Electrical Behaviour of Heavily-Doped Ion Implanted Layers in Silicon
Some of the practical conseguences of carrying out high fluence implantation (⩾ 5 × 1014 ions/cm2) on an industrial scale have been examined.
The annealing behaviour of heavily doped ion implanted layers of boron, phosphorus, arsenic and antimony have been studied as a function of doping rate, dose and ion-beam energy. The experiments, which were all carried out on 5 cm diameter wafers, involved the measurement of the spatial distribution of resistivity across the surface of the implanted specimens at a series of annealing temperatures. Under certain conditions, it was found that the uniformity and the reproducibility of the dopant-induced electrical activity, which are such an important characteristic of ion implantation, appear to be significantly degraded. The loss of uniformity was found to be associated with the variations in temperature which can occur across the surface of the wafers during such high-fluence implantations.
Persistent implantation temperature memory effects were observed in the electrical behaviour of antimony even after prolonged annealing at 1200°C. Boron, phosphorus and arsenic layers exhibit similar non-uniformities after implantation. For these important dopants, however, there is much less evidence of the implantation history after annealing in the temperature range 950° – 1200° C and good uniformity and reproducibility of the electrical resistivity have been obtained after doping with high-intensity ion beams under conditions corresponding to large temperature variations across the surface of the wafers.
Measurements of the heating effects of ion beams on both thermally-bonded and thermally-isolated silicon wafers have also been made and compared with theory. It is shown that the cumulative heating effect of implanting with even modest beam intensities can be significant. For example, at a typical implantation energy of 100 keV a beam current gensity of only 1 µA/cm2 may result in a temperature rise of over 100°C. The concept of IRRADIANCE is introduced as a measure of the ion-beam power density. It is seen that it is this parameter, in conjunction with the fluence and the nature of the sample mounting, which determine the temperature rise. It is also shown that in certain circumstances the beam irradiance can be reduced without a commensurate decrease in the throughput of implanted wafers.
KeywordsAnnealing Behaviour Target Chamber Sheet Resistivity Doping Uniformity Diameter Wafer
Unable to display preview. Download preview PDF.
- (1).Mayer, J.W. et al. — Ion Implantation in Semiconductors, Chapter 5. Academic Press, 1970.Google Scholar
- (2).Robertson, G.I. — Rotating scan for ion implantation. Abstract No. 223, Extended Abstracts Vol. 74–1, Electrochemical Society, May, 1974.Google Scholar
- (3).Freeman, J.H. — A variable geometry separator and low energy heavy ion accelerator. Part I, General Principles, AERER 6254.Google Scholar
- (4).Freeman, J.H., Caldecourt, L.R., Done, K.C.W., and Francis, R.J. — An industrial scale ion implantation facility. AERER 6496.Google Scholar
- (5).Dearnaley, G., Freeman, J.H., Nelson, R.S., and Stephen, J. — Ion Implantation, p.421. Defects in Crystalline Solids Series, North-Holland, 1973.Google Scholar
- (7).Drum, C.H. — Diffusion of ion implanted antimony and arsenic in silicon. Abstract No. 84, Extended Abstracts Vol. 74–1, Electrochemical Society, May, 1974.Google Scholar