Carbon diffusion and nanocrystalline diamond formation in carbon ion-implanted oxides studied by nuclear elastic scattering

https://doi.org/10.1016/S0168-583X(00)00675-3Get rights and content

Abstract

We have shown that MeV implantation of carbon into fused quartz and sapphire followed by thermal annealing in a suitable environment can result in the formation of diamond. Using cross-sectional transmission electron microscopy (TEM) and secondary ion mass spectroscopy (SIMS), we determined (in a previous paper) that, following annealing, there was a redistribution of carbon from the original implantation depth, depending on the annealing environment, annealing time and annealing temperature. In our search, for the optimum implantation and annealing parameters to maximize the yield of diamond, we have used backscattering spectrometry (BS), with MeV hydrogen, to profile the implanted carbon, taking advantage of the large C(p,p)C scattering cross-section at around 1.73 MeV. We studied samples of fused quartz and sapphire implanted with carbon to a range of doses and annealed in forming gas, oxygen and argon. We show that in an oxygen environment, there is significant carbon loss in fused quartz but not in sapphire while in the other environments no significant loss is reported. We conclude that redistribution of carbon, the formation of nanocrystalline diamond (as seen in cross-sectional TEM) and possible carbon loss is determined both by the mobility of carbon in the host matrix at the prevailing annealing temperatures and, most importantly, the annealing ambient.

Introduction

Many studies involving the synthesis and characterization of nanometer-sized clusters of atoms have been carried out in the last few years with a view to creating materials with properties which can be varied according to size. A perfect example is the observed variation in bandgap with cluster size [1], which allows for the tuning of luminescence wavelengths according to the desired application.

Recently, we succeeded in creating nanocrystalline diamond directly from carbon implantation of fused quartz followed by thermal annealing [2], [3]. The formation of diamond was found to be possible within a narrow range of doses and only when the samples were annealed in forming gas (4% hydrogen in argon) for 1 h at 1100°C. Using techniques such as absorption spectroscopy, transmission electron microscopy (TEM) and Raman spectroscopy [3], we determined that significant carbon loss occurred when samples were annealed in oxygen or argon but not in forming gas. In addition, sapphire showed less carbon loss compared to fused quartz for the same annealing conditions.

In this report, we use the backscattering technique, taking advantage of the large C(p,p)C scattering cross-section at around 1.73 MeV [4], to profile the implanted carbon and thus gain valuable information on the distribution of carbon and any loss thereof.

Section snippets

Experimental

The samples studied were optical grade fused quartz supplied by Esco Products and sapphire (α-Al2O3) from Crystal Systems. In both cases, samples were implanted with 1 MeV carbon ions at room temperature using the 1.7 MeV ion implanter located in the Research School of Physical Sciences, The Australian National University. The source material was compacted carbon powder, which was sputtered by cesium atoms and accelerated down a tandem accelerator into the specimen chamber through an aperture

Experimental results and discussion

Fig. 1 shows RBS spectra for fused quartz samples implanted to doses of 0.5, 5 and 6×1017Ccm−2 and annealed for various durations in argon, oxygen and forming gas annealing environments. Results for unimplanted fused quartz and the as-implanted samples are included for comparison. The peak observed at ∼1190 keV is due to the implanted carbon as confirmed by the non-Rutherford modified-RUMP [5] simulation using 1.78 MeV protons (smooth curve) and the nominal carbon depth scale shown, calculated

Conclusion

The substrate, annealing environment and annealing time all determine the amount of carbon retained. To maximize the formation of nanodiamond, all these parameters must be optimized. Because there is less carbon loss in sapphire, this material may be ideal for maximizing nanodiamond yield. We propose that one way to achieve higher yields would be to implant overlapping ranges of carbon and hydrogen. In this way, formation of nanocrystalline diamond would not depend on the inward diffusion of

Acknowledgements

We thank the department of Electronic Materials Engineering, Australian National University, for allowing us to access their ion implantation facilities.

References (6)

  • R. Amirikas et al.

    Nucl. Instr. and Meth. B

    (1993)
  • L.R. Doolittle

    Nucl. Instr. Meth. B

    (1985)
  • A.P. Alivisatos

    MRS Bull.

    (1995)
There are more references available in the full text version of this article.

Cited by (0)

View full text