Real-Time Studies of Interface Structural Dynamics
- PDF / 406,158 Bytes
- 8 Pages / 420.48 x 639 pts Page_size
- 65 Downloads / 243 Views
REAL-TIME STUDIES OF INTERFACE STRUCTURAL DYNAMICS WALTER P. LOWE * AND ROY CLARKE** *AT&T Bell Laboratories, 600 Mountain Ave., Murray Hill, NJ 07974 "*Randall Laboratory of Physics, University of Michigan, Ann Arbor, MI 48109
ABSTRACT We present dynamic structural studies of thin films and their interface with underlying substrates using real-time x-ray diffraction. Using synchrotron light we have observed, in real-time, interface dynamics in semiconductor systems such as GexSi(1-x) /Si. The measurements show that under large temperature changes thin epitaxial layers may behave cooperatively to modify the overall strain profile. Dynamic behavior is exhibited in a series of discontinuities in the perpendicular lattice constant of the overlayer.
INTRODUCTION Structural changes that have long range correlations but occur on short time scales are difficult to study. One such place where these types of studies are important is the interface kinetics of semiconductors. The term kinetics itself implies that the crystal lattice must be measured rapidly with a probe capable of resolving minute changes in the atomic order or correlated perturbations within the lattice. In semiconductor processing the associated time scale of dynamic changes in interface structure is known to be strongly temperature dependent. Structural changes are expected to occur in less than 1 second for samples at elevated temperatures. As will be shown later certain kinetic interface effects only occur during rapid temperature changes. Thus these effects are truly kinetic - - depending on the time rate of change of temperature rather than the temperature itself. A set of criteria is established to qualify structural probes that can time-resolve interface kinetics. Probe candidates must have high spatial resolution with the ability to sample large specimen areas. In high quality interfaces, kinetic effects will be long range. Furthermore it is not always possible to "cycle" the specimen to make multiple measurements. In some cases the excitation, such as temperature for instance, cannot be cycled and stabilized at high enough rates to time-resolve the kinetics. In other cases the sample may undergo irreversible changes. In reality these requirements make x-ray diffraction the only likely candidate; however tntil now x-ray diffraction experiments have not been done under these extreme requirements of temporal resolution, 4-space resolution and parallel rather than sequential data acquisition. The experiments discussed here were done at the NSLS on beam line X16B. Bend magnet
Mat. Res. Soc. Symp. Proc. Vol. 237. 01992 Materials Research Society
368
lines are modest with respect to their brilliance, however X16B does provide an opportunity to demonstrate the usefulness of dispersive scattering geometry and area detectors.
SCATTERING GEOMETRY In order to time-resolve the kinetics of semiconductor interfaces, we use a scattering geometry that measures a range of diffraction vectors, -, but does not require translations or rotations of the x-ray beam
Data Loading...
