panel displays, particle detectors, etc. [2–4]. DLC films are extensively used as protective coatings in both magnetic and optical data storage devices [4,5]. Ion beams have been used to modify the optical, electrical and structural properties of DLC films [6–8]. Baptista et al. reported the formation of sp2-bonded hard amorphous carbon network in 400 keV N+-irradiated amorphous carbon and fullerene films [9]. Ion irradiation using various ions of tens of keV
* Corresponding author. Tel: +91 33 2473 4971x200; fax: +91 33 2473 2805. E-mail addresses: [email protected], [email protected] (B.N. Dev). 1 Present address: Department of Electrical and Computer Engineering, National University of Singapore, 117576, Singapore. 0169-4332/$ – see front matter ß 2009 Published by Elsevier B.V. doi:10.1016/j.apsusc.2008.11.061
energy has been found to reduce the intrinsic stress present in DLC films [10]. The growth of microcrystalline graphite in DLC films following 85 MeV Ni ion irradiation has been reported [11]. Since DLC films find applications in electronic and electro-mechanical systems the behavior of these films under extreme conditions of heat, radiation, etc. is important for the reactor based and space applications of these devices. The existence of carbon in sp, sp2 and sp3 hybridization with the possibility to obtain systems having different percentages of carbon bonding, makes ion irradiation of various carbon allotropes interesting for both basic as well as applied fields of research [12]. Ion irradiation introduces a wide range of defects in a controlled manner and is used to tailor the structural and even the magnetic properties of materials. Recent reports of magnetism in ionirradiated carbon systems have stimulated renewed interest in ion irradiation of various carbon allotropes [14–16]. Changes in the structure and hybridization cause changes in physical, electrical and other properties of DLC films. In order to investigate the ion beam induced changes in the structure and hybridization in DLC films, we have performed ion irradiation experiments as a function of the fluence of 2 MeV N+ ions. The
S. Mathew et al. / Applied Surface Science 255 (2009) 4796–4800
projected range of 2 MeV nitrogen ions is more than the thickness of the DLC film and the modifications of structure and hybridization are mainly due to the defects produced during the passage of ions through the film. The ions are eventually buried in the underlying substrate. Using high-resolution transmission electron microscopy (HRTEM) and Raman spectroscopy we have investigated the structural changes in our DLC films. X-ray photoelectron spectroscopy (XPS) has been employed for understanding the nature and strength of the carbon bonding.
4797
Table 1 The D, G peak positions and I(D)/I(G) ratios for the pristine and 2 MeV N+-irradiated samples Irradiation fluence (ions cm 0 (pristine) 1 1014 1 1015 5 1015
2
)
D (cm 1320 1364 1404 1415
1
)
G (cm 1539 1538 1553 1564
1
)
I(D)/I(G) 0.44 0.66 1.40 2.0
The Raman spectrum of DLC films consists of a broad skewed peak centered at about 1500–1550 cm 1[1]. The Raman spectra from the pristine DLC film and the films irradiated at fluences of 1 1014 , 1 1015 and 5 1015 ions/cm2(hereafter samples A, B, C
and D, respectively) are shown in Fig. 1(a)–(d). The spectra are fitted using two Lorentzians and a linear background as shown in Fig. 1. The component in the low wave number region is identified as the D mode and the one at higher wave number region is the G mode. The G peak is due to bond stretching of all pairs of sp2 atoms in both carbon rings and chains. The D peak is caused by the breathing modes of sp2 atoms in the aromatic sixfold carbon rings. In the pristine sample the D and the G peaks are found to be at 1320 and 1539 cm 1, respectively. In the case of purely sp2 hybridized systems, such as graphite, the G peak is expected to be at 1580 cm 1. But in the case of DLC films the position of the G peak is found to shift to a lower wave number depending on the amount of tetrahedral bonding [13]. It can be seen that the D and G peaks are shifted to higher wave numbers in the irradiated samples [Fig. 1(a)–(d)]. The position of D and G peaks and the ratio of the intensities of D peak to G peak (I(D)/I(G)) are given in Table 1. The large width of the peaks can be associated to distortions of sp2 bond angle [9]. The I(D)/I(G) ratio in the pristine film (sample A) is 0.44 and it increases to 2.0 in the case of sample D. The Raman frequencies of idealized five-, six- and seven-membered p-bonded carbon rings have been calculated theoretically by Doyle and Jamson and the D mode is estimated to be at 1444 cm 1 in the case of fivefold rings [17]. Since five- and seven-membered ring structures lack inversion symmetry, the vibrational modes with A symmetry dominate, i.e., an increase in the intensity of D mode is expected [17,18]. The shifting of the G peak to higher wave number with increasing ion fluence indicates an increase of sp2 hybridization in the irradiated samples. The D mode is at 1320 cm 1 in the pristine sample. It has shifted to 1415 cm 1 in the sample irradiated at a fluence of 5 1015 ions/cm2. This shifting of D mode is an indication of the presence of non-six-membered ring structures in the sample [18]. All the features in the Raman spectra, namely the shifting of the D and G peaks along with the increase of I(D)/I(G) ratio point towards an increase of sp2 bonding and
Fig. 1. Raman spectrum of (a) pristine DLC/Si(1 0 0) sample, and samples irradiated at a fluence of (b) 1 1014 ions/cm2, (c) 1 1015 ions/cm2 and (d) 5 1015 ions/cm2.
Fig. 2. XPS spectra from (a) a pristine DLC/Si(1 0 0) sample, and samples irradiated at a fluence of (b) 1 1014 ions/cm2, (c) 1 1015 ions/cm2 and (d) 5 1015 ions/cm2.
2. Experiment Diamond-like carbon thin films were deposited on mirror polished Si(1 0 0) substrates at room temperature using asymmetric capacitively coupled rf (13.56 MHz) plasma system utilizing pure hydrogen (H2), helium (He) and methane (CH4) gases. The thickness of the films were estimated to be 100 nm. Ion irradiations were carried out using the 3 MV 9SDH2 tandem Pelletron accelerator facility in our institute. The samples were irradiated uniformly by rastering the ion beam on the sample. The ion fluence used for irradiation were 1 1014 , 1 1015 and 5 1015 ions/cm2. The beam current during irradiation was kept at 20 nA. The pressure in the irradiation chamber was 1 10 6 mbar. Raman spectra were recorded at room temperature in the back-scattering geometry using photons of 514.5 nm wavelength as exciting radiation, U1000 monochromator and a CCD detector. XPS measurements were made using Omicron Multiprobe spectrometer fitted with an EA 125 hemispherical analyzer. A monochromatized Al Ka X-ray source operating at a base pressure of 2 10 10 mbar was used for the measurements. Plan view TEM measurements were carried out using 200 kV (JEOL 2010) HRTEM using GATAN TV camera for data acquisition with point to point resolution of 0.19 nm and lattice resolution of 0.14 nm. The sample preparation for TEM was carried out by using an ultrasonic disc cutter, a laping and polishing system, a dimple grinder and a precision ion polishing system for the final thinning of the sample. 3. Results and discussions
4798
S. Mathew et al. / Applied Surface Science 255 (2009) 4796–4800
Fig. 3. XPS spectra of N 1s and F 1s electrons extracted from the spectra in Fig. 2. (a), (b), (c) and (d) correspond to those in Fig. 2.
graphite clustering in the irradiated samples. To estimate the nature of hybridization we have carried out XPS measurements and the irradiation induced graphite clustering is investigated using HRTEM. These results are discussed below. XPS spectra of the pristine and the irradiated samples are shown in Fig. 2. Apart from the C 1s peak, the pristine sample shows the presence of N 1s (not visible on this scale), O 1s and F 1s peaks. These peaks are also seen in the irradiated samples with varying intensities. The core level spectra of N 1s and F 1s are shown in Fig. 3. The C 1s spectra are shown in Fig. 4. In the pristine sample the ratio of intensity of F 1s to C 1s is 0.3 and N 1s to C 1s is 0.06. With irradiation, the intensities of F 1s and N 1s are found to be decreasing and in sample D both these peaks are absent. In sample D the F 1s peak is absent, while the N 1s peak has nearly vanished (Fig. 3). The fitting of the spectra in Fig. 4 is done by a x2 iteration programme using a convolution of Lorentzian–Gaussian function with a Sherly background [19]. The core level spectra of C 1s in all the samples were fitted with components positioned
around 284.3 and 285.0, 286.2 and 288.8 eV. The peaks observed at 284.3 and 285.0 eV correspond to sp2 and sp3 hybridized carbon atoms, respectively [20]. The peaks at 286.2 and 288.8 eV can be due to C–O and C O, respectively [21,22]. The above two peaks observed in the higher binding energy side of C 1s spectrum can also be due to CN and CF bondings, i.e., CN can give a peak at 286.2 eV and CF at 288.8 eV [22,23]. The intensity of the peak at 288.8 eV in the pristine sample is 2% of the total intensity of C 1s peak and is barely visible in samples B and C. In sample D this peak has an intensity of 7% of the total C 1s peak intensity and is clearly visible. The intensities of F 1s/C 1s and N 1s/C 1s have reduced with ion fluence and in sample D the fluorine and the nitrogen peaks were absent [Fig. 3(a)–(d)]. In C 1s spectra, the intensities of the 286.2 and 288.8 eV peaks have not reduced with increasing ion fluence. This indicates that F and N are surface impurities and the peaks at 286.2 and 288.8 eV were predominantly due to C–O and C O bond, respectively. In the pristine sample the intensity ratio of O 1s/C 1s is 0.4 and it has became 0.6 in sample D (in earlier XPS
Fig. 4. XPS spectra of C 1s electrons extracted from the spectra in Fig. 2: (a) pristine DLC/Si(1 0 0) sample, and samples irradiated at a fluence of (b) 1 1014 ions/cm2, (c) 1 1015 ions/cm2 and (d) 5 1015 ions/cm2.
Fig. 5. The variation of sp2/sp3 ratio of pristine and irradiated DLC films estimated from the spectra in Fig. 4.
S. Mathew et al. / Applied Surface Science 255 (2009) 4796–4800
experiments, this trend has also been observed for halogenadsorbed Si surfaces, even without ion irradiation. For example, F desorption from Si–F bonds leads to silicon oxide formation [25]). We do not observe a noticeable shift of binding energies of C 1s peak position after irradiation. The binding energy positions and FWHMs of all the constituent peaks were kept constant for fitting the spectra of the pristine and the irradiated samples as shown in Fig. 4. The line shape of the XPS spectrum gives information about the chemical bonding environments in the sample [20,24]. The ratio of the intensities of sp2 to sp3 (sp2/sp3) in the pristine sample is 0.93 and it increases to 1.33 in sample D. The variation of sp2/sp3 ratio is plotted in Fig. 5. Both XPS and Raman spectroscopy measurements clearly indicate an increase of trigonal bonding with the ion fluence. Our Raman spectroscopy results show the features, based on which earlier workers speculated the formation of graphite clusters in the ion-irradiated DLC films [11]. We have used HRTEM measurements for direct verification of the presence of graphite clusters in our irradiated samples. Transmission electron micrographs of the pristine sample (sample A) and the sample irradiated at a fluence of 5 1015 ions/cm2(sample D) are shown in Fig. 6(a) and (b), respectively. The appearance of nanometer (nm) sized ( 2 nm) clusters are clearly visible in the irradiated sample. A HRTEM image of the irradiated sample (D) is shown in Fig. 7. From HRTEM, the observed lattice spacing of 0.34 nm in clusters is found to be that of graphite. A filtered image [using a fast Fourier transform (FFT) technique] from a single cluster is shown in the inset of Fig. 7. This clearly shows that MeV nitrogen irradiation induces the formation of graphite clusters in DLC films. Very recently Choi et al. observed the formation graphite clusters in DLC films after rapid thermal annealing at 900 C in nitrogen ambient [26]. Talapatra et al. observed the formation of nm sized graphite
4799
Fig. 7. A HRTEM image of an irradiated DLC film — the same sample as in Fig. 6(b). A FFT filtered image of one of the observed graphite clusters is shown in the inset.
clusters in nanodiamonds after keV N and C ion irradiation [27] and also in thermally annealed samples [28]. sp2 clustering in tetrahedral amorphous carbon samples induced by thermal annealing have also been reported [29]. In the present study of MeV N+ irradiation of DLC films the N+ ions pass through the DLC film and get buried into the substrate silicon; the increase of trigonal bonding in the DLC films can be mainly due to the defects produced during ion irradiation. The nuclear and the electronic energy loss of 2 MeV N+ ions in an amorphous carbon target with the density of DLC is estimated (using SRIM 2003 [30]) to be 1.354 keV/nm and 0.648 keV/nm, respectively. The effect of ion irradiation on DLC films can be related to the film microstructure. Ion irradiation of any solid induces the breakage of bonds, creation of interstitials and vacancies, and also loss of hydrogen in the case of DLC films [11]. An estimate about the amount of hydrogen in the DLC samples can be obtained from the photoluminescence (PL) background in the first order Raman spectra. The ratio of the slope of the fitted linear background of the Raman spectrum to the intensity of the G peak can be used as a measure of the bonded H content in the film [31]. The amount of hydrogen content in the pristine film is 40% and the Raman spectra analysis did not show a significant loss of hydrogen in our irradiated samples. Amorphous carbons with the same sp2/sp3 ratio and H content are reported to show different optical, mechanical and electrical properties according to the clustering of sp2 phase [13]. The recent experiments had demonstrated room temperature ferromagnetism in heat-treated and ion-irradiated (100 keV C or N) nanodiamond samples [15,28]. In both of these cases, graphite nanoclusters were formed either by heat treatment or ion irradiation [27,28], and the authors attribute the observed ferromagnetism to the presence of graphite nanoclusters. In our case, from the formation of graphite nanoclusters along with a mixed sp2–sp3 hybridization in the ion irradiated DLC samples, one may speculate the possibility of magnetic ordering in these irradiated DLC films. 4. Conclusions
Fig. 6. Transmission electron micrograph of (a) a pristine DLC film and (b) a DLC film irradiated at a fluence of 5 1015 ions/cm2.
MeV N+ irradiation on DLC films leads to an increase of sp2/sp3 hybridization ratio. With increasing ion fluence the sp2/sp3 ratio is found to be increasing and samples irradiated at a fluence of 5 1015 ions/cm2 show a 40% increase in comparison with the pristine sample. Raman spectroscopy results, namely, the shifting of the positions of the D and the G peaks, along with the increase of the I(D)/I(G) ratio, are also an indication of an increase of sp2 bonding in the ion irradiated DLC samples. TEM images from the
4800
S. Mathew et al. / Applied Surface Science 255 (2009) 4796–4800
irradiated samples show the formation of nanometer sized clusters. HRTEM results reveal the details within the clusters and show that the clusters are indeed graphite clusters, consistent with the increase of sp2 bonding. From the observed formation of graphite nanoclusters, along with a mixed sp2–sp3 hybridization in the ion-irradiated DLC films, one is tempted to speculate about the possibility of magnetic ordering in the irradiated DLC films. Further experiments such as the quantification of hydrogen content in the film, conductivity and hardness measurements, investigation of magnetic ordering, etc. can provide further insights into the basic understanding and the technological applications of these ionirradiated DLC films.
[8] [9] [10] [11]
Acknowledgment
[19] [20]
We acknowledge Prof. S.N. Sahu and Mr. S.N. Sarangi for Raman spectroscopy measurements.
[21] [22]
References
[23] [24] [25]
[1] J. Robertson, Mater. Sci. Eng. R 37 (2002) 129. [2] R. Hauert, Diamond. Relat. Mater. 12 (2003) 583. [3] N.R. Ray (Ed.). Applications of Nanocrystalline Diamond and Diamond Like Carbon materials, Proceedings of IWAncDLC-2006 (Saha Institute of Nuclear Physics, Kolkata, 2006). [4] A. Erdemir, C. Donnet, J. Phys. D 39 (2006) R311. [5] C. Casiraghi, J. Robertson, A.C. Ferrari, Mater. Today 10 (2007) 44. [6] M.S. Dresselhaus, R. Kalish, Ion Implantation in Diamond, Graphite and Related Materials, Chapter 9, Springer Series in Materials Science 22, Berlin, 1992. [7] V.D. Vankar, N. Dilawar, Vacuum 47 (1996) 1275.
[12] [13] [14] [15] [16] [17] [18]
[26] [27] [28] [29] [30] [31]
D.K. Avasti, Vacuum 47 (1996) 165. D.L. Baptista, F.C. Zawislak, Diamond. Relat. Mater. 13 (2004) 1791. D.H. Lee, S. Fayeulle, K.C. Walter, M. Nastasi, Nucl. Instr. Meth. B 148 (1999) 216. H.C. Barshilia, S. Sah, B.R. Mehta, V.D. Vankar, D.K. Avasthi, G.K. Jaipal, Mehta, Thin Solid Films 258 (1995) 123. G. Compagnini, S. Battiato, O. Puglisi, G.A. Baratta, G. Strazzulla, Carbon 43 (2005) 3025. A.C. Ferrari, J. Robertson, Phys. Rev. B. 61 (2000) 14095. S. Mathew, B. Satpati, B. Joseph, B.N. Dev, R. Nirmala, S.K. Mallik, R. Kesavamoorthy, Phys. Rev. B 75 (2007) 75426. S. Talapatra, P.G. Ganesan, T. Kim, T.R. Vajtai, H. Huang, M. Shima, G. Ramanath, D. Srivastava, S.C. Deevi, P.M. Ajayan, Phys. Rev. Lett. 95 (2005) 97201. P. Esquinazi, D. Spemann, R. Hhne, A. Setzer, K.H. Han, T. Butz, Phys. Rev. Lett. 91 (2003) 227201. T.E. Doyle, J.R. Dennison, Phys. Rev. B. 51 (1995) 196. M.P. Siegal, D.R. Tallant, L.J. Martinez-Miranda, J.C. Barbour, R.L. Simpson, D.L. Overmyer, Phys. Rev. B 61 (2000) 10451. D.A. Shirly, Phys. Rev. 55 (1972) 4709. J.F. Morar, F.J. Himpsel, G. Hollinger, J.L. Jordan, G. Hughes, F.R. McFeely, Phys. Rev. B 33 (1986) 1340. Y. Taki, O. Takai, Thin Solid Films 316 (1998) 45. K. Yamamoto, Y. Koga, S. Fujiwara, F. Kokai, J.I. Kleiman, K.K. Kim, Thin Solid Films 339 (1999) 38. A. Tressaud, T. Shirasaki, G. Nans, E. Papirer, Carbon 40 (2002) 217. R. Haerle, E. Riedo, A. Pasquarello, A. Baldereschi, Phys. Rev. B 65 (2001) 45101. K. Sekar, G. Kuri, D.P. Mahapatra, B.N. Dev, J.V. Ramana, S. Kumar, V.S. Raju, Surf. Sci. 302 (1994) 25. W.S. Choi, B. Hong, Renew. Energ. 33 (2008) 226. S. Talapatra, J.-Y. Cheng, N. Chakrapani, S. Trasobares, A. Cao, R. Vajtai, M.B. Huang, P.M. Ajayan, Nanotechnology 17 (2006) 305. S. Talapatra, T. Kim, B.Q. Wei, S. Kar, R. Vajtai, G.V.S. Sastry, M. Shima, D. Srivastava, P.M. Ajayan, Nanopage 1 (2006) 315. J.O. Orwa, I. Andrienko, J.L. Peng, S. Prawer, Y.B. Zhang, S.P. Lau, J. Appl. Phys. 96 (2004) 6286. J.F. Ziegler, J.P. Biersack, U. Littmark, SRIM 2003—a version of the TRIM program. B. Marchon, J. Gui, J.K. Grannen, G.C. Rauch, IEEE Trans. Magn. 33 (1997) 3148.