Dopant Incorporation during Rapid Solidification
- PDF / 2,322,946 Bytes
- 10 Pages / 417.6 x 639 pts Page_size
- 12 Downloads / 288 Views
C. W. WHITE, D. M. ZEHNER, J. NARAYAN, 0. W. HOLLAND, B. R. APPLETON Solid State Division, Oak Ridge National Laboratory, Oak Ridge, TN 37830 and S. R. WILSON Motorola Inc.,
Phoenix,
AZ
ABSTRACT Incorporation of Group III, IV, V dopants in silicon occurs as a result of solute trapping during laser annealing. Distribution coefficients and substitutional solubilities are far greater than equilibrium values, and can be functions of growth velocity and crystal orientation. Mechanisms limiting dopant incorporation at high concentrations are identified and discussed.
INTRODUCTION In pulsed laser annealing of ion implanted silicon, the rapid deposition of energy into the near surface region leads to melting, followed by liquid phase epitaxial regrowth from the underlying substrate [1-4]. During recrystallization the velocity of the liquid-solid interface can be several meters per second [5] which leads to solidification conditions that are far from equilibrium at the interface. At these velocities, dopant incorporation occurs by means of solute trapping and dopant concentrations far in excess of the equilibrium solubility limit can be achieved [6]. Here we review some of the systematic studies of the incorporation of Group III, IV and V dopants in silicon during rapid solidification. The behavior of these impurities has been studied at both low and high dopant concentrations. At low dopant concentrations, the dopant profiles measured after laser annealing can be compared to model calculations of dopant redistribution in order to determine the distribution coefficient (k') from the liquid during solidification [6]. For each impurity there is a maximum concentration (Cmax) which can be incorporated substitutionally in the lattice during pulsed laser annealing [6]. The mechanisms which limit the incorporation of dopant into the crystal lattice are discussed as well as the dependence of growth kinetics upon interface velocity [7,8] and crystal orientation [9]. EXPERIMENTAL Group III, IV and V impurities were implanted into both (100) and (111) sili15 17 2 con to doses in the range of 10 -10 /cm . Implantation energies were chosen to give a projected range of '800 A. Laser annealing was carried out using a pulsed ruby laser (6943 AO, 'U12 x 10-9 s) or a pulsed XeCl laser (3080 A, 'v35 x 9 10s). By changing the energy density obtained from these lasers one can vary the regrowth velocity in the range of 2-6 m/s as determined from heat flow calculations. Analysis of the implanted region was carried out using Rutherford *Research sponsored by the Division of Materials Sciences, U.S. Department of Energy under contract W-7405-eng-26 with Union Carbide Corporation. Mat. Res.Soc.Syrup. Proc. Vol. 13 (1983) Published by Elsevier Science Publishing Co.,Inc.
288
1021 5
o IMPLANTED 0 LASERANNEALED
'-05
-
1020
0 0
0 00 52
0
5
Z
0 0
0
CALCULATED
0 0
0
0
1 15
a
o 2 Io'l
SOLIDSOLUBILITY= LIMIT
Fig. 1. Dopant profiles for In (125 15 2 keV, 1.2 x 10 /cm ) in (100) Si compared The dashed proto model calculations. file
Data Loading...
