Kazuyoshi Tomita

A new buffer layer for high quality GaN growth by metalorganic vapor phase epitaxy

A new buffer layer to grow high-quality GaN films was proposed. The new buffer layer consisted of a thin (20–30 nm) InN layer deposited at low temperature (∼600 °C). GaN films were grown on (1120)-oriented (A-face) sapphire substrates using a conventional GaN buffer layer and an InN buffer layer by atmospheric pressure metalorganic vapor phase epitaxy. Dislocations in the GaN films were observed by cross-sectional transmission electron microscopy (TEM). The dislocation densities were measured from the TEM observation and were ∼4×109 and ∼6×108 cm−2 for epilayers with the GaN and the InN buffer, respectively. The low dislocation density by the InN buffer was attributed to relaxation of the stress in the GaN epilayers due to the low melting point of InN. GaN epilayers using the InN buffer also showed good electrical properties.

Reduction of Mg segregation in a metalorganic vapor phase epitaxial grown GaN layer by a low-temperature AlN interlayer

The redistribution behavior of Mg in a sequentially regrown GaN epilayer on a p-type doped GaN template was studied. All samples in this study were regrown by metalorganic vapor phase epitaxy on the sapphire substrates. A high density and a slow tail of Mg concentration were observed in a nominally undoped layer due to the surface segregation. We found that the insertion of a low-temperature (LT) AlN interlayer was effective to suppress the Mg redistribution in the GaN regrown layer. Analyzing the temperature dependence of the surface segregation, the activation energy of the Mg segregation was estimated to be 0.63eV in GaN and 2.47eV in a LT-AlN layer, respectively.

LPE Growth and Surface Morphology of InxGa1-xAsyP1-y (y≤0.01) on (100) GaAs

The effects of the growth conditions (growth time, growth temperature and melt concentration) on the surface morphology of an LPE layer of InxGa1-xAsyP1-y(y≤0.01) on a (100)GaAs substrate are studied. The change in morphology originates from the relation between the growth temperature (Tg) and the equilibrium saturation temperature (Ts) of the source melt, which is in equilibrium with the quaternary solid but is not in equilibrium with the GaAs substrate. The equilibrium temperature was determined from growth layer thickness data, as well as by analysing the phase diagram. It was found that partial growth of the quaternary layer can take place on GaAs even at Tg>Ts, and this is supported by partial melt-back of the substrate. A mirror-like smooth surface was obtained even if the supersaturation ΔT was as low as 1°C.

Self-Separation of Freestanding GaN from Sapphire Substrates by Hydride Vapor Phase Epitaxy

Freestanding GaN wafers were produced by a newly developed self-separation method. Thick GaN layers were grown using hydride vapor phase epitaxy on a sapphire substrate with GaN seeds. The separation of the thick GaN layers took place during the growth sequence at the interface of GaN/sapphire, because of thermal stress and lattice mismatch between GaN and sapphire. The size of the freestanding GaN wafers was 23 mm x 22 mm. The threading dislocation density at the top surface was 10 6 cm -2 to ∼10 7 cm -2 .

Uniform growth of Te-doped AlxGa1t−xAs (x = 0-0.65) on (111) B GaP substrates by liquid phase epitaxy

Abstract Doping elements play an important role in the early stage of heteroepitaxial growth by liquid-phase epitaxy (LPE). It is found that uniform growth of AlGaAs on chemically-etched (111)B GaP substrates can be obtained by doping with group VI elements, Te or Se, into the melt for LPE. From Nomarski microscopic observation of the heterointerface after selective etching of epitaxial layers, it can be explained that such a dopant has an important effect on suppressing the meltback of GaP substrates and causes an acceleration of two-dimensional growth. Other dopants such as Mg, Zn, Si, Ge or Sn produce only hexagonal island growth, similar to the case of undoped growth. Moreover, multilayer growth has a clear effect of reducing the dislocation density in the epitaxial layer.