Optical property changes in sapphire induced by triple-energy Cu and O implantation

Optical property changes in sapphire induced by triple-energy Cu and O implantation

Surface and Coatings Technology 158 – 159 (2002) 720–724 Optical property changes in sapphire induced by triple-energy Cu and O implantation M. Ikeya...

174KB Sizes 0 Downloads 7 Views

Surface and Coatings Technology 158 – 159 (2002) 720–724

Optical property changes in sapphire induced by triple-energy Cu and O implantation M. Ikeyama*, S. Nakao, M. Tazawa National Institute of Advanced Industrial Science and Technology (AIST), Institute for Structural and Engineering Materials, AIST Chubu, 2266-98 Anagahora, Shimoshidami, Moriyama-ku, Nagoya 463-8560, Japan

Abstract Triple-energy Cu and O ions were implanted into sapphire, changing the ratios of Cu and O doses and the implantation sequence as follows: Cu only (1:0); Cuq1 y2 O (1:0.5); CuqO (1:1); 1 y2 OqCu (0.5:1); and OqCu (1:1) at room temperature (300 K). Optical property changes induced by the ion implantations and successive heat treatments were studied. Specific optical absorption is clearly observed at approximately 590 nm, which is attributed to Cu nano-particles, for Cu-implanted sapphire after annealing at 1070 and 1270 K. The intensity is drastically changed between 1070 and 1270 K. A broad absorption centered at approximately 300 nm is also observed for samples annealed at 770 and 1070 K. For Cu and O co-implantation, an increase in O dose leads to lower absorption on the whole. The absorption at approximately 590 nm is observed after annealing at 1070 or 1270 K for Cuq1y 2O and CuqO, but not for 1y 2OqCu and OqCu implantation. In general, optical absorption of sapphire increases after annealing, especially in the short-wavelength region, and this can be attributed to copper oxide formation. The formation of Cu, Cu2O and CuO nanoparticles was confirmed by XRD measurements. CuO and Cu2 O are easily formed at a lower annealing temperature, whereas Cu nanoparticle formation requires a higher annealing temperature. The sequence of ion implantation for Cu and O affects the optical absorption and nanoparticle formation. 䊚 2002 Elsevier Science B.V. All rights reserved. Keywords: Ion implantation; Optical property; Triple energy; Sapphire; Nanoparticles

1. Introduction Nano-particles have attracted much attention recently because of the possibility of their application in luminescent and non-linear optical devices. Ion implantation is one of the techniques useful for the formation of nanoparticles. Many studies have been carried out to prepare nanoparticles by ion implantation and have revealed their luminescent w1x and non-linear optical properties w2–5x. We have studied the formation of metal or metal-oxide nanoparticles by metal implantation or co-implantation of metal and oxygen into silica glass or sapphire w6–8x. In our previous study w9x, optical property changes in silica glass induced by Cu and O multi-energy (triple-energy) implantation with successive annealing were studied, and the formation of copper or copper oxide nanoparticles was revealed. The *Corresponding author. Tel.: q81-52-736-7284; fax: q81-52-7367406. E-mail address: [email protected] (M. Ikeyama).

formation of Cu nanoparticles was enhanced by the multi-energy ion implantation. Silica glass is a typical amorphous material. Some differences in the formation of nanoparticles in crystalline materials, such as sapphire, can be expected. In this study, we studied optical property changes in sapphire induced by triple-energy Cu and O ion implantation and subsequent heat treatments. 2. Experiment Optically flat sapphire samples were subjected to triple-energy implantation with 1.2-, 1.7- and 2.4-MeV Cuq and 0.4-, 0.7- and 1.2-MeV Oq ions. The energy values were chosen in order to obtain the same projection ranges for O and Cu. The dose of each Cu implantation was 1=1017 ionsycm2. The dose of O ions and the implantation sequence for Cu and O were changed to yield Cu, Cuq1 y 2O, 1 y 2OqCu, CuqO and OqCu, where 1 y 2O and O indicate a dose of 0.5=1017 and 1=1017 O ionsycm2, respectively. For the

0257-8972/02/$ - see front matter 䊚 2002 Elsevier Science B.V. All rights reserved. PII: S 0 2 5 7 - 8 9 7 2 Ž 0 2 . 0 0 2 5 8 - X

M. Ikeyama et al. / Surface and Coatings Technology 158 – 159 (2002) 720–724

Fig. 1. (a) Optical absorption and (b) XRD patterns of sapphire samples after triple-energy (1.2, 1.7 and 2.4 MeV) Cu-ion implantation at room temperature (300 K) and successive annealing for 1 h in air. The dose was 1=1017 ionsycm2 for each implantation. The annealing temperature is shown in the figure.

implantation energy, we adopted an energy increase mode, i.e. the lowest energy implantation was first, the second was the middle energy and the last was the highest energy implantation. All implantations were performed at room temperature (approx. 300 K). After the implantation, each sample was cut into four pieces and three pieces of the sample were annealed at 770, 1070 and 1270 K for 1 h in air, respectively. Optical absorption was measured with a spectrophotometer (JASCO V-570). X-Ray diffraction (XRD) measurements were also performed with an X-ray diffractometer (Mac Science MXP-3) to identify the particles formed. 3. Results and discussion The results of optical absorption and XRD measurements for triple-energy Cu implanted sapphire are shown

721

in Fig. 1. It is very clear that the absorption band at approximately 590 nm, which is attributed to the surface plasmon resonance absorption of Cu nanoparticles, is observed for the sample annealed at 1270 K (Fig. 1a). The absorption band is very small for the sample annealed at 1070 K and almost non-existent for the 770K-annealed and as-implanted samples. The formation of Cu nanoparticles in the sample annealed at 1270 K was confirmed by XRD measurements, as shown in Fig. 1b. Thus, the optical absorption at approximately 590 nm can be attributed to the surface plasmon resonance of Cu nanoparticles w2,5,6x. On the other hand, a broad absorption centered at approximately 300 nm is very clear for samples annealed at 770 and 1070 K. It is believed that this absorption is caused by an increase in the metal Cu fraction and is possibly due to the formation of very small Cu particles w6x. There is no peak in the XRD patterns for samples annealed at 770 and 1070 K, although they show broad absorption at approximately 300 nm, which is attributed to the metallic Cu fraction with no surface plasmon resonance absorption. Cheshnovsky et al. w10x reported that Cu particles smaller than 0.7 nm in diameter have no surface plasmon resonance absorption. Therefore, the size of the metallic Cu particles in our samples is assumed to be smaller than 0.7 nm in diameter, and this might be too small to be detected by XRD measurements. The results of optical absorption measurements for triple-energy Cu- and O-implanted sapphire are shown in Fig. 2. Absorption at approximately 590 nm is observed for Cuq1 y 2O and CuqO implantation after annealing at 1070 or 1270 K (Fig. 2a,b). However, for 1 y 2OqCu and OqCu implantation (Fig. 2c,d), there are no clear absorption bands at 590 nm, and a broad absorption at approximately 450 nm is observed for asimplanted and 770-K-annealed samples. The sequence of implantation of Cu and O ions is a very effective method for changing the optical properties of sapphire. In general, the optical absorption of sapphire increases after annealing, especially for the short-wavelength region. To observe the effects of the dose ratio and implantation sequence of Cu and O ions on the optical absorption in detail, the data of Fig. 2 are plotted according to annealing condition in Fig. 3. Co-implantation with Cu and O ions reduced the absorption on the whole compared with Cu implantation for all annealing conditions. Even after annealing at 1270 K, the absorption at approximately 590 nm attributed to Cu nanoparticles is very small for Cuq1 y 2O and CuqO implantation, and almost non-existent for 1 y 2OqCu and OqCu implantation. This is the same as for silica glass samples reported in our previous study w9x. It is very clear that O implantation obstructs the formation of Cu nanoparticles by Cu implantation and

722

M. Ikeyama et al. / Surface and Coatings Technology 158 – 159 (2002) 720–724

Fig. 2. Optical absorption of sapphire samples of (a) Cuq1y2O, (b) CuqO, (c) 1y2OqCu and (d) OqCu, prepared by triple-energy (1.2, 1.7 and 2.4 MeV) Cu- and (0.4, 0.7 and 1.2 MeV) O-ion implantation at room temperature (300 K) and successive annealing for 1 h in air. The dose was 1=1017 ionsycm2 for each Cu implantation and 0.5=1017 (labeled by 1 y2 O) or 1=1017 (labeled by O) ionsycm2 for each O implantation. The annealing temperature is shown in the figure.

annealing. This is confirmed by the drastic difference in absorbance between Cu only implantation and CuqO implantation for annealing at 770 and 1070 K, and the absorption centered at 300 nm, which represents the amount of metallic Cu, as mentioned above. O implantation has very large influence; however, there is almost no change between Cuq1 y 2O and CuqO implantation. For silica glass, there was a large difference between Cuq1 y 2O and CuqO implantation w9x. On the other hand, there is a clear difference between 1 y 2OqCu and OqCu implantation, especially in the case of asimplanted and 770-K-annealed samples. Obviously, an increase in O dose leads to lower absorption on the whole. With increasing annealing temperature, the difference becomes smaller. As mentioned above, the sequence of Cu and O implantation clearly affects the optical absorption property and Cu nanoparticle formation. In order to identify particles formed in the sapphire samples, results of XRD measurements for the same

samples as in Figs. 2 and 3 are shown in Fig. 4, except for the as-implanted samples, for which there was almost no change from the original sapphire sample. The peaks at 36.58 due to Cu2O(111) and at 43.08 due to Cu(111) are observed for all samples after annealing at 1070 K, and their intensity increases with increasing annealing temperature. Apart from the peaks at 36.58 due to Cu2O(111) and at 43.08 due to Cu(111), a very small peak at 39.48 due to CuO(200) is observed for the Cuq1 y 2O sample annealed at 770 K that increases for annealing at 1070 K; however, this peak is not clear after annealing at 1270 K, and another peak at 29.78 due Cu2O(110) is obvious. For CuqO and OqCu implantation, the peak at 39.48 due to CuO(200) is observed, even for samples annealed at 770 K, but this peak disappears with increasing annealing temperature. Instead of CuO(200), the peaks due to Cu2O(111) and Cu2O(110) increase with annealing temperature and the peaks at 31.78 and 48.48 due to CuO(110) and CuO(202), respectively,

M. Ikeyama et al. / Surface and Coatings Technology 158 – 159 (2002) 720–724

723

Fig. 3. Optical absorption of sapphire samples: (a) as-implanted; and after annealing at (b) 770; (c) 1070; and (d) and 1270 K with triple-energy Cu, Cuq1y2O, 1y2OqCu, CuqO and OqCu implantations. The energy of the Cu ions was 1.2, 1.7 and 2.4 MeV, and the dose was 1=1017 ionsycm2. The energy of the O ions was 0.4, 0.7 and 1.2 MeV, and the dose was 5=1016 or 1=1017 ionsycm2 for the label 1 y2 O or O, respectively.

appear for annealing at 1270 K. For 1 y 2OqCu implantation, CuO(111) is formed after annealing at 1070 and 1270 K. To compare Fig. 3 and Fig. 1b, it can be said that copper oxide particles consisting of CuO and Cu2O are easily formed at a lower annealing temperature and that Cu nanoparticle formation requires the higher temperature induced by Cu and O co-implantation into sapphire. In Fig. 2, the optical absorption of sapphire increases after annealing, especially for the short-wavelength region. This can be attributed to the formation of the copper oxides Cu2Oand CuO, because they have large optical absorption of visible light at wavelengths shorter than 500 nm. For silica glass, Cu nanoparticles are formed even without annealing, and the formation of copper oxide particles requires a higher annealing temperature w9x. There is a very large difference between sapphire and silica glass for nanoparticle formation induced by ion implantation. Nanoparticle formation might be affected

by the difference in crystallinity for sapphire and silica glass, which are single-crystal and amorphous, respectively. We have studied the crystal orientation of nanoparticles in sapphire using precise XRD measurement, including pole figures. Recently, we have found that the particles are highly oriented corresponding to the orientation of sapphire. These results will be presented elsewhere. 4. Summary Triple-energy Cu and O ions were implanted into sapphire, changing the ratios of Cu and O doses and the implantation sequence at room temperature (300 K). Optical absorption and XRD measurement for asimplanted and annealed samples were performed. Specific optical absorption at approximately 590 nm is clearly observed for Cu-implanted sapphire after

724

M. Ikeyama et al. / Surface and Coatings Technology 158 – 159 (2002) 720–724

Fig. 4. XRD patterns for the same samples as in Figs. 2 and 3, except for as-implanted ones.

annealing at 1070 and 1270 K. The intensity is drastically changed between 1070 and 1270 K. A broad absorption centered at approximately 300 nm is also observed for samples annealed at 770 and 1070 K. For Cu and O co-implantation, an increase in O dose leads to lower absorption on the whole. Absorption at approximately 590 nm is observed after annealing at 1070 or 1270 K for Cuq1 y 2O and CuqO implantation, but no clear absorption is observed for 1 y 2OqCu and OqCu implanted samples. In general, the optical absorption of sapphire increases after annealing, especially for the short-wavelength region, which can be attributed to copper oxide formation. The formation of Cu, Cu2O and CuO nanoparticles was confirmed by XRD measurements. CuO and Cu2O are easily formed at a lower annealing temperature, whereas Cu nanoparticle formation requires a higher annealing temperature. The sequence of ion implantation for Cu and O affects the optical absorption and nanoparticle formation.

References w1x T.S. Iwayama, S. Nakao, K. Saitoh, Appl. Phys. Lett. 65 (1994) 1814. w2x K. Uchida, S. Kaneko, S. Omi, et al., J. Opt. Soc. Am. B 11 (1994) 1236. w3x K. Fukumi, A. Chayahara, K. Kadono, et al., J. Appl. Phys. 75 (1994) 3075. w4x M. Ando, K. Kadono, M. Haruta, T. Sakaguchi, M. Miya, Nature 374 (1995) 625. w5x Y. Takeda, V.T. Gritsyna, N. Umeda, C.G. Lee, N. Kishimoto, Nucl. Instrum. Methods B 148 (1999) 1029. w6x S. Nakao, Y. Miyagawa, K. Saitoh, et al., Jpn. J. Appl. Phys. 36 (1997) 7681. w7x S. Nakao, K. Saitoh, M. Ikeyama, et al., Nucl. Instrum. Methods B 141 (1998) 246. w8x S. Nakao, M. Ikeyama, M. Tazawa, et al., Proceedings of the 1998 International Conference on Ion Implantation Technology, 1999, p. 771. w9x M. Ikeyama, S. Nakao, M. Tazawa, K. Kadono, K. Kamada, Nucl. Instrum. Methods B 175–177 (2001) 652. w10x O. Cheshnovsky, K.J. Taylor, J. Conceicao, R.E. Smalley, Phys. Rev. Lett. 64 (1990) 1785.