Epitaxial Silicon-Carbon Alloy Growth by Laser Induced Melting and Solidification
- PDF / 1,248,872 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 72 Downloads / 193 Views
tions which challenge our understanding of solid, liquid, and vapor phase crystal growth mechanisms. For pulsed laser induced melting and regrowth, these questions relate to the interdependence of interface morphology, elemental segregation, second phase precipitation, and strain incorporation. The objective of this work is to characterize the microstructure of SiCl1 1 alloys as a function of composition, providing a basis for understanding these complex interactions. EXPERIMENTAL PROCEDURE Double-side polished, {001} silicon wafers (p > 250 f-cm) were RCA cleaned prior to carbon implantation. Samples with nominal peak concentrations of 0.35 at.% C, over :120 nm (as predicted by TRIM [1]), were produced by implantation of 2.22 x 1014, 3.60 x 1014 and 1.14 x 1015 12C+/cm2 at 5, 12.5, and 25 keV respectively. Multiples of this base dose were used to produce samples with nominal peak concentrations of 0.35 through 3.85 at.% C. Except for the 0.35 at.% samples, these implantations produced a 110 nm amorphous layer as measured by 2.2 MeV 4He2+ channeling. Following carbon implantation, the wafers were cleaned in an RF-generated oxygen plasma to remove carbon-based impurities which may have accumulated on the surface during implantation. Immediately prior to laser irradiation, samples were dipped in HF:11 20 (1:10); Fourier-transform infrared spectroscopy (FTIR) 127
Mat. Res. Soc. Symp. Proc. Vol. 398 01996 Materials Research Society
showed that this reduced oxygen incorporation. Sample irradiation was carried out at room temperature, in air, using a XeCl excimer laser pulse (308 nm, 30 ns). Output from the laser was expanded to fill an energy homogenizer and focused to a rectangle approximately 0.3 x 0.5 cm2 . Upon irradiation with fluences of 1.15 to 1.3 J/cm2 , the surface remained molten for 70 to 90 ns as monitored by changes in surface reflectivity. These melt durations were matched with numerical solutions to the heat equation for pure silicon, predicting melt depths of 210 to 260 nm and average interface velocities during regrowth of 4.7 to 5.0 m/s. Following irradiation, samples were analyzed with a variety of tools including transmission electron microscopy (TEM), Fourier-transform infrared spectroscopy (FTIR), secondary-ion mass spectrometry (SIMS), ion channeling, and high-resolution x-ray diffraction (HRXRD). SIMS depth profiles were acquired with primary 0+ incident at 3 keV and collection of secondary
12
C 28 Si+ and 29Si28Si+. Concentrations were calibrated to an accuracy of approx-
imately ±10% by normalizing the ratio of these signals by the integrated ratio acquired from the analysis of a 35 kV, 5 x 1015 C+/cm 2 implant. The substitutional carbon areal density was monitored by room temperature FTIR measurement of the substitutional 12C local vibrational mode (LVM) absorbance at 605 cm-'. Local vibrational modes result from vibration of substitutional impurities lighter than the host at frequencies higher than the maximum optic phonon frequency; since these phonons cannot propagate through the hos
Data Loading...
