Molecular beam epitaxy growth of the dilute nitride with a helical resonator plasma source
Drawing of plasma source showing helical resonator, coupling loop, and pyrolytic boron nitride (PBN) discharge tube. The discharge tube is in diameter. A constriction made from PBN at the upstream end of the discharge tube prevents the plasma from expanding back into the stainless steel gas feed line. A three-element baffle (inset) at the open end of the discharge tube blocks the line of sight to the discharge, suppressing the ion flux. The cylindrical Ta ground shield that surrounds the helix is not shown.
Ion current from the plasma source, measured with a Faraday cup in a test chamber. The ion current is measured as a function of gas pressure in the discharge tube and the net forward rf power supplied to the discharge.
(▵) and (◻) measured with the watt meter and (●), calculated as , as a function of frequency. The two vertical lines indicate the frequencies of the two operating modes of the plasma source discussed in this article: on resonance (left) and off resonance (right).
Emission spectra for on-resonance operation at (dashed line) and off-resonance operation at (solid line).
Emission intensities for atomic N ( line with background from molecular emission subtracted) (▴) and ( line) (엯). Left plot is measured with the plasma operating in the off-resonance mode; the right plot in the on-resonance mode.
N content of on-resonance (∎) and off-resonance (▴) grown samples as determined by XRD for varying power levels and pressures. and .
Effect of As coating of the PBN discharge tube on the N concentration for on-resonance (∎ and ◻) and off-resonance (▴ and ▵) operations. The samples represented by the solid markers are the samples in Fig. 4, grown at . The hollow symbols are samples grown after the removing the As coating. The N concentrations have been scaled to eliminate differences caused by (a) differences in growth rate, (b) changes in source to substrate distance, and (c) differences in gas flow for the growths after the discharge tube was cleaned.
Photoluminescence spectra measured at for GaNAs samples grown with an As coated pyrolytic boron nitride discharge tube (a) and with a clean discharge tube (b). Samples grown with the plasma source operating on resonance are indicated by solid lines and off resonance with dashed lines. The N concentrations determined by x-ray diffraction are indicated in the figures. The arrows indicate the band gaps estimated from the N concentrations (Ref. 9). All of the samples are as grown, with no postgrowth annealing.
Photoluminescence peak energies measured at as function of [N] for on resonance (∎ and ◻) and off resonance (▴ and ▵); samples represented by solid markers were grown with an As-contaminated discharge tube, the open circles with a clean discharge tube. The solid line shows the expected band gap adjusted to the measurement temperature. Error bars are shown for three representative samples, , based on fits to high resolution x-ray diffraction scans. A representative XRD scan is shown in the inset for a sample with .
SIMS depth profile of GaNAs sample. The slow decay of the carbon and oxygen concentrations with depth suggestions that these elements result from sample surface contamination after growth. For depths lower than the epilayer nitrogen is below the detection limit, .
Article metrics loading...
Full text loading...