Pulsed laser deposition of carbon nitride thin films from graphite targets

CarbonVol.36,No. 5-6, pp. 771-774, 1998 0 1998Elwier Science Ltd Printed inGreatBritain. All rights reserved

0008-6223/9x$19.00+ 0.00

PII: SOOO8-6223(98)00004-9

PULSED LASER DEPOSITION OF CARBON NITRIDE THIN FILMS FROM GRAPHITE TARGETS Y. SUDA,~** T. NAKAZONO,~ K. EBIHARA,~ K. BABA’ and S. AOQUI~ BDepartment of Electrical Engineering, Sasebo National College of Technology, l-l Okishin-machi, Sasebo, Nagasaki 857-l 1,Japan bDepartment of Electrical Engineering and Computer Science, Kumamoto University, 2-39-l Kurokami, Kumamoto 860, Japan ‘Technology Center of Nagasaki, 2-1303-8 Ikeda, Omura, Nagasaki 856, Japan dKumamoto Institute of Technology, Ikeda, Kumamoto 860, Japan (Received 30 October 1997; accepted in revised form 9 December 1997)

Abstract-Carbon nitride thin films were synthesized on Si(lO0) substrates by a pulsed Nd:YAG laser deposition. The laser beam is incident on the high-purity graphite targets. The films are grown using an energy density 3.8 J cm-* at a laser repetition rate of IO Hz. The nitrogen gas pressure in the chamber is 10.0 Pa. Morphology features of the films have been obtained by employing the technique of scanning electron microscopy. Auger electron spectroscopy has been used to obtain compositional information about the films. The N/C composition ratio was found to vary from zero to 0.32 depending on deposition conditions. IR absorption spectra show two characteristic bands: a broad band composed of the graphite G-band and disordered D-band of carbon, and another associated with C-N triple bonds. Raman spectra have also been used to characterize the films. 0 1998 Elsevier Science Ltd. All rights reserved. Key Word-A. Carbon composites, C. infrared spectroscopy, C. Raman spectroscopy, electron spectroscopy (SEM).

C. scanning

1. INTRODUCTION The hardest materials known are diamond and cubic boron nitride, with hardness values of around 100 and 50 GPa, respectively. Recent calculations by Liu and Cohen [ 1,2] suggested that a hypothetical material, g-C,N4, may be as hard as diamond. This produced a surge of activity in the synthesis of carbon nitride (hereafter referred to as CN,) thin films. Various CN, films have been prepared by different deposition techniques including reactive DC magnetron sputtering [3,4], RF sputtering [5-71, chemical vapor deposition [8,9], ion beam assisted deposition [l&12] and pulsed laser deposition (PLD) [l3-171. The synthesis of pure g-C3N, remains an open challenge. In this paper we report our experimental results on synthesizing CN, films using pulsed laser deposition. All samples were analyzed by a field-emission secondary electron microscope (FE-SEM), Auger electron spectroscopy (AES), Fourier transform infrared spectroscopy (FT-IR) and a micro-Raman spectrometer.

Fig. 1. Schematic diagram of the Nd:YAG laser deposition system. YM600: wavelength of 532 nm, pulse duration of 6.5 ns, maximum output energy of 340 mJ). The laser beam was focused on the high purity (more than 99.999%) graphite targets at 45”, which were placed in the center of the stainless deposition chamber ((21400 mm x 370 mm). The radiated area was kept at about 2.8 mm’. The laser energy density was fixed at 3.8 Jcm- 2. The targets were rotated at about 20 rpm to avoid pitting during the deposition. Single crystal Si ( 100) substrates of size approximately 4 cm2 were ultrasonically cleaned in consecutive baths of ethanol and rinsed in high-purity deionized water prior to loading in the deposition chamber. Prior to actual deposition, the Si( 100) substrates were cleaned in situ by reverse sputter-etching to remove any residual contamination. The Si( 100) substrates were

2. EXPERIMENTAL The PLD apparatus used in these experiments is shown schematically in Fig. I [ 181. A deposition chamber was evacuated by a turbo-molecular pump and a rotary pump. The laser used in the present study was a pulsed Nd:YAG laser (Lumonics *Correspondingauthor.Tel: +81956 313261; Fax: f81 956 33 2895; e-mail: [email protected] 771

Y. SUIM r,nl

112

located at a distance of 60 mm from the facing target and were heated up to 650°C by an IR lamp. The substrate temperature was measured by a thermocouple. The thermocouple was used as an input to a programmable temperature controller that drove the input power to the IR lamp. An external RF bias at 13.56 MHz or negative DC bias was applied to the substrate holder. The gas pressure was varied from a base pressure (below 4.0x lob4 Pa) to 10.0 Pa ( 100% nitrogen). After 36 OOOG72000 laser shots at 10 Hz repetition rate, the deposition process was completed. The film thickness was about 2000 A and the deposition rate was about 33 A min-’ without bias voltage. Table I shows the deposition conditions for the preparation of CN, thin films. The surface morphology was observed by a fieldemission secondary electron microscope ( FE-SEM: JEOL JSM-6300F). The composition and structure of the CN, films were examined by Auger electron spectroscopy (AES: JEOL JUMP-30), Fourier transform infrared spectroscopy (FT-IR: JEOL JIR-5500) and a micro-Raman spectrometer (Renishaw system 2000). The Raman spectra were obtained using an Ar+ ion laser operated at 514.5 nm (2.41 eV) at a power of 40 mW.

l&ml

(a) room temperature

3. RESULTS AND DISCUSSION The surface morphology of the CN, films on the Si( 100) substrate was examined by FE-SEM, as shown in Fig. 2. The film prepared at room temperature (Fig. 2(a)) is homogeneous with the occasional incorporation of spherical particles ejected from targets due to surface heating above the melting point, while the film prepared at 650°C (Fig. 2(b)) consists of many rugged particles of which sizes are about 50 nm. AES analysis showed that the films are free (within the detection limit of this technique) from impurities and only signals from C and N are observed. Figure 3 shows the variation of the N/C composition ratio measured by AES in the films as a function of the substrate temperature T,. An N/C composition ratio of

0.28

higher

was substrate

found

at

room

temperatures,

Table 1. Deposition Laser

Target Rotating speed of target Substrate Target-substrate distance Ultimate pressure Gas pressure Substrate temperature DC bias voltage Deposition time

conditions

temperature. the N/C

For

(b)

650

“C

Fig. 2. FE-SEM micrographs of CN, films deposited on Si( 100) at.substrate temperatures of (a) room temperature and (b) 650°C (P,,=lO.OPa, &=3.8 J cm-‘, d=6.0cm).

the

composition

for CN, thin films

Pulsed Nd:YAG laser Wavelength 1.= 532 nm Pulse width z = 6.5 ns Energy density Ed = 3.8 J cm ’ Repetition rate 10 Hz C (purity 99.999%) -20 rpm Si( 100) 6.0 cm <4.0x lo-” Pa 1.0~~10.0 Pa (100% nitrogen) 7, = room temperature 650°C v*=o- -15ov 60 120 minutes

0

100

200

300

400

500

Substrate temperature Fig. 3. Variation

600

700

(“C >

of the N/C composition ratio as a function of the substrate temperature.

Pulsed laser deposition of carbon nitride thin films from graphite targets

600

800

1000 1200 1400 1600 1800 2000

Wave number (cm -‘) Fig. 4. Raman spectrum for the CN, film deposited at P,, = 10.0 Pa (r, = room temperature, V. = - 75 V). ratio decreases

to become 0.18 at 650°C. Our experi-

mental result is similar to that of the DC magnetron sputtering. The above N/C composition ratio of 0.28 is far less than the 1.33 expected for the ideal g-C,N,. We have also investigated the possibility of increasing the N/C composition ratio in the films by using low energy ion bombardment. However, since a negative RF self-bias was applied to the substrate, an extensive re-sputtering took place. Further investigation of the influence of ion bombardment on the CN, film growth was performed by applying an external DC bias voltage V, to the substrate. The maximum N/C composition ratio of 0.32 was obtained at V,= -75 V. The Raman-scattering technique measures certain vibration modes of a material which are determined from both symmetry selection rules and conservation of momentum. A Raman spectrum for the CN, film deposited at P,,= 10.0 Pa (T, = room temperature, V,= - 75 V) is shown in Fig. 4. The peaks at around 1390 and 1600 cm-’ are probably associated with disordered carbon (D-band) and graphitic sp*-bonded carbon (G-band) [ 19,201, respectively. This spectrum looks similar to that of the CN, film obtained by Narayan et al. [ 141. Figure 5 shows the FT-IR absorption spectrum of the CN, film deposited at P,,= 10.0 Pa (T,=room temperature, V,= -75 V). As can be seen in the figure, the absorption peak centered at about 22OOcm-’ is due to the stretching vibration of a C=N triple bond [7,20], suggesting that the C and N atoms are chemically bonded in the film. The most distinguishing spectral feature of this film was the broad, poorly resolved band covering the range 1200-l 700 cm- ‘. This spectral band region contains sp* carbon vibrational modes which become IR active with nitrogen addition. 4. CONCLUSIONS

In this study, CN, thin films were deposited using a pulsed Nd:YAG laser with an external DC bias.

773

GD 11 4000

I

I

2800

I

I

2000

I

1600

I

1200

I

I

800

400

Wave number (cm-‘) Fig. 5. FT-IR absorption spectrum of the CN, film deposited at PN2=10.0 Pa (T,=room temperature, V,= -75 V).

The following conclusions can be drawn from the results: (1) The N/C composition ratio is affected by the substrate temperature and reaches as much as 0.32. (2) Raman results show the presence of both disordered carbon (D-band) and graphitic sp*-bonded carbon (G-band). (3) FT-IR measurement indicates the presence of a C=N triple bond. AcknowledgementspThe authors wish to thank Drs T. lkegami and Y. Yamagata of Kumamoto University for their helpful discussions, and Dr H. Abe of the Ceramic Research Center of Nagasaki and R. Hatada of the Technology Center of Nagasaki for their technical assistance with the data.

REFERENCES Liu, A. Y. and Cohen, M. L., Science, 1989, 245, 841. Liu, A. Y. and Cohen, M. L., Phys. Rev. B, 1990, 41, 10727. Li, D., Chung, Y. W., Wong, M. S. and Sproul, W. D., J. A&. Phys., 1993,14, 219. Shostrom, H., Ivanov, I., Johansson, M., Hultman, L., Sundgren, J.-E., Hainsworth, S., Page, T. F. and Walenberg, L. R., Thin Solid Films, 1994, 246. 103. Torng, C. J., Siversten, J. M., Judy, J. H. a& Chang, C., J. Mater. Res.. 1990. 5. 2490. Kumar, S. and Tansley; T. L., Solid State Commun., 1994,88, 803. Kumar, S. and Tansley, T. L., Thin Solid Films, 1995, 256,44.

Ricci, M., Trinquecoste, M., Auguste, F., Canet, R., Delhaes, P., Guimon, C., Guillouzo, G. P., Nysten, B. and Issi, J. P., J. Mater. Res., 1993, 8, 480. 9. Durrant, S. F., Rangel, E. C. and Moraes, M. A. J. Vat. Sci. Technol. A, 1995, 13, 1901. IO. Rossi, F., Andre, B., van Veen, A., Mijnarends, P. Schut, H., Labohm, F., Delplancke, M. P., Dunlop, and Anger, E., Thin Solid Films, 1994, 253, 85. 11. Rossi, F., Andre, B., van Veen, A., Mijnarends, P.

B., E., H. E.,

774

12. 13. 14.

15.

Y.

%JDA

Schut, H., Labohm, F., Dunlop, H., Delplancke, M. P. and Hubbard, K., J. Mater. Res., 1994, 9, 2440. Ogata, K., Chubaci, J. F. D. and Fujimoto, F., J. Appl. Phys., 1994, 16, 3791. Niu, C., Lu, Y. Z. and Lieber, C. M., Scienc~r, 1993, 261, 334. Narayan, J., Reddy, J., Biunno, N., Kanetkar, S. M.. Tiwari, P. and Parikh, N.. Muter. Sci. Eng. B, 1994, 26, 49. Ren, Z. M., Du, Y. C., Qiu, Y., Wu, J. D., Ying, Z. F.. Xiong, X. X. and Li, F. M.. Phys. Rev. E. 1995,51,5274.

Cl Ul.

16. Zhang, Z. J.. Fan, S. and Lieber, C. M., Appl. Phys. Letf., 1995, 66, 3582. 17. Saitou, K. and Koga, K., Nucl. Instrum. Methods Phys. Rex. B, 1997, 121, 400. 18. Suda, Y., Nakazono, T., Ebihara, K. and Baba, K., Thin Solid Films, 1996, 2811282, 324. 19. Tsai, H. and Bogy, D. B., J. Vat. Sci. Technol. A, 1987, 5, 3281. 20. Kobayashi, S., Nozaki, S., Morisaki, H., Fukui, S. and Masaki, S., Thin Solid Films, 1996, 281/282, 289.