- Email: [email protected]
Amorphization of carbon materials studied by X-ray photoelectron spectroscopy
NIM B Beam Interactions with Materials & Atoms
Nuclear Instruments and Methods in Physics Research B 242 (2006) 445–447 www.elsevier.com/locate/nimb
Amorphization of carbon materials studied by X-ray photoelectron spectroscopy Katsumi Takahiro
a,*
, Atsushi Terai a, Shinnosuke Oizumi a, Kiyoshi Kawatsura a, Shunya Yamamoto b, Hiroshi Naramoto c
a b
Kyoto Institute of Technology, Department of Chemistry and Materials Technology, Matsugasaki, Sakyo, Kyoto 606-8585, Japan Department of Materials Development, Japan Atomic Energy Research Institute, 1233 Watanuki, Takasaki, Gunma 370-1292, Japan c Advanced Science Research Center, Japan Atomic Energy Research Institute, 1233 Watanuki, Takasaki, Gunma 370-1292, Japan Available online 25 October 2005
Abstract X-ray photoelectron spectroscopy (XPS) and Raman spectroscopy have been applied to investigate amorphization processes by ion irradiation for various carbons, including highly oriented pyrolytic graphite (HOPG), isotropic graphite, glassy carbon (GC) and C60. It is found that the asymmetry of the XPS C 1s line increases as the irradiation dose increases. The origin of the asymmetry appeared on the C 1s line is discussed. We conclude that the asymmetry of the C 1s line does not relate to the increase in the size of a graphitic layer, but relates to structural disorders, such as a bond angle disorder. 2005 Elsevier B.V. All rights reserved. PACS: 61.80.Jh; 61.43.Er; 61.48.+c; 79.60. i Keywords: Amorphization; X-ray photoelectron spectroscopy; Carbon; Asymmetry; Raman
1. Introduction Ion implantation into carbon materials, such as graphite, diamond, C60 and glassy carbon, results in changing their chemical and physical properties. Crystalline-toamorphous transformation called ‘‘amorphization’’ causes a drastic change in their properties [1]. Amorphization is one of the most important phenomena in the fields of ion beam modification of carbons. Raman spectroscopy is a standard tool for the structural characterization of ion-implanted carbons. In case of ionimplanted graphite, for example, the structural changes are well characterized by the broadening of two prominent lines in a Raman spectrum, including the Raman active E2g line at 1580 cm 1 (referred to as G band) and the disorder induced line at 1355 cm 1 (referred to as D band) [2].
*
Corresponding author. Tel./fax: +81 75 724 7507. E-mail address: [email protected] (K. Takahiro).
0168-583X/$ - see front matter 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.nimb.2005.09.009
However, Raman spectroscopy becomes insensitive to the changes in the structure of carbons after amorphization occurs because the Raman spectrum exhibits a broad band in the range of 900–1800 cm 1 [2]. In carbon materials, the environment around a carbon atom should affect its electronic structure on core-level as well as valence band. Therefore, the examination of the electronic structure can provide information about the structure of carbons. X-ray photoelectron spectroscopy (XPS) is widely used to study the electronic structure and can be applied to the structural characterization of carbons [3–5]. In our previous work for deuteron-irradiated C60 films, we found that the shape of a C 1s XPS spectrum became asymmetric as the deuteron dose increased [6]. We speculated that the asymmetry resulted from the growth of a graphitic layer in the amorphized C60 films according to the analysis of Raman spectra. However, the origin of the asymmetry remains unclear. In the present work, amorphization processes have been investigated by using XPS
446
K. Takahiro et al. / Nucl. Instr. and Meth. in Phys. Res. B 242 (2006) 445–447
and Raman spectroscopy for various carbons, including highly oriented pyrolytic graphite (HOPG), isotropic graphite, glassy carbon (GC) and C60, to clarify the origin of the asymmetry appeared on the C 1s line. 2. Experimental procedure HOPG, manufactured by Molecular Device Tools for Nano Technology, Russia, was cleavaged with a Scotch Tape just before XPS measurements. Isotropic graphite (ISO-880U grade) was supplied from Toyo Tanso, Japan. Glassy carbon (Tokai Carbon, Japan; GC-30 grade) was mechanically polished to mirror surface with 1 lm diamond slurry on a cloth lap. C60 films were prepared onto Si substrates by vacuum evaporation of C60 powders of 99.98% purity, manufactured by Term, USA. XPS analysis using monochromatized Al Ka (hm = 1486.6 eV) radiation was performed with JEOL 9010. The base pressure in the analysis chamber for XPS measurement was about 1.0 · 10 7 Pa. The binding energies were calibrated by the Au 4f7/2 (83.9 eV) and Ag 3d5/2 (368.3 eV). The mean free path of an electron escaping from a C 1s level is 2–3 nm in carbons [7]. The carbon samples were irradiated with 1 keV Ar ions produced by a sputter-etching gun, which is equipped with JEOL 9010. The XPS analysis can be done without breaking vacuum after ion irradiation. Raman spectroscopy was performed by backscattering from the sample using an Ar laser operating at 514.5 nm with a laser power of 2.5 mW at a 10 lm spot. Higherenergy irradiation is needed to examine the damaged layer by using Raman spectroscopy because the information depth for carbon is approximately 100 nm, much larger than the depth of a damaged layer created by 1 keV Ar ions. The carbon samples were also irradiated with 100 keV Ar ions using the ion implanter at TIARA of JAERI Takasaki so that radiation damage distributes almost uniformly throughout the depth of 0–100 nm in the carbons. The depth distribution of damage was calculated by using the TRIM code [8]. 3. Results and discussion Fig. 1 shows C 1s core-level XPS spectra for HOPG irradiated with 1 keV Ar ions at various doses. The peak position of the C 1s line shifts slightly toward lower binding energy by 0.1 eV after irradiation. Significantly, the line width becomes larger as the irradiation dose increases. In addition, the shape of the C 1s line is found to be asymmetric upon irradiation. The same tendency for the change in a line shape was obtained in all carbon materials investigated. In Fig. 2, the half width at half maximum (HWHM) values of the C 1s line for HOPG, ISO-880U, GC-30 and C60/Si are plotted as a function of irradiation dose. In all kinds of samples examined, the HWHM values for the low-energy side, referred to as ‘‘HWHM-low’’, increases
Fig. 1. XPS C 1s spectra for HOPG irradiated with 1 keV Ar ions at various doses.
abruptly from 0.3 eV to 0.5 eV with irradiation at a dose of 0.05 · 1016 cm 2. Above that dose, the HWHM values are almost constant (0.5 eV) for any dose. In contrast, the HWHM values for the high-energy side, ‘‘HWHM-high’’, become larger gradually with the dose and reach 0.8–0.9 eV, depending on the sample, at a dose of 0.5 · 1016 cm 2. The HWHM-high at 1 · 1016 cm 2 for HOPG and ISO-880U is approximately 0.8 eV, slightly smaller than that for GC and C60/Si (0.9 eV), suggesting the difference in structure between the former and latter samples even when irradiated up to 1 · 1016 cm 2, corresponding to 3 dpa (displacements per atom). For all the samples, the difference between the HWHM-low and HWHM-high, that is, the asymmetry of the C 1s line, increases with increasing the dose. We shall discuss the evolution of the asymmetry accompanied by ion irradiation. Cheung [3] pointed out that the asymmetry of the C 1s line comes from the delocalization of p-electron on a graphitic component in an amorphous carbon. Therefore, the increased asymmetry can be explained by the increase in size of a graphitic domain. In fact, in our previous work, the growth of a graphitic layer was observed in the amorphized C60 films [6]. This is referred to as ‘‘size effect’’ hereafter. As shown below, however, ion irradiation should lead to the reduction in size of a graphitic domain in graphitic materials such as HOPG and ISO-880U. Information about the size of a graphitic layer in an ion-irradiated carbon can be obtained from a Raman spectrum since the
K. Takahiro et al. / Nucl. Instr. and Meth. in Phys. Res. B 242 (2006) 445–447
447
Fig. 3. Intensity ratio ID/IG in Raman spectra for HOPG and C60/Si as a function of irradiation dose of 100 keV Ar ions and accumulated damage in the unit of dpa. The solid and broken lines are drawn to guide the eyes.
Fig. 2. HWHM values of XPS C 1s lines on a lower binding energy side (HWHM-low) and on a higher binding energy side (HWHM-high) for HOPG, ISO-880U, GC-30 and C60/Si as a function of irradiation dose of 1 keV Ar ions and accumulated damage in the unit of dpa. The solid, broken and dotted lines are drawn to guide the eyes.
peak height ratio of the D band to G band, ID/IG, in the spectrum depends on the size of a graphitic layer [9,10]. Fig. 3 shows intensity ratio ID/IG in Raman spectra for HOPG and C60/Si as a function of irradiation dose of 100 keV Ar ions. In case of HOPG, the ID/IG increases as the dose increases up to 4 · 1014 cm 2, corresponding to 0.25 dpa, indicating that the size of a graphitic layer becomes smaller [9,10] as has been expected. On the other hand, the ID/IG for C60/Si decreases with the dose, suggesting the increase in size of a graphitic layer, which is consistent with the previous work [6]. According to the findings by Cheung [3], the shape of the C 1s line might be related to the size effect. However, this is not the case for HOPG. Thus, evolution of the asymmetry of the C 1s line shown in Fig. 2 does not relate to the increase in the size of a graphitic layer, but relates to structural disorders, such as a bond angle disorder, produced by ion irradiation. In summary, we have investigated the shapes of an XPS C 1s line for various carbon materials irradiated with 1 keV
Ar ions to clarify the origin of the asymmetry of the C 1s line for carbons amorphized by ion irradiation. The origin of the asymmetry has not been well understood yet, but we can exclude the size effect from it. The emergence of some higher-energy components corresponding to structural disorder peaks may result in the asymmetry of a C 1s line, although the higher-energy components are unidentified. XPS analysis can be applied to the analysis of the structure of amorphous carbons if the origin of the asymmetry of the C 1s line is completely understood. Acknowledgement This work was performed under the Joint Research Program at TIARA of JAERI Takasaki. References [1] M.S. Dresselhaus, R. Kalish, Ion Implantation in Diamond, Graphite and Related Materials, Springer-Verlag, Berlin, 1992, and references therein. [2] B.S. Elman, M.S. Dresselhaus, G. Dresselhaus, E.W. Maby, H. Mazurek, Phys. Rev. B 24 (1981) 1027. [3] T.T.P. Cheung, J. Appl. Phys. 53 (1982) 6857. [4] M. Ramm, M. Ata, K.W. Brzezinka, T. Gross, W. Unger, Thin Solid film 354 (1999) 106. [5] D.-Q. Yang, E. Sacher, Surf. Sci. 504 (2002) 125. [6] R. Ookawa, K. Takahiro, K. Kawatsura, F. Nishiyama, S. Yamamoto, H. Naramoto, Nucl. Instr. and Meth. B 206 (2003) 175. [7] S. Tanuma, C.J. Powell, D.R. Penn, Surf. Interface Anal. 21 (1994) 165. [8] J.F. Ziegler, J.P. Biersack, U. Littmark, The Stopping and Ranges of Ions in Solids, Pergamon, New York, 1985. [9] F. Tuinstra, J.L. Koening, J. Chem. Phys. 53 (1970) 1126. [10] A.C. Ferrari, J. Robertson, Phys. Rev. B 61 (2000) 14095.
Nuclear Instruments and Methods in Physics Research B 242 (2006) 445–447 www.elsevier.com/locate/nimb
Amorphization of carbon materials studied by X-ray photoelectron spectroscopy Katsumi Takahiro
a,*
, Atsushi Terai a, Shinnosuke Oizumi a, Kiyoshi Kawatsura a, Shunya Yamamoto b, Hiroshi Naramoto c
a b
Kyoto Institute of Technology, Department of Chemistry and Materials Technology, Matsugasaki, Sakyo, Kyoto 606-8585, Japan Department of Materials Development, Japan Atomic Energy Research Institute, 1233 Watanuki, Takasaki, Gunma 370-1292, Japan c Advanced Science Research Center, Japan Atomic Energy Research Institute, 1233 Watanuki, Takasaki, Gunma 370-1292, Japan Available online 25 October 2005
Abstract X-ray photoelectron spectroscopy (XPS) and Raman spectroscopy have been applied to investigate amorphization processes by ion irradiation for various carbons, including highly oriented pyrolytic graphite (HOPG), isotropic graphite, glassy carbon (GC) and C60. It is found that the asymmetry of the XPS C 1s line increases as the irradiation dose increases. The origin of the asymmetry appeared on the C 1s line is discussed. We conclude that the asymmetry of the C 1s line does not relate to the increase in the size of a graphitic layer, but relates to structural disorders, such as a bond angle disorder. 2005 Elsevier B.V. All rights reserved. PACS: 61.80.Jh; 61.43.Er; 61.48.+c; 79.60. i Keywords: Amorphization; X-ray photoelectron spectroscopy; Carbon; Asymmetry; Raman
1. Introduction Ion implantation into carbon materials, such as graphite, diamond, C60 and glassy carbon, results in changing their chemical and physical properties. Crystalline-toamorphous transformation called ‘‘amorphization’’ causes a drastic change in their properties [1]. Amorphization is one of the most important phenomena in the fields of ion beam modification of carbons. Raman spectroscopy is a standard tool for the structural characterization of ion-implanted carbons. In case of ionimplanted graphite, for example, the structural changes are well characterized by the broadening of two prominent lines in a Raman spectrum, including the Raman active E2g line at 1580 cm 1 (referred to as G band) and the disorder induced line at 1355 cm 1 (referred to as D band) [2].
*
Corresponding author. Tel./fax: +81 75 724 7507. E-mail address: [email protected] (K. Takahiro).
0168-583X/$ - see front matter 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.nimb.2005.09.009
However, Raman spectroscopy becomes insensitive to the changes in the structure of carbons after amorphization occurs because the Raman spectrum exhibits a broad band in the range of 900–1800 cm 1 [2]. In carbon materials, the environment around a carbon atom should affect its electronic structure on core-level as well as valence band. Therefore, the examination of the electronic structure can provide information about the structure of carbons. X-ray photoelectron spectroscopy (XPS) is widely used to study the electronic structure and can be applied to the structural characterization of carbons [3–5]. In our previous work for deuteron-irradiated C60 films, we found that the shape of a C 1s XPS spectrum became asymmetric as the deuteron dose increased [6]. We speculated that the asymmetry resulted from the growth of a graphitic layer in the amorphized C60 films according to the analysis of Raman spectra. However, the origin of the asymmetry remains unclear. In the present work, amorphization processes have been investigated by using XPS
446
K. Takahiro et al. / Nucl. Instr. and Meth. in Phys. Res. B 242 (2006) 445–447
and Raman spectroscopy for various carbons, including highly oriented pyrolytic graphite (HOPG), isotropic graphite, glassy carbon (GC) and C60, to clarify the origin of the asymmetry appeared on the C 1s line. 2. Experimental procedure HOPG, manufactured by Molecular Device Tools for Nano Technology, Russia, was cleavaged with a Scotch Tape just before XPS measurements. Isotropic graphite (ISO-880U grade) was supplied from Toyo Tanso, Japan. Glassy carbon (Tokai Carbon, Japan; GC-30 grade) was mechanically polished to mirror surface with 1 lm diamond slurry on a cloth lap. C60 films were prepared onto Si substrates by vacuum evaporation of C60 powders of 99.98% purity, manufactured by Term, USA. XPS analysis using monochromatized Al Ka (hm = 1486.6 eV) radiation was performed with JEOL 9010. The base pressure in the analysis chamber for XPS measurement was about 1.0 · 10 7 Pa. The binding energies were calibrated by the Au 4f7/2 (83.9 eV) and Ag 3d5/2 (368.3 eV). The mean free path of an electron escaping from a C 1s level is 2–3 nm in carbons [7]. The carbon samples were irradiated with 1 keV Ar ions produced by a sputter-etching gun, which is equipped with JEOL 9010. The XPS analysis can be done without breaking vacuum after ion irradiation. Raman spectroscopy was performed by backscattering from the sample using an Ar laser operating at 514.5 nm with a laser power of 2.5 mW at a 10 lm spot. Higherenergy irradiation is needed to examine the damaged layer by using Raman spectroscopy because the information depth for carbon is approximately 100 nm, much larger than the depth of a damaged layer created by 1 keV Ar ions. The carbon samples were also irradiated with 100 keV Ar ions using the ion implanter at TIARA of JAERI Takasaki so that radiation damage distributes almost uniformly throughout the depth of 0–100 nm in the carbons. The depth distribution of damage was calculated by using the TRIM code [8]. 3. Results and discussion Fig. 1 shows C 1s core-level XPS spectra for HOPG irradiated with 1 keV Ar ions at various doses. The peak position of the C 1s line shifts slightly toward lower binding energy by 0.1 eV after irradiation. Significantly, the line width becomes larger as the irradiation dose increases. In addition, the shape of the C 1s line is found to be asymmetric upon irradiation. The same tendency for the change in a line shape was obtained in all carbon materials investigated. In Fig. 2, the half width at half maximum (HWHM) values of the C 1s line for HOPG, ISO-880U, GC-30 and C60/Si are plotted as a function of irradiation dose. In all kinds of samples examined, the HWHM values for the low-energy side, referred to as ‘‘HWHM-low’’, increases
Fig. 1. XPS C 1s spectra for HOPG irradiated with 1 keV Ar ions at various doses.
abruptly from 0.3 eV to 0.5 eV with irradiation at a dose of 0.05 · 1016 cm 2. Above that dose, the HWHM values are almost constant (0.5 eV) for any dose. In contrast, the HWHM values for the high-energy side, ‘‘HWHM-high’’, become larger gradually with the dose and reach 0.8–0.9 eV, depending on the sample, at a dose of 0.5 · 1016 cm 2. The HWHM-high at 1 · 1016 cm 2 for HOPG and ISO-880U is approximately 0.8 eV, slightly smaller than that for GC and C60/Si (0.9 eV), suggesting the difference in structure between the former and latter samples even when irradiated up to 1 · 1016 cm 2, corresponding to 3 dpa (displacements per atom). For all the samples, the difference between the HWHM-low and HWHM-high, that is, the asymmetry of the C 1s line, increases with increasing the dose. We shall discuss the evolution of the asymmetry accompanied by ion irradiation. Cheung [3] pointed out that the asymmetry of the C 1s line comes from the delocalization of p-electron on a graphitic component in an amorphous carbon. Therefore, the increased asymmetry can be explained by the increase in size of a graphitic domain. In fact, in our previous work, the growth of a graphitic layer was observed in the amorphized C60 films [6]. This is referred to as ‘‘size effect’’ hereafter. As shown below, however, ion irradiation should lead to the reduction in size of a graphitic domain in graphitic materials such as HOPG and ISO-880U. Information about the size of a graphitic layer in an ion-irradiated carbon can be obtained from a Raman spectrum since the
K. Takahiro et al. / Nucl. Instr. and Meth. in Phys. Res. B 242 (2006) 445–447
447
Fig. 3. Intensity ratio ID/IG in Raman spectra for HOPG and C60/Si as a function of irradiation dose of 100 keV Ar ions and accumulated damage in the unit of dpa. The solid and broken lines are drawn to guide the eyes.
Fig. 2. HWHM values of XPS C 1s lines on a lower binding energy side (HWHM-low) and on a higher binding energy side (HWHM-high) for HOPG, ISO-880U, GC-30 and C60/Si as a function of irradiation dose of 1 keV Ar ions and accumulated damage in the unit of dpa. The solid, broken and dotted lines are drawn to guide the eyes.
peak height ratio of the D band to G band, ID/IG, in the spectrum depends on the size of a graphitic layer [9,10]. Fig. 3 shows intensity ratio ID/IG in Raman spectra for HOPG and C60/Si as a function of irradiation dose of 100 keV Ar ions. In case of HOPG, the ID/IG increases as the dose increases up to 4 · 1014 cm 2, corresponding to 0.25 dpa, indicating that the size of a graphitic layer becomes smaller [9,10] as has been expected. On the other hand, the ID/IG for C60/Si decreases with the dose, suggesting the increase in size of a graphitic layer, which is consistent with the previous work [6]. According to the findings by Cheung [3], the shape of the C 1s line might be related to the size effect. However, this is not the case for HOPG. Thus, evolution of the asymmetry of the C 1s line shown in Fig. 2 does not relate to the increase in the size of a graphitic layer, but relates to structural disorders, such as a bond angle disorder, produced by ion irradiation. In summary, we have investigated the shapes of an XPS C 1s line for various carbon materials irradiated with 1 keV
Ar ions to clarify the origin of the asymmetry of the C 1s line for carbons amorphized by ion irradiation. The origin of the asymmetry has not been well understood yet, but we can exclude the size effect from it. The emergence of some higher-energy components corresponding to structural disorder peaks may result in the asymmetry of a C 1s line, although the higher-energy components are unidentified. XPS analysis can be applied to the analysis of the structure of amorphous carbons if the origin of the asymmetry of the C 1s line is completely understood. Acknowledgement This work was performed under the Joint Research Program at TIARA of JAERI Takasaki. References [1] M.S. Dresselhaus, R. Kalish, Ion Implantation in Diamond, Graphite and Related Materials, Springer-Verlag, Berlin, 1992, and references therein. [2] B.S. Elman, M.S. Dresselhaus, G. Dresselhaus, E.W. Maby, H. Mazurek, Phys. Rev. B 24 (1981) 1027. [3] T.T.P. Cheung, J. Appl. Phys. 53 (1982) 6857. [4] M. Ramm, M. Ata, K.W. Brzezinka, T. Gross, W. Unger, Thin Solid film 354 (1999) 106. [5] D.-Q. Yang, E. Sacher, Surf. Sci. 504 (2002) 125. [6] R. Ookawa, K. Takahiro, K. Kawatsura, F. Nishiyama, S. Yamamoto, H. Naramoto, Nucl. Instr. and Meth. B 206 (2003) 175. [7] S. Tanuma, C.J. Powell, D.R. Penn, Surf. Interface Anal. 21 (1994) 165. [8] J.F. Ziegler, J.P. Biersack, U. Littmark, The Stopping and Ranges of Ions in Solids, Pergamon, New York, 1985. [9] F. Tuinstra, J.L. Koening, J. Chem. Phys. 53 (1970) 1126. [10] A.C. Ferrari, J. Robertson, Phys. Rev. B 61 (2000) 14095.