Nonhydrostatic Stress Effects on Solid Phase Epitaxial Growth in Silicon
- PDF / 383,860 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 99 Downloads / 245 Views
William B. Carter and Michael J. Aziz Division of Applied Sciences, Harvard University, Cambridge MA 02138
ABSTRACT The dependence of solid phase epitaxial growth in Si on uniaxial compression applied perpendicular to the amorphous-crystal interface is investigated. Long, thin pure Si bars of square cross section are ion-implanted to produce amorphous layers on the end faces. The bars are placed end-to-end and uniaxially loaded at temperature to partially regrow the amorphous layers. The resulting growth rates are measured ex situ by re-heating the samples on a hot stage and using time-resolved reflectivity to deduce interface depths. Preliminary results are that uniaxial compression is more effective than hydrostatic pressure for enhancing the growth rate, in qualitative but not quantitative agreement with previously made predictions.
INTRODUCTION Crystal growth frequently occurs under conditions of nonhydrostatic stress. For example, during epitaxial growth of a thin film on a substrate, differences in lattice parameter or thermal expansion between film and substrate can lead to significant stresses during growth. Studies of the thermodynamics of such growth have primarily focused on the effects of stress on the driving forces that govern the growth rate and stability of the growing film. However, little attention has been paid to the effect of stress on the mobilities of the interfaces or the atomic species involved in growth. This is largely due to the difficulties associated with sustaining non-hydrostatic stress
in the presence of strain-relief mechanisms such as creep or dislocation injection. Solid phase epitaxial growth (SPEG) of amorphous Si on crystalline Si has proved to be a good test case for understanding the effect of stress on atomic mobilities. SPEG in pure Si is a clean process in which the thermodynamics of both amorphous and crystal phases are well understood; only one atomic species is involved; and the direction of growth is easily controlled. In addition, non-hydrostatic stress experiments on SPEG in Si are possible because SPEG rates are measurable at temperatures below the ductile-brittle transition [6] in crystalline Si. Olson and Roth [1] measured the temperature dependence of the interface velocity, v, in SPEG, and found Arrhenius behavior over 10 orders of magnitude in growth rate. Given this wide range, a simple transition state theory (TST) model can be used to describe the growth rate. This model assumes that a single, unimolecular, defect-mediated mechanism controls the SPEG rate in the temperature range of interest. For the case of SPEG, the driving force for growth is virtually constant over the range of temperature and pressure covered by the experiments, in which case the pressure dependence of SPEG is given by AV=-kBT ap(l where kB is Boltzmann's constant, T is temperature, P is pressure, and AV* is the activation volume. A negative AV* characterizes an enhancement of the rate with pressure, whereas a 87 Mat. Res. Soc. Symp. Proc. Vol. 356 0 1995 Materials Research Society
Data Loading...
